train.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "72O4rhDEZrIq",
        "outputId": "f6081b4b-bd38-4891-f0c5-378f0df34056"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Cloning into 'DeepVCP-Pointcloud-Registration'...\n",
            "remote: Enumerating objects: 1333, done.\u001b[K\n",
            "remote: Counting objects: 100% (36/36), done.\u001b[K\n",
            "remote: Compressing objects: 100% (29/29), done.\u001b[K\n",
            "remote: Total 1333 (delta 11), reused 19 (delta 7), pack-reused 1297\u001b[K\n",
            "Receiving objects: 100% (1333/1333), 138.82 MiB | 15.24 MiB/s, done.\n",
            "Resolving deltas: 100% (344/344), done.\n"
          ]
        }
      ],
      "source": [
        "! git clone https://github.com/niladridutt/DeepVCP-Pointcloud-Registration.git"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OUGR4V4kDY15",
        "outputId": "f98057c9-a400-4d78-abac-3db1c02f9869"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "--2023-04-15 10:12:35--  https://shapenet.cs.stanford.edu/media/modelnet40_normal_resampled.zip\n",
            "Resolving shapenet.cs.stanford.edu (shapenet.cs.stanford.edu)... 171.67.77.19\n",
            "Connecting to shapenet.cs.stanford.edu (shapenet.cs.stanford.edu)|171.67.77.19|:443... connected.\n",
            "WARNING: cannot verify shapenet.cs.stanford.edu's certificate, issued by ‘CN=InCommon RSA Server CA,OU=InCommon,O=Internet2,L=Ann Arbor,ST=MI,C=US’:\n",
            "  Issued certificate has expired.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 1705117335 (1.6G) [application/zip]\n",
            "Saving to: ‘modelnet40_normal_resampled.zip’\n",
            "\n",
            "modelnet40_normal_r 100%[===================>]   1.59G  58.1MB/s    in 29s     \n",
            "\n",
            "2023-04-15 10:13:04 (56.1 MB/s) - ‘modelnet40_normal_resampled.zip’ saved [1705117335/1705117335]\n",
            "\n"
          ]
        }
      ],
      "source": [
        "!wget https://shapenet.cs.stanford.edu/media/modelnet40_normal_resampled.zip --no-check-certificate"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "tcNwoHRlmtTL"
      },
      "outputs": [],
      "source": [
        "! unzip -q modelnet40_normal_resampled.zip"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "VpM8oiQsaMz2"
      },
      "outputs": [],
      "source": [
        "!pip install --upgrade https://github.com/unlimblue/KNN_CUDA/releases/download/0.2/KNN_CUDA-0.2-py3-none-any.whl -q"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nWFanNlZbNz2",
        "outputId": "76a9918b-62d3-4d6f-b5cf-13539c2509d2"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\u001b[?25l     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0.0/146.0 kB\u001b[0m \u001b[31m?\u001b[0m eta \u001b[36m-:--:--\u001b[0m\r\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m146.0/146.0 kB\u001b[0m \u001b[31m4.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h"
          ]
        }
      ],
      "source": [
        "! pip install Ninja -q"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MnFkroz4c8Ht",
        "outputId": "e3cecb93-6de6-4345-d933-556e00683786"
      },
      "outputs": [],
      "source": [
        "! pip install trimesh -q"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lWF9CPBba-kv",
        "outputId": "4c7ed8e4-530a-44e8-e6df-ece38b3019f5"
      },
      "outputs": [
        {
          "ename": "ImportError",
          "evalue": "cannot import name 'CustomDataset2' from 'CustomDataset2' (\\\\evs2.cs.ucl.ac.uk\\student-msc\\cgvi\\2022\\bingwang\\GitHub\\DeepVCP-Pointcloud-Registration\\CustomDataset2.py)",
          "output_type": "error",
          "traceback": [
            "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[1;31mImportError\u001b[0m                               Traceback (most recent call last)",
            "Cell \u001b[1;32mIn[1], line 18\u001b[0m\n\u001b[0;32m     16\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mKITTIDataset\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m KITTIDataset\n\u001b[0;32m     17\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mCustomDataset\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m CustomDataset\n\u001b[1;32m---> 18\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mCustomDataset2\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m CustomDataset2\n\u001b[0;32m     19\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mdeepVCP_loss\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m deepVCP_loss\n\u001b[0;32m     20\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mmatplotlib\u001b[39;00m\n",
            "\u001b[1;31mImportError\u001b[0m: cannot import name 'CustomDataset2' from 'CustomDataset2' (\\\\evs2.cs.ucl.ac.uk\\student-msc\\cgvi\\2022\\bingwang\\GitHub\\DeepVCP-Pointcloud-Registration\\CustomDataset2.py)"
          ]
        }
      ],
      "source": [
        "import os\n",
        "import numpy as np\n",
        "import torch\n",
        "from torch import nn\n",
        "from torch.optim import Adam\n",
        "from torch.utils.data import DataLoader\n",
        "from torch.optim.lr_scheduler import ReduceLROnPlateau\n",
        "from scipy.spatial.transform import Rotation as R\n",
        "import time\n",
        "import pickle\n",
        "import argparse\n",
        "from utils import *\n",
        "\n",
        "from deepVCP import DeepVCP\n",
        "from ModelNet40Dataset import ModelNet40Dataset\n",
        "from KITTIDataset import KITTIDataset\n",
        "from CustomDataset import CustomDataset\n",
        "from CustomDataset2 import CustomDataset2\n",
        "from deepVCP_loss import deepVCP_loss\n",
        "import matplotlib\n",
        "matplotlib.use('Agg')\n",
        "from matplotlib import pyplot as plt"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "6eg9nCTqdPTA"
      },
      "outputs": [],
      "source": [
        "dataset = 'modelnet'\n",
        "retrain_path = 'store'\n",
        "model_path = 'final_model.pt'\n",
        "full_dataset = True"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "T2R0MObvcqHG",
        "outputId": "e6ab55fc-13e2-45ff-fd7e-455511bce501"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Params: epochs: 10, batch: 1, lr: 0.001, alpha: 0.5\n",
            "\n",
            "device: cuda\n"
          ]
        }
      ],
      "source": [
        "num_epochs = 10\n",
        "batch_size = 1\n",
        "lr = 0.001\n",
        "# loss balancing factor \n",
        "alpha = 0.5\n",
        "\n",
        "print(f\"Params: epochs: {num_epochs}, batch: {batch_size}, lr: {lr}, alpha: {alpha}\\n\")\n",
        "\n",
        "# check if cuda is available\n",
        "device = 'cuda' if torch.cuda.is_available() else 'cpu'\n",
        "print(f\"device: {device}\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JTia5bfVcxG_",
        "outputId": "70299cfc-fe26-463d-cdf2-b86eacc5dbf4"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "# Total clouds 2\n"
          ]
        }
      ],
      "source": [
        "# root = 'modelnet40_normal_resampled/'\n",
        "# shape_names = np.loadtxt(root+\"modelnet10_shape_names.txt\", dtype=\"str\")\n",
        "# train_data= ModelNet40Dataset(root=root, augment=True, full_dataset=full_dataset, split='train')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bWiQwvqcdA7p",
        "outputId": "86a82273-537d-4db2-b8d9-e98ed20385d6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "# Total clouds 2\n"
          ]
        }
      ],
      "source": [
        "# test_data = ModelNet40Dataset(root=root, augment=True, full_dataset=full_dataset,  split='test')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yQpQfDt_c27b",
        "outputId": "e162b74f-da03-4d21-acda-547aad9b3bdf"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train dataset size:  2\n",
            "Test dataset size:  2\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "DeepVCP(\n",
              "  (FE1): feat_extraction_layer(\n",
              "    (sa1): PointNetSetAbstraction(\n",
              "      (mlp_convs): ModuleList(\n",
              "        (0): Conv2d(6, 16, kernel_size=(1, 1), stride=(1, 1))\n",
              "        (1): Conv2d(16, 16, kernel_size=(1, 1), stride=(1, 1))\n",
              "        (2): Conv2d(16, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "      )\n",
              "      (mlp_bns): ModuleList(\n",
              "        (0-1): 2 x BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "        (2): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      )\n",
              "    )\n",
              "    (sa2): PointNetSetAbstraction(\n",
              "      (mlp_convs): ModuleList(\n",
              "        (0): Conv2d(6, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "        (1): Conv2d(32, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "      )\n",
              "      (mlp_bns): ModuleList(\n",
              "        (0): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "        (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      )\n",
              "    )\n",
              "    (sa3): PointNetSetAbstraction(\n",
              "      (mlp_convs): ModuleList(\n",
              "        (0): Conv2d(6, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "        (1): Conv2d(64, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "      )\n",
              "      (mlp_bns): ModuleList(\n",
              "        (0): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "        (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      )\n",
              "    )\n",
              "  )\n",
              "  (WL): weighting_layer(\n",
              "    (fc1): Sequential(\n",
              "      (0): Linear(in_features=32, out_features=16, bias=True)\n",
              "      (1): ReLU()\n",
              "    )\n",
              "    (fc2): Sequential(\n",
              "      (0): Linear(in_features=16, out_features=8, bias=True)\n",
              "      (1): ReLU()\n",
              "    )\n",
              "    (fc3): Sequential(\n",
              "      (0): Linear(in_features=8, out_features=1, bias=True)\n",
              "      (1): Softplus(beta=1, threshold=20)\n",
              "    )\n",
              "  )\n",
              "  (DFE): feat_embedding_layer(\n",
              "    (fc1): Linear(in_features=35, out_features=32, bias=True)\n",
              "    (fc2): Linear(in_features=32, out_features=32, bias=True)\n",
              "    (fc3): Linear(in_features=32, out_features=32, bias=True)\n",
              "    (max_pool): MaxPool1d(kernel_size=32, stride=32, padding=0, dilation=1, ceil_mode=False)\n",
              "  )\n",
              "  (cpg): cpg(\n",
              "    (conv1): Conv3d(32, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n",
              "    (conv2): Conv3d(16, 4, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n",
              "    (conv3): Conv3d(4, 1, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n",
              "    (softmax): Softmax(dim=-1)\n",
              "  )\n",
              ")"
            ]
          },
          "execution_count": 7,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "train_loader = DataLoader(dataset=train_data, batch_size=batch_size, shuffle=False)\n",
        "test_loader = DataLoader(dataset=test_data, batch_size=batch_size, shuffle=False)\n",
        "\n",
        "\n",
        "num_train = len(train_data)\n",
        "num_test = len(test_data)\n",
        "print('Train dataset size: ', num_train)\n",
        "print('Test dataset size: ', num_test)\n",
        "\n",
        "use_normal = dataset == \"modelnet\"\n",
        "\n",
        "# Initialize the model\n",
        "model = DeepVCP(use_normal=use_normal)\n",
        "if torch.cuda.device_count() > 1:\n",
        "    print(\"Let's use\", torch.cuda.device_count(), \"GPUs!\")\n",
        "    # dim = 0 [30, xxx] -> [10, ...], [10, ...], [10, ...] on 3 GPUs\n",
        "    model = nn.DataParallel(model)\n",
        "\n",
        "model.to(device)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DYuFCk8MdSHp",
        "outputId": "fdf7aedf-b807-4479-96fd-91a92a2e3e7b"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\u001b[1;30;43mStreaming output truncated to the last 5000 lines.\u001b[0m\n",
            "Loss: 0.2841099506198117\n",
            "rotation error:  tensor(29.6124, dtype=torch.float64)\n",
            "translation error:  tensor(0.6462, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.70760750770569 seconds ---\n",
            "Processing file: bed_0282\n",
            "feature extraction time:  8.560503959655762\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.031216859817504883\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.005066394805908203\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3314, 7600,  633,  ..., 7412, 3811, 9548], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04758477210998535\n",
            "Loss: 0.3077676974373139\n",
            "rotation error:  tensor(20.3923, dtype=torch.float64)\n",
            "translation error:  tensor(1.0480, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.977498054504395 seconds ---\n",
            "Processing file: bed_0283\n",
            "feature extraction time:  9.18882131576538\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02904820442199707\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.00287628173828125\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7277, 5750, 7247,  ..., 2004, 2377, 7150], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.06188225746154785\n",
            "Loss: 0.18307932614115885\n",
            "rotation error:  tensor(76.5210, dtype=torch.float64)\n",
            "translation error:  tensor(0.2408, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.68598747253418 seconds ---\n",
            "Processing file: bed_0284\n",
            "feature extraction time:  8.523370742797852\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.034439802169799805\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.004461765289306641\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3628, 9623, 7242,  ..., 7829, 9932, 1809], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05596041679382324\n",
            "Loss: 0.2579073541676978\n",
            "rotation error:  tensor(33.0924, dtype=torch.float64)\n",
            "translation error:  tensor(0.8402, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.14065670967102 seconds ---\n",
            "Epoch: [0/10], Batch: 389, Loss: 0.2579073541676978\n",
            "Processing file: bed_0285\n",
            "feature extraction time:  9.185311317443848\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.027977466583251953\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.004014253616333008\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([8411, 6362,  445,  ..., 6360, 1312,  862], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05568051338195801\n",
            "Loss: 0.15159963066590354\n",
            "rotation error:  tensor(20.5446, dtype=torch.float64)\n",
            "translation error:  tensor(0.4600, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.580511808395386 seconds ---\n",
            "Processing file: bed_0286\n",
            "feature extraction time:  8.512573719024658\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03381848335266113\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0040340423583984375\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 883, 3039, 2119,  ..., 6238, 8760, 1364], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04926609992980957\n",
            "Loss: 0.21870116510732693\n",
            "rotation error:  tensor(3.4046, dtype=torch.float64)\n",
            "translation error:  tensor(0.6495, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.110195875167847 seconds ---\n",
            "Processing file: bed_0287\n",
            "feature extraction time:  9.177715301513672\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025980234146118164\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028748512268066406\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6552, 5412, 7550,  ...,  926,  828, 1231], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05530095100402832\n",
            "Loss: 0.2640757141070025\n",
            "rotation error:  tensor(29.1719, dtype=torch.float64)\n",
            "translation error:  tensor(0.6553, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.57657527923584 seconds ---\n",
            "Processing file: bed_0288\n",
            "feature extraction time:  8.572754383087158\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.034533023834228516\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003229856491088867\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3475, 6623,    3,  ..., 1328, 7357, 9514], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04861044883728027\n",
            "Loss: 0.17950378388084126\n",
            "rotation error:  tensor(17.1054, dtype=torch.float64)\n",
            "translation error:  tensor(0.4017, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.067465782165527 seconds ---\n",
            "Processing file: bed_0289\n",
            "feature extraction time:  9.107971668243408\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024658679962158203\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002882719039916992\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7750, 6033, 5946,  ..., 9540, 9713, 9498], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.057329654693603516\n",
            "Loss: 0.28658364779715195\n",
            "rotation error:  tensor(42.3525, dtype=torch.float64)\n",
            "translation error:  tensor(0.8812, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.696540355682373 seconds ---\n",
            "Epoch: [0/10], Batch: 394, Loss: 0.28658364779715195\n",
            "Processing file: bed_0290\n",
            "feature extraction time:  8.610717535018921\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030567169189453125\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.00316619873046875\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9354, 1898, 5884,  ..., 7515,   71, 3668], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.051548004150390625\n",
            "Loss: 0.29518884240145404\n",
            "rotation error:  tensor(17.5377, dtype=torch.float64)\n",
            "translation error:  tensor(0.8565, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.038516759872437 seconds ---\n",
            "Processing file: bed_0291\n",
            "feature extraction time:  9.298762321472168\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02512192726135254\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003146648406982422\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9222, 1675,  303,  ..., 2335, 4209, 7468], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04783773422241211\n",
            "Loss: 0.23428666491364825\n",
            "rotation error:  tensor(17.9997, dtype=torch.float64)\n",
            "translation error:  tensor(0.8655, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.813585996627808 seconds ---\n",
            "Processing file: bed_0292\n",
            "feature extraction time:  8.556830644607544\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03155064582824707\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003234386444091797\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([8527,   11, 8274,  ..., 3789, 9918, 7693], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.049582719802856445\n",
            "Loss: 0.26978403311917826\n",
            "rotation error:  tensor(16.8217, dtype=torch.float64)\n",
            "translation error:  tensor(0.8420, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.984619617462158 seconds ---\n",
            "Processing file: bed_0293\n",
            "feature extraction time:  9.24456238746643\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025644540786743164\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031576156616210938\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 287, 4331, 7803,  ..., 4622, 7639, 5153], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05682682991027832\n",
            "Loss: 0.11683387119148096\n",
            "rotation error:  tensor(340.7969, dtype=torch.float64)\n",
            "translation error:  tensor(0.2373, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.78012180328369 seconds ---\n",
            "Processing file: bed_0294\n",
            "feature extraction time:  8.579015016555786\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.029948711395263672\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031905174255371094\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5219, 9498, 2861,  ..., 6580, 2634, 1344], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05217933654785156\n",
            "Loss: 0.30435531336977717\n",
            "rotation error:  tensor(346.1740, dtype=torch.float64)\n",
            "translation error:  tensor(0.5284, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.02916193008423 seconds ---\n",
            "Epoch: [0/10], Batch: 399, Loss: 0.30435531336977717\n",
            "Processing file: bed_0295\n",
            "feature extraction time:  9.236961126327515\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.023737430572509766\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003224611282348633\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9951, 7929, 5398,  ..., 7633, 8603, 7233], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05423784255981445\n",
            "Loss: 0.3375000695454967\n",
            "rotation error:  tensor(20.1520, dtype=torch.float64)\n",
            "translation error:  tensor(0.8575, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.735342502593994 seconds ---\n",
            "Processing file: bed_0296\n",
            "feature extraction time:  8.608820676803589\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02930450439453125\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003111124038696289\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6088, 9988, 1562,  ..., 2996, 8446, 4075], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0526275634765625\n",
            "Loss: 0.20372720723826881\n",
            "rotation error:  tensor(26.6872, dtype=torch.float64)\n",
            "translation error:  tensor(0.6993, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.044450283050537 seconds ---\n",
            "Processing file: bed_0297\n",
            "feature extraction time:  9.147016763687134\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02524256706237793\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028905868530273438\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([  10, 3772, 7493,  ..., 6181, 7617,  528], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04947328567504883\n",
            "Loss: 0.3189979129966745\n",
            "rotation error:  tensor(12.8512, dtype=torch.float64)\n",
            "translation error:  tensor(0.9534, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.66093635559082 seconds ---\n",
            "Processing file: bed_0298\n",
            "feature extraction time:  8.55413556098938\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03354525566101074\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0035724639892578125\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6130,  305, 7742,  ...,  838,  306, 6892], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.052248477935791016\n",
            "Loss: 0.24893948753871886\n",
            "rotation error:  tensor(181.1618, dtype=torch.float64)\n",
            "translation error:  tensor(0.3150, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.99471354484558 seconds ---\n",
            "Processing file: bed_0299\n",
            "feature extraction time:  9.086682081222534\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02579641342163086\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030181407928466797\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 168, 2170, 7259,  ..., 6368,  617, 4842], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05946660041809082\n",
            "Loss: 0.1602850011022358\n",
            "rotation error:  tensor(296.7850, dtype=torch.float64)\n",
            "translation error:  tensor(0.3648, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.57151508331299 seconds ---\n",
            "Epoch: [0/10], Batch: 404, Loss: 0.1602850011022358\n",
            "Processing file: bed_0300\n",
            "feature extraction time:  8.469149827957153\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.026019811630249023\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003261566162109375\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6480, 5293, 9739,  ..., 1513, 6296, 1724], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.051229238510131836\n",
            "Loss: 0.19538607806477265\n",
            "rotation error:  tensor(27.0119, dtype=torch.float64)\n",
            "translation error:  tensor(0.5893, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.997525930404663 seconds ---\n",
            "Processing file: bed_0301\n",
            "feature extraction time:  9.379190921783447\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02927231788635254\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002986431121826172\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7685, 4153, 2375,  ..., 6087,  215, 8437], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.057105302810668945\n",
            "Loss: 0.28942903495473815\n",
            "rotation error:  tensor(187.3297, dtype=torch.float64)\n",
            "translation error:  tensor(0.4340, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.94060730934143 seconds ---\n",
            "Processing file: bed_0302\n",
            "feature extraction time:  8.575546264648438\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03159904479980469\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003236532211303711\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6181, 7579,    1,  ..., 9321, 1644, 7299], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.049486637115478516\n",
            "Loss: 0.2877499869584203\n",
            "rotation error:  tensor(273.2207, dtype=torch.float64)\n",
            "translation error:  tensor(0.7370, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.166428804397583 seconds ---\n",
            "Processing file: bed_0303\n",
            "feature extraction time:  9.111280679702759\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024582862854003906\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0034677982330322266\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([  60, 9991, 7560,  ..., 9962, 3778, 1743], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05821728706359863\n",
            "Loss: 0.19189496695949618\n",
            "rotation error:  tensor(49.1569, dtype=torch.float64)\n",
            "translation error:  tensor(0.4726, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.602962970733643 seconds ---\n",
            "Processing file: bed_0304\n",
            "feature extraction time:  8.538745164871216\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03304791450500488\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0034034252166748047\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4046,  490, 8008,  ..., 7831, 6678, 4957], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04692363739013672\n",
            "Loss: 0.14289509598263145\n",
            "rotation error:  tensor(12.7212, dtype=torch.float64)\n",
            "translation error:  tensor(0.4312, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.091428756713867 seconds ---\n",
            "Epoch: [0/10], Batch: 409, Loss: 0.14289509598263145\n",
            "Processing file: bed_0305\n",
            "feature extraction time:  9.277526617050171\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02487015724182129\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031785964965820312\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4339, 9931, 5119,  ..., 3929, 3622, 8912], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05358290672302246\n",
            "Loss: 0.20672103665893982\n",
            "rotation error:  tensor(4.6520, dtype=torch.float64)\n",
            "translation error:  tensor(0.5064, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.746152877807617 seconds ---\n",
            "Processing file: bed_0306\n",
            "feature extraction time:  8.602496147155762\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030324935913085938\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003253459930419922\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1185, 8765, 3211,  ..., 3596, 2729, 3111], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05294013023376465\n",
            "Loss: 0.18573778980801808\n",
            "rotation error:  tensor(23.7804, dtype=torch.float64)\n",
            "translation error:  tensor(0.3535, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.03181028366089 seconds ---\n",
            "Processing file: bed_0307\n",
            "feature extraction time:  9.204020738601685\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024774789810180664\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031328201293945312\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3019, 9666, 8061,  ..., 7185, 4249, 7155], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05500340461730957\n",
            "Loss: 0.1663950284935009\n",
            "rotation error:  tensor(6.1527, dtype=torch.float64)\n",
            "translation error:  tensor(0.4743, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.749830961227417 seconds ---\n",
            "Processing file: bed_0308\n",
            "feature extraction time:  8.620213747024536\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.034114837646484375\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0051462650299072266\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1267, 6290, 3310,  ..., 2666, 4825, 6738], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0516660213470459\n",
            "Loss: 0.2192706629203277\n",
            "rotation error:  tensor(8.9636, dtype=torch.float64)\n",
            "translation error:  tensor(0.6562, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.062750577926636 seconds ---\n",
            "Processing file: bed_0309\n",
            "feature extraction time:  9.257349014282227\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024634838104248047\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028443336486816406\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5196,    1, 2867,  ..., 6609, 9406, 2800], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.050894975662231445\n",
            "Loss: 0.19312889423189444\n",
            "rotation error:  tensor(70.3166, dtype=torch.float64)\n",
            "translation error:  tensor(0.5514, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.81258988380432 seconds ---\n",
            "Epoch: [0/10], Batch: 414, Loss: 0.19312889423189444\n",
            "Processing file: bed_0310\n",
            "feature extraction time:  8.61951732635498\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03134298324584961\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.005288362503051758\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2054, 9994, 5624,  ..., 7684, 3586, 1546], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.047399282455444336\n",
            "Loss: 0.24213133936112358\n",
            "rotation error:  tensor(13.5598, dtype=torch.float64)\n",
            "translation error:  tensor(0.5666, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.058990001678467 seconds ---\n",
            "Processing file: bed_0311\n",
            "feature extraction time:  9.333644151687622\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025416851043701172\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030150413513183594\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 586, 3481, 8201,  ..., 3099, 5915, 6819], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.057648658752441406\n",
            "Loss: 0.24467768491441297\n",
            "rotation error:  tensor(21.1703, dtype=torch.float64)\n",
            "translation error:  tensor(0.8109, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.8994083404541 seconds ---\n",
            "Processing file: bed_0312\n",
            "feature extraction time:  8.581432342529297\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.027516603469848633\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031232833862304688\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1798, 8768, 4567,  ..., 6002, 6086, 8940], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05785393714904785\n",
            "Loss: 0.41278213530557517\n",
            "rotation error:  tensor(61.8555, dtype=torch.float64)\n",
            "translation error:  tensor(1.0677, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.18935251235962 seconds ---\n",
            "Processing file: bed_0313\n",
            "feature extraction time:  9.214290142059326\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025962114334106445\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028433799743652344\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6250, 9816, 3758,  ..., 7401, 1125, 6895], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04879331588745117\n",
            "Loss: 0.17678567001526363\n",
            "rotation error:  tensor(11.3181, dtype=torch.float64)\n",
            "translation error:  tensor(0.6227, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.639129400253296 seconds ---\n",
            "Processing file: bed_0314\n",
            "feature extraction time:  8.7091805934906\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.029314756393432617\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031681060791015625\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 164, 9626, 7684,  ..., 5485, 7887, 3603], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05230545997619629\n",
            "Loss: 0.202367670896005\n",
            "rotation error:  tensor(20.2393, dtype=torch.float64)\n",
            "translation error:  tensor(0.5094, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.107134103775024 seconds ---\n",
            "Epoch: [0/10], Batch: 419, Loss: 0.202367670896005\n",
            "Processing file: bed_0315\n",
            "feature extraction time:  9.278443813323975\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024439096450805664\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028352737426757812\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9709, 1874, 4993,  ..., 5067, 2768, 3725], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05272102355957031\n",
            "Loss: 0.2796740594999618\n",
            "rotation error:  tensor(163.6709, dtype=torch.float64)\n",
            "translation error:  tensor(0.5874, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.817811012268066 seconds ---\n",
            "Processing file: bed_0316\n",
            "feature extraction time:  8.621192693710327\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030591964721679688\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0033562183380126953\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2434,  102, 8314,  ..., 7166, 1225, 2132], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0555112361907959\n",
            "Loss: 0.1261479157881039\n",
            "rotation error:  tensor(5.9168, dtype=torch.float64)\n",
            "translation error:  tensor(0.4713, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.994933605194092 seconds ---\n",
            "Processing file: bed_0317\n",
            "feature extraction time:  9.366506338119507\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025348901748657227\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029337406158447266\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9659, 9568, 5500,  ..., 7342, 3184, 9008], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.058173418045043945\n",
            "Loss: 0.19879304795390132\n",
            "rotation error:  tensor(177.3354, dtype=torch.float64)\n",
            "translation error:  tensor(0.3306, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.96207284927368 seconds ---\n",
            "Processing file: bed_0318\n",
            "feature extraction time:  9.004555940628052\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030725955963134766\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003385782241821289\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1604, 7399, 7728,  ..., 1011, 7467, 3881], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.048406362533569336\n",
            "Loss: 0.09882425912209022\n",
            "rotation error:  tensor(13.6133, dtype=torch.float64)\n",
            "translation error:  tensor(0.3117, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.310372352600098 seconds ---\n",
            "Processing file: bed_0319\n",
            "feature extraction time:  9.313664197921753\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024603843688964844\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031337738037109375\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7886, 3533, 1103,  ..., 3229, 7951, 7777], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04911994934082031\n",
            "Loss: 0.2455804654937445\n",
            "rotation error:  tensor(70.8560, dtype=torch.float64)\n",
            "translation error:  tensor(0.5054, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.864971160888672 seconds ---\n",
            "Epoch: [0/10], Batch: 424, Loss: 0.2455804654937445\n",
            "Processing file: bed_0320\n",
            "feature extraction time:  8.860247611999512\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0298159122467041\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031290054321289062\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9893,  524, 9798,  ..., 4251, 8864, 3257], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.052390098571777344\n",
            "Loss: 0.16574512948044226\n",
            "rotation error:  tensor(5.9179, dtype=torch.float64)\n",
            "translation error:  tensor(0.6404, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.13121461868286 seconds ---\n",
            "Processing file: bed_0321\n",
            "feature extraction time:  9.359748840332031\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02637004852294922\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0037183761596679688\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4788, 1025, 7080,  ..., 8128, 8454, 3858], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05527448654174805\n",
            "Loss: 0.20487677649201586\n",
            "rotation error:  tensor(12.6596, dtype=torch.float64)\n",
            "translation error:  tensor(0.5879, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 19.058616161346436 seconds ---\n",
            "Processing file: bed_0322\n",
            "feature extraction time:  8.967934608459473\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.031020402908325195\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0036802291870117188\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4111, 3654, 7689,  ..., 7657, 8889, 9952], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05853533744812012\n",
            "Loss: 0.18985739568518026\n",
            "rotation error:  tensor(283.9329, dtype=torch.float64)\n",
            "translation error:  tensor(0.3822, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.202033042907715 seconds ---\n",
            "Processing file: bed_0323\n",
            "feature extraction time:  9.29975414276123\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02498650550842285\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029954910278320312\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9940, 5688, 3344,  ..., 7451, 9475, 9858], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.053093671798706055\n",
            "Loss: 0.13157094158331256\n",
            "rotation error:  tensor(6.7988, dtype=torch.float64)\n",
            "translation error:  tensor(0.3256, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.82974338531494 seconds ---\n",
            "Processing file: bed_0324\n",
            "feature extraction time:  8.945627689361572\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03049159049987793\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003355264663696289\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6710, 3090,    4,  ..., 8520,  919,  332], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.048139095306396484\n",
            "Loss: 0.10911860845048785\n",
            "rotation error:  tensor(10.6975, dtype=torch.float64)\n",
            "translation error:  tensor(0.3152, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.0825412273407 seconds ---\n",
            "Epoch: [0/10], Batch: 429, Loss: 0.10911860845048785\n",
            "Processing file: bed_0325\n",
            "feature extraction time:  9.27826738357544\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024698495864868164\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028498172760009766\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1400, 4346,  434,  ..., 4837, 5449, 4010], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.051525115966796875\n",
            "Loss: 0.23523108553102598\n",
            "rotation error:  tensor(22.5555, dtype=torch.float64)\n",
            "translation error:  tensor(0.6181, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.824804067611694 seconds ---\n",
            "Processing file: bed_0326\n",
            "feature extraction time:  8.935609340667725\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0317845344543457\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003040790557861328\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5369, 1338, 9881,  ..., 8295, 7727, 4060], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05537557601928711\n",
            "Loss: 0.04684440863536451\n",
            "rotation error:  tensor(24.6607, dtype=torch.float64)\n",
            "translation error:  tensor(0.0799, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.18111300468445 seconds ---\n",
            "Processing file: bed_0327\n",
            "feature extraction time:  9.385977506637573\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024651050567626953\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002858400344848633\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6068, 6317, 7636,  ..., 3686, 4214, 8208], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0518803596496582\n",
            "Loss: 0.23806750467368243\n",
            "rotation error:  tensor(198.1646, dtype=torch.float64)\n",
            "translation error:  tensor(0.3119, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.882861137390137 seconds ---\n",
            "Processing file: bed_0328\n",
            "feature extraction time:  9.033601999282837\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.031190872192382812\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.004132747650146484\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 162, 9381, 9889,  ...,   31, 9343, 4879], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.050428152084350586\n",
            "Loss: 0.19655463861879197\n",
            "rotation error:  tensor(11.6755, dtype=torch.float64)\n",
            "translation error:  tensor(0.4205, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.064891815185547 seconds ---\n",
            "Processing file: bed_0329\n",
            "feature extraction time:  9.290555000305176\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02487802505493164\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002930164337158203\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2839, 8488, 4549,  ..., 5172, 5626, 6685], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05829596519470215\n",
            "Loss: 0.21461280032121305\n",
            "rotation error:  tensor(268.9576, dtype=torch.float64)\n",
            "translation error:  tensor(0.5247, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.80522656440735 seconds ---\n",
            "Epoch: [0/10], Batch: 434, Loss: 0.21461280032121305\n",
            "Processing file: bed_0330\n",
            "feature extraction time:  8.986548900604248\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03273296356201172\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0036208629608154297\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4110, 9667, 2128,  ..., 3213, 5436, 5954], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05068492889404297\n",
            "Loss: 0.3197667171157949\n",
            "rotation error:  tensor(306.9451, dtype=torch.float64)\n",
            "translation error:  tensor(1.1588, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.06041193008423 seconds ---\n",
            "Processing file: bed_0331\n",
            "feature extraction time:  9.216983556747437\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025319576263427734\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003108501434326172\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 376, 6859, 1344,  ..., 1553, 9650, 3921], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05947756767272949\n",
            "Loss: 0.14702442859724624\n",
            "rotation error:  tensor(6.2963, dtype=torch.float64)\n",
            "translation error:  tensor(0.4274, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.684355974197388 seconds ---\n",
            "Processing file: bed_0332\n",
            "feature extraction time:  9.117952585220337\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03181743621826172\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.000782012939453125\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4448, 5834, 6029,  ..., 2887, 8031, 7704], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04926156997680664\n",
            "Loss: 0.1090989666082569\n",
            "rotation error:  tensor(20.5064, dtype=torch.float64)\n",
            "translation error:  tensor(0.4220, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.20978546142578 seconds ---\n",
            "Processing file: bed_0333\n",
            "feature extraction time:  9.152915000915527\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024996042251586914\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003336668014526367\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2228, 9648, 9404,  ...,  516, 3239,  574], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05038619041442871\n",
            "Loss: 0.11948172974636313\n",
            "rotation error:  tensor(34.9407, dtype=torch.float64)\n",
            "translation error:  tensor(0.4196, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.58596110343933 seconds ---\n",
            "Processing file: bed_0334\n",
            "feature extraction time:  8.888134241104126\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030169248580932617\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003125905990600586\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5044, 8732, 2634,  ..., 4218, 4605, 4414], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.051680803298950195\n",
            "Loss: 0.15557063778246308\n",
            "rotation error:  tensor(70.6333, dtype=torch.float64)\n",
            "translation error:  tensor(0.4539, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.003114223480225 seconds ---\n",
            "Epoch: [0/10], Batch: 439, Loss: 0.15557063778246308\n",
            "Processing file: bed_0335\n",
            "feature extraction time:  9.175395488739014\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.023848533630371094\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029120445251464844\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5012, 2388, 6168,  ..., 7600, 8395, 5007], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04848051071166992\n",
            "Loss: 0.31881477381679785\n",
            "rotation error:  tensor(194.3158, dtype=torch.float64)\n",
            "translation error:  tensor(0.7458, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.5832359790802 seconds ---\n",
            "Processing file: bed_0336\n",
            "feature extraction time:  9.002799272537231\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03008723258972168\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003495931625366211\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 320, 7554, 9707,  ..., 2196, 7744, 2211], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04977703094482422\n",
            "Loss: 0.3899195616400617\n",
            "rotation error:  tensor(395.0280, dtype=torch.float64)\n",
            "translation error:  tensor(1.2034, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.133765697479248 seconds ---\n",
            "Processing file: bed_0337\n",
            "feature extraction time:  9.230058908462524\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0250856876373291\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002935171127319336\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9915,  296, 7851,  ..., 1144, 6045, 3831], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.054535627365112305\n",
            "Loss: 0.20539798613880048\n",
            "rotation error:  tensor(311.0589, dtype=torch.float64)\n",
            "translation error:  tensor(0.3509, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.612420558929443 seconds ---\n",
            "Processing file: bed_0338\n",
            "feature extraction time:  8.677046060562134\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03014850616455078\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003170490264892578\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 887, 9893, 4433,  ..., 6607, 3180, 9168], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05434131622314453\n",
            "Loss: 0.23721320560281936\n",
            "rotation error:  tensor(184.8021, dtype=torch.float64)\n",
            "translation error:  tensor(0.5462, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.937475442886353 seconds ---\n",
            "Processing file: bed_0339\n",
            "feature extraction time:  9.132946014404297\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024401426315307617\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002831697463989258\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6175,  999,  491,  ..., 6620, 6108,  339], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0584261417388916\n",
            "Loss: 0.13453817690844624\n",
            "rotation error:  tensor(7.3014, dtype=torch.float64)\n",
            "translation error:  tensor(0.3657, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.531527280807495 seconds ---\n",
            "Epoch: [0/10], Batch: 444, Loss: 0.13453817690844624\n",
            "Processing file: bed_0340\n",
            "feature extraction time:  8.799201726913452\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03547787666320801\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0033516883850097656\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7987,  154, 4396,  ..., 3058, 4568, 4582], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05103349685668945\n",
            "Loss: 0.09316281836858771\n",
            "rotation error:  tensor(11.1472, dtype=torch.float64)\n",
            "translation error:  tensor(0.3549, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.125730752944946 seconds ---\n",
            "Processing file: bed_0341\n",
            "feature extraction time:  9.281420707702637\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024968624114990234\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029096603393554688\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2082, 9966, 9801,  ..., 9230, 2222, 7919], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05271649360656738\n",
            "Loss: 0.3804385426804787\n",
            "rotation error:  tensor(107.8728, dtype=torch.float64)\n",
            "translation error:  tensor(1.0037, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.83678126335144 seconds ---\n",
            "Processing file: bed_0342\n",
            "feature extraction time:  8.962871313095093\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03609943389892578\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.004631519317626953\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 540, 6720,  741,  ...,  642, 3750, 8728], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.057847023010253906\n",
            "Loss: 0.25815121071780406\n",
            "rotation error:  tensor(226.9090, dtype=torch.float64)\n",
            "translation error:  tensor(0.7624, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.243868827819824 seconds ---\n",
            "Processing file: bed_0343\n",
            "feature extraction time:  9.265146970748901\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025540828704833984\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030031204223632812\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5149, 1453, 2546,  ..., 1333, 8205, 2269], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.047926902770996094\n",
            "Loss: 0.3820399354288956\n",
            "rotation error:  tensor(272.8623, dtype=torch.float64)\n",
            "translation error:  tensor(1.0120, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.753278255462646 seconds ---\n",
            "Processing file: bed_0344\n",
            "feature extraction time:  8.676216840744019\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03014230728149414\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003148317337036133\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([  30, 9781, 2860,  ..., 8765, 1485,  451], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.048416852951049805\n",
            "Loss: 0.39935367582810755\n",
            "rotation error:  tensor(57.7380, dtype=torch.float64)\n",
            "translation error:  tensor(1.2210, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.988117456436157 seconds ---\n",
            "Epoch: [0/10], Batch: 449, Loss: 0.39935367582810755\n",
            "Processing file: bed_0345\n",
            "feature extraction time:  9.214775085449219\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025168180465698242\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0027315616607666016\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1964, 9521, 8307,  ..., 8146, 9686, 2955], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.049028873443603516\n",
            "Loss: 0.3598313917373728\n",
            "rotation error:  tensor(356.2674, dtype=torch.float64)\n",
            "translation error:  tensor(1.1803, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.68886971473694 seconds ---\n",
            "Processing file: bed_0346\n",
            "feature extraction time:  8.72647500038147\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03185606002807617\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031943321228027344\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7517, 6200, 7666,  ..., 5466,  909, 6316], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0562281608581543\n",
            "Loss: 0.2311662911803909\n",
            "rotation error:  tensor(86.1327, dtype=torch.float64)\n",
            "translation error:  tensor(0.5251, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.06922197341919 seconds ---\n",
            "Processing file: bed_0347\n",
            "feature extraction time:  9.182446479797363\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02383279800415039\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003320455551147461\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2275, 5929, 4148,  ..., 2329, 9590, 4739], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05907773971557617\n",
            "Loss: 0.14352614864468238\n",
            "rotation error:  tensor(7.1459, dtype=torch.float64)\n",
            "translation error:  tensor(0.3735, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.663382291793823 seconds ---\n",
            "Processing file: bed_0348\n",
            "feature extraction time:  8.566805124282837\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.029791831970214844\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0035216808319091797\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([  63, 9502, 2864,  ..., 9971, 5591, 1379], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04945969581604004\n",
            "Loss: 0.14936148247478853\n",
            "rotation error:  tensor(16.3202, dtype=torch.float64)\n",
            "translation error:  tensor(0.5451, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.973828315734863 seconds ---\n",
            "Processing file: bed_0349\n",
            "feature extraction time:  9.193490505218506\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0244143009185791\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031173229217529297\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([8666, 2327, 5581,  ..., 7192, 6804,    2], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05280733108520508\n",
            "Loss: 0.09298188659105819\n",
            "rotation error:  tensor(58.6209, dtype=torch.float64)\n",
            "translation error:  tensor(0.2252, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.64970827102661 seconds ---\n",
            "Epoch: [0/10], Batch: 454, Loss: 0.09298188659105819\n",
            "Processing file: bed_0350\n",
            "feature extraction time:  8.58372449874878\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.029900550842285156\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.00315093994140625\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1206,    1, 9802,  ..., 8423, 9711, 5160], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0596165657043457\n",
            "Loss: 0.26581878049629304\n",
            "rotation error:  tensor(357.9814, dtype=torch.float64)\n",
            "translation error:  tensor(0.5211, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.913393259048462 seconds ---\n",
            "Processing file: bed_0351\n",
            "feature extraction time:  9.199293375015259\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025145292282104492\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0027844905853271484\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7398, 8580, 2548,  ..., 7768, 3430, 3874], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.052362918853759766\n",
            "Loss: 0.3437217624298157\n",
            "rotation error:  tensor(20.2370, dtype=torch.float64)\n",
            "translation error:  tensor(1.1697, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.72137975692749 seconds ---\n",
            "Processing file: bed_0352\n",
            "feature extraction time:  8.469182968139648\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03351926803588867\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0034940242767333984\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7569, 9922,   97,  ..., 3607, 4639, 2850], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05476212501525879\n",
            "Loss: 0.18212012171371536\n",
            "rotation error:  tensor(8.1224, dtype=torch.float64)\n",
            "translation error:  tensor(0.3378, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.035979747772217 seconds ---\n",
            "Processing file: bed_0353\n",
            "feature extraction time:  9.17512845993042\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024777889251708984\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002926349639892578\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9628, 1841, 6039,  ..., 6012, 9925, 9754], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04491591453552246\n",
            "Loss: 0.20657473753619982\n",
            "rotation error:  tensor(58.3757, dtype=torch.float64)\n",
            "translation error:  tensor(0.5180, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.765169382095337 seconds ---\n",
            "Processing file: bed_0354\n",
            "feature extraction time:  8.487433195114136\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03078293800354004\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0038847923278808594\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2852, 9637, 9592,  ...,  605, 6066, 3002], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.045636653900146484\n",
            "Loss: 0.14135803401585778\n",
            "rotation error:  tensor(26.5478, dtype=torch.float64)\n",
            "translation error:  tensor(0.4179, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.10464644432068 seconds ---\n",
            "Epoch: [0/10], Batch: 459, Loss: 0.14135803401585778\n",
            "Processing file: bed_0355\n",
            "feature extraction time:  9.14905333518982\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.023996353149414062\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002898693084716797\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 804, 5448, 8055,  ..., 2989, 6001, 2162], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05558896064758301\n",
            "Loss: 0.0986595910538999\n",
            "rotation error:  tensor(21.8389, dtype=torch.float64)\n",
            "translation error:  tensor(0.4166, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.630839109420776 seconds ---\n",
            "Processing file: bed_0356\n",
            "feature extraction time:  8.581138372421265\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030878782272338867\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002929210662841797\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9783, 6866, 1883,  ..., 4454, 2642, 5499], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05532431602478027\n",
            "Loss: 0.21794714686556782\n",
            "rotation error:  tensor(198.1189, dtype=torch.float64)\n",
            "translation error:  tensor(0.4153, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.02333116531372 seconds ---\n",
            "Processing file: bed_0357\n",
            "feature extraction time:  9.220643281936646\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024639129638671875\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002971649169921875\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4018, 6233, 1290,  ..., 7245, 3142, 1609], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.049573659896850586\n",
            "Loss: 0.37485232537550495\n",
            "rotation error:  tensor(13.2298, dtype=torch.float64)\n",
            "translation error:  tensor(0.8693, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.651135206222534 seconds ---\n",
            "Processing file: bed_0358\n",
            "feature extraction time:  8.366138458251953\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.023336410522460938\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003053426742553711\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9879, 3843, 1424,  ..., 8042, 8940, 4449], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.054230690002441406\n",
            "Loss: 0.19001855465583872\n",
            "rotation error:  tensor(8.9934, dtype=torch.float64)\n",
            "translation error:  tensor(0.5642, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.88335657119751 seconds ---\n",
            "Processing file: bed_0359\n",
            "feature extraction time:  9.21714472770691\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024680614471435547\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0026955604553222656\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6188, 6760, 3008,  ..., 1806, 5335,  160], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04962420463562012\n",
            "Loss: 0.29571262336531523\n",
            "rotation error:  tensor(123.7321, dtype=torch.float64)\n",
            "translation error:  tensor(0.7825, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.788957118988037 seconds ---\n",
            "Epoch: [0/10], Batch: 464, Loss: 0.29571262336531523\n",
            "Processing file: bed_0360\n",
            "feature extraction time:  8.523344993591309\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025506973266601562\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030672550201416016\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   2, 8368, 9843,  ..., 7202, 2299, 8377], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04680180549621582\n",
            "Loss: 0.31347981206158654\n",
            "rotation error:  tensor(6.7518, dtype=torch.float64)\n",
            "translation error:  tensor(1.0818, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.036698818206787 seconds ---\n",
            "Processing file: bed_0361\n",
            "feature extraction time:  9.262052536010742\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02445840835571289\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030651092529296875\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   1, 1429, 5176,  ...,  650, 6114, 5857], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0441281795501709\n",
            "Loss: 0.3324602732388179\n",
            "rotation error:  tensor(212.6972, dtype=torch.float64)\n",
            "translation error:  tensor(0.7445, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.696890830993652 seconds ---\n",
            "Processing file: bed_0362\n",
            "feature extraction time:  8.490662097930908\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024899959564208984\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028944015502929688\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1695, 5558, 3709,  ..., 3217, 3729, 2015], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0551910400390625\n",
            "Loss: 0.31453792833788746\n",
            "rotation error:  tensor(17.6902, dtype=torch.float64)\n",
            "translation error:  tensor(0.9877, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.123239517211914 seconds ---\n",
            "Processing file: bed_0363\n",
            "feature extraction time:  9.288549184799194\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02506542205810547\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0032606124877929688\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5467,    2, 9704,  ..., 1098, 4092, 7130], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04692363739013672\n",
            "Loss: 0.2667392007018593\n",
            "rotation error:  tensor(23.7956, dtype=torch.float64)\n",
            "translation error:  tensor(0.9337, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.825915098190308 seconds ---\n",
            "Processing file: bed_0364\n",
            "feature extraction time:  8.498001098632812\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.032160043716430664\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003786802291870117\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4294, 9727, 8357,  ..., 7814, 3322, 8787], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05304908752441406\n",
            "Loss: 0.1621920363081619\n",
            "rotation error:  tensor(20.4690, dtype=torch.float64)\n",
            "translation error:  tensor(0.4746, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.911256074905396 seconds ---\n",
            "Epoch: [0/10], Batch: 469, Loss: 0.1621920363081619\n",
            "Processing file: bed_0365\n",
            "feature extraction time:  9.1985342502594\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024393081665039062\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029747486114501953\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9941,  421, 6375,  ..., 7658, 5573,  973], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.055098533630371094\n",
            "Loss: 0.2601826420099321\n",
            "rotation error:  tensor(280.6156, dtype=torch.float64)\n",
            "translation error:  tensor(0.6122, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.60987615585327 seconds ---\n",
            "Processing file: bed_0366\n",
            "feature extraction time:  8.53864312171936\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025986433029174805\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003320932388305664\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9807, 3822,   33,  ..., 5681,  116, 6947], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.055132389068603516\n",
            "Loss: 0.14869523465466414\n",
            "rotation error:  tensor(12.5084, dtype=torch.float64)\n",
            "translation error:  tensor(0.5688, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.92985439300537 seconds ---\n",
            "Processing file: bed_0367\n",
            "feature extraction time:  9.261123418807983\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0251619815826416\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003274679183959961\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7028, 5294,  107,  ..., 4785, 3361, 5549], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05154109001159668\n",
            "Loss: 0.4294778881850214\n",
            "rotation error:  tensor(9.9135, dtype=torch.float64)\n",
            "translation error:  tensor(1.1720, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.82888913154602 seconds ---\n",
            "Processing file: bed_0368\n",
            "feature extraction time:  8.481667041778564\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025130510330200195\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028791427612304688\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4121,  243, 7254,  ..., 5948, 8805, 3185], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05651545524597168\n",
            "Loss: 0.12373931757689903\n",
            "rotation error:  tensor(22.1130, dtype=torch.float64)\n",
            "translation error:  tensor(0.3377, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.99416971206665 seconds ---\n",
            "Processing file: bed_0369\n",
            "feature extraction time:  9.240771532058716\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024852752685546875\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029451847076416016\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5215, 9700, 8320,  ..., 5682, 6813, 7463], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0469968318939209\n",
            "Loss: 0.296376968386862\n",
            "rotation error:  tensor(13.3132, dtype=torch.float64)\n",
            "translation error:  tensor(0.7222, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.792343378067017 seconds ---\n",
            "Epoch: [0/10], Batch: 474, Loss: 0.296376968386862\n",
            "Processing file: bed_0370\n",
            "feature extraction time:  8.446621179580688\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024061203002929688\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028591156005859375\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([  11, 2563, 3974,  ..., 2056,   25, 5979], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04719996452331543\n",
            "Loss: 0.34599284100472033\n",
            "rotation error:  tensor(12.2300, dtype=torch.float64)\n",
            "translation error:  tensor(0.8313, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.004992246627808 seconds ---\n",
            "Processing file: bed_0371\n",
            "feature extraction time:  9.360718488693237\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025362014770507812\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002796649932861328\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7847, 6652, 9861,  ..., 2494, 1736, 5722], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.044628143310546875\n",
            "Loss: 0.24758624552370442\n",
            "rotation error:  tensor(7.6679, dtype=torch.float64)\n",
            "translation error:  tensor(0.7921, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.888225078582764 seconds ---\n",
            "Processing file: bed_0372\n",
            "feature extraction time:  8.572967767715454\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03249382972717285\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.00357818603515625\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2429, 7146, 4704,  ..., 7605, 6825, 4313], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04842329025268555\n",
            "Loss: 0.1883153250923062\n",
            "rotation error:  tensor(33.9169, dtype=torch.float64)\n",
            "translation error:  tensor(0.7358, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.14964199066162 seconds ---\n",
            "Processing file: bed_0373\n",
            "feature extraction time:  9.194804906845093\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025531291961669922\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029344558715820312\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([  19, 6264, 2574,  ..., 3837, 6686,   11], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05092000961303711\n",
            "Loss: 0.33939520078784824\n",
            "rotation error:  tensor(43.7492, dtype=torch.float64)\n",
            "translation error:  tensor(1.1742, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.639583349227905 seconds ---\n",
            "Processing file: bed_0374\n",
            "feature extraction time:  8.500428438186646\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025691986083984375\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003169536590576172\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([  10, 7688, 7629,  ..., 9440, 2239, 9004], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.052345991134643555\n",
            "Loss: 0.22220965357878386\n",
            "rotation error:  tensor(43.1460, dtype=torch.float64)\n",
            "translation error:  tensor(0.7490, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.957311630249023 seconds ---\n",
            "Epoch: [0/10], Batch: 479, Loss: 0.22220965357878386\n",
            "Processing file: bed_0375\n",
            "feature extraction time:  9.259795904159546\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02541041374206543\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028569698333740234\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3575, 6153, 5231,  ..., 1017, 7048, 1638], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05677533149719238\n",
            "Loss: 0.1653402916367001\n",
            "rotation error:  tensor(15.5594, dtype=torch.float64)\n",
            "translation error:  tensor(0.5228, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.85219717025757 seconds ---\n",
            "Processing file: bed_0376\n",
            "feature extraction time:  8.548120021820068\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02528238296508789\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002814054489135742\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6394, 2423, 2819,  ..., 7720, 3159, 3869], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05393815040588379\n",
            "Loss: 0.29742761871340434\n",
            "rotation error:  tensor(165.3098, dtype=torch.float64)\n",
            "translation error:  tensor(0.8567, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.121922969818115 seconds ---\n",
            "Processing file: bed_0377\n",
            "feature extraction time:  9.292306184768677\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0247957706451416\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002977132797241211\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   3, 9389, 6387,  ..., 6170, 5831, 7961], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.052580833435058594\n",
            "Loss: 0.21853166815702052\n",
            "rotation error:  tensor(268.2000, dtype=torch.float64)\n",
            "translation error:  tensor(0.4856, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.83018660545349 seconds ---\n",
            "Processing file: bed_0378\n",
            "feature extraction time:  8.539453506469727\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024640560150146484\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028536319732666016\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2648, 7084, 4686,  ..., 7883, 1726, 3110], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05216097831726074\n",
            "Loss: 0.10869250233213326\n",
            "rotation error:  tensor(8.0743, dtype=torch.float64)\n",
            "translation error:  tensor(0.3230, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.004350900650024 seconds ---\n",
            "Processing file: bed_0379\n",
            "feature extraction time:  9.26367712020874\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024593353271484375\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003088712692260742\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3545, 1845, 5444,  ..., 1194, 2514, 1192], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05682563781738281\n",
            "Loss: 0.16681329851869187\n",
            "rotation error:  tensor(12.3056, dtype=torch.float64)\n",
            "translation error:  tensor(0.5280, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.798194408416748 seconds ---\n",
            "Epoch: [0/10], Batch: 484, Loss: 0.16681329851869187\n",
            "Processing file: bed_0380\n",
            "feature extraction time:  8.591075420379639\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03167724609375\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003515005111694336\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 440, 8385, 8350,  ..., 9788, 4172, 6584], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0548090934753418\n",
            "Loss: 0.03711680063067997\n",
            "rotation error:  tensor(6.8732, dtype=torch.float64)\n",
            "translation error:  tensor(0.0936, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.119250535964966 seconds ---\n",
            "Processing file: bed_0381\n",
            "feature extraction time:  9.231102228164673\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024529457092285156\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002829313278198242\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 361, 2909, 9690,  ..., 6318, 4119, 9477], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04819321632385254\n",
            "Loss: 0.40343820829533145\n",
            "rotation error:  tensor(343.8714, dtype=torch.float64)\n",
            "translation error:  tensor(0.9227, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.761046409606934 seconds ---\n",
            "Processing file: bed_0382\n",
            "feature extraction time:  8.584667444229126\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03368091583251953\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002947568893432617\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5778, 9202, 2906,  ..., 2712, 3297, 2150], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.057167768478393555\n",
            "Loss: 0.3019545986641249\n",
            "rotation error:  tensor(71.3943, dtype=torch.float64)\n",
            "translation error:  tensor(0.6271, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.1357684135437 seconds ---\n",
            "Processing file: bed_0383\n",
            "feature extraction time:  9.23103141784668\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024654865264892578\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002859354019165039\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 610, 2226, 6804,  ..., 8929, 3437, 7913], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04933309555053711\n",
            "Loss: 0.33856634141827713\n",
            "rotation error:  tensor(280.8137, dtype=torch.float64)\n",
            "translation error:  tensor(1.0253, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.665504455566406 seconds ---\n",
            "Processing file: bed_0384\n",
            "feature extraction time:  8.410625219345093\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025448083877563477\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003198385238647461\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9958, 7428, 4052,  ..., 2306, 4924, 7719], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05306720733642578\n",
            "Loss: 0.3718126855619446\n",
            "rotation error:  tensor(53.4722, dtype=torch.float64)\n",
            "translation error:  tensor(0.9527, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.94289207458496 seconds ---\n",
            "Epoch: [0/10], Batch: 489, Loss: 0.3718126855619446\n",
            "Processing file: bed_0385\n",
            "feature extraction time:  9.278321266174316\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02492213249206543\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029990673065185547\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 880, 3365, 3360,  ..., 3615,  472, 5206], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05075407028198242\n",
            "Loss: 0.21451203234376165\n",
            "rotation error:  tensor(269.7897, dtype=torch.float64)\n",
            "translation error:  tensor(0.5420, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.84402370452881 seconds ---\n",
            "Processing file: bed_0386\n",
            "feature extraction time:  8.570831060409546\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.034337520599365234\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0037212371826171875\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9761, 4376, 2925,  ..., 1639, 3204, 1183], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04469871520996094\n",
            "Loss: 0.2832472885876399\n",
            "rotation error:  tensor(250.4803, dtype=torch.float64)\n",
            "translation error:  tensor(0.9395, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.015159368515015 seconds ---\n",
            "Processing file: bed_0387\n",
            "feature extraction time:  9.140963315963745\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02540874481201172\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030672550201416016\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3155, 6595, 1480,  ..., 6099, 5018, 3882], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.06373429298400879\n",
            "Loss: 0.2135525363946321\n",
            "rotation error:  tensor(130.7513, dtype=torch.float64)\n",
            "translation error:  tensor(0.5143, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.689882516860962 seconds ---\n",
            "Processing file: bed_0388\n",
            "feature extraction time:  8.521830558776855\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.033951759338378906\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0038843154907226562\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7670, 6247, 9911,  ..., 9750, 7493, 9777], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04991412162780762\n",
            "Loss: 0.08846653831302603\n",
            "rotation error:  tensor(8.2308, dtype=torch.float64)\n",
            "translation error:  tensor(0.2971, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.107847690582275 seconds ---\n",
            "Processing file: bed_0389\n",
            "feature extraction time:  9.298644304275513\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024569272994995117\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029952526092529297\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   8, 7626, 7677,  ..., 8295,  893, 5870], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05456995964050293\n",
            "Loss: 0.3828688787915803\n",
            "rotation error:  tensor(137.6064, dtype=torch.float64)\n",
            "translation error:  tensor(1.0496, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.847081661224365 seconds ---\n",
            "Epoch: [0/10], Batch: 494, Loss: 0.3828688787915803\n",
            "Processing file: bed_0390\n",
            "feature extraction time:  8.522751569747925\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03320455551147461\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.00272369384765625\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7357, 4419,  922,  ..., 2514, 7780, 9838], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05220770835876465\n",
            "Loss: 0.20790213889200282\n",
            "rotation error:  tensor(55.4097, dtype=torch.float64)\n",
            "translation error:  tensor(0.5252, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.940842151641846 seconds ---\n",
            "Processing file: bed_0391\n",
            "feature extraction time:  9.195320129394531\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02561807632446289\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031807422637939453\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6441, 4403, 7707,  ...,  811, 8068, 5698], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.06358098983764648\n",
            "Loss: 0.013584540679189636\n",
            "rotation error:  tensor(4.7078, dtype=torch.float64)\n",
            "translation error:  tensor(0.0502, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.681607961654663 seconds ---\n",
            "Processing file: bed_0392\n",
            "feature extraction time:  8.545870542526245\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03433799743652344\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0016579627990722656\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5308, 6265, 2797,  ..., 9878, 1158, 9953], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04951286315917969\n",
            "Loss: 0.12444160076901344\n",
            "rotation error:  tensor(5.1703, dtype=torch.float64)\n",
            "translation error:  tensor(0.2453, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.040789127349854 seconds ---\n",
            "Processing file: bed_0393\n",
            "feature extraction time:  9.119259119033813\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024564027786254883\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0027933120727539062\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 423, 9721, 4188,  ..., 7594, 4087, 6345], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.057805538177490234\n",
            "Loss: 0.21348971765982183\n",
            "rotation error:  tensor(24.4018, dtype=torch.float64)\n",
            "translation error:  tensor(0.5622, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.63105058670044 seconds ---\n",
            "Processing file: bed_0394\n",
            "feature extraction time:  8.57664680480957\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.034207820892333984\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003046751022338867\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2950, 9730, 5400,  ..., 2785, 1479, 8991], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04702258110046387\n",
            "Loss: 0.18186414844190746\n",
            "rotation error:  tensor(348.0366, dtype=torch.float64)\n",
            "translation error:  tensor(0.6301, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.13434147834778 seconds ---\n",
            "Epoch: [0/10], Batch: 499, Loss: 0.18186414844190746\n",
            "Processing file: bed_0395\n",
            "feature extraction time:  9.23216438293457\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.026397705078125\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002825021743774414\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4497, 1069, 7511,  ..., 5831,  958, 8092], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.050252676010131836\n",
            "Loss: 0.4105233182350959\n",
            "rotation error:  tensor(171.8541, dtype=torch.float64)\n",
            "translation error:  tensor(0.6961, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.736205101013184 seconds ---\n",
            "Processing file: bed_0396\n",
            "feature extraction time:  8.602598905563354\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03238630294799805\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.004912614822387695\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9340,  262, 4236,  ..., 9810, 9540, 6230], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04946541786193848\n",
            "Loss: 0.28901932517152706\n",
            "rotation error:  tensor(140.6100, dtype=torch.float64)\n",
            "translation error:  tensor(0.7354, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.133753061294556 seconds ---\n",
            "Processing file: bed_0397\n",
            "feature extraction time:  9.258933067321777\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.023805856704711914\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002905130386352539\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4288, 3542, 1242,  ...,  522, 7550, 1742], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.051656484603881836\n",
            "Loss: 0.2517310312042292\n",
            "rotation error:  tensor(333.2851, dtype=torch.float64)\n",
            "translation error:  tensor(0.9355, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.814656734466553 seconds ---\n",
            "Processing file: bed_0398\n",
            "feature extraction time:  8.529314994812012\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03140544891357422\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028040409088134766\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1639, 2828, 4704,  ..., 1400, 6292, 8391], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.045276641845703125\n",
            "Loss: 0.21391700256801405\n",
            "rotation error:  tensor(14.6174, dtype=torch.float64)\n",
            "translation error:  tensor(0.6109, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.06883192062378 seconds ---\n",
            "Processing file: bed_0399\n",
            "feature extraction time:  9.256962776184082\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025818586349487305\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003506183624267578\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9790, 1570, 6958,  ..., 5758, 7847, 7866], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.046504974365234375\n",
            "Loss: 0.24115475821326116\n",
            "rotation error:  tensor(18.5277, dtype=torch.float64)\n",
            "translation error:  tensor(0.6180, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.743638277053833 seconds ---\n",
            "Epoch: [0/10], Batch: 504, Loss: 0.24115475821326116\n",
            "Processing file: bed_0400\n",
            "feature extraction time:  8.509730339050293\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03479647636413574\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0032072067260742188\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5408,    5, 4308,  ..., 3639, 9200, 4654], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04134416580200195\n",
            "Loss: 0.16658209997994206\n",
            "rotation error:  tensor(14.5027, dtype=torch.float64)\n",
            "translation error:  tensor(0.5688, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.993276834487915 seconds ---\n",
            "Processing file: bed_0401\n",
            "feature extraction time:  9.23034381866455\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024995803833007812\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002927541732788086\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9678, 4682, 1211,  ..., 9786, 4047, 1257], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.056043148040771484\n",
            "Loss: 0.20563952409110037\n",
            "rotation error:  tensor(320.2936, dtype=torch.float64)\n",
            "translation error:  tensor(0.4980, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.693276405334473 seconds ---\n",
            "Processing file: bed_0402\n",
            "feature extraction time:  8.588412761688232\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.031491994857788086\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.004048347473144531\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 986, 4173, 5973,  ..., 3563, 2187, 5010], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.051839351654052734\n",
            "Loss: 0.4295133629071849\n",
            "rotation error:  tensor(180.2568, dtype=torch.float64)\n",
            "translation error:  tensor(1.3148, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.990813732147217 seconds ---\n",
            "Processing file: bed_0403\n",
            "feature extraction time:  9.16646933555603\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.026976585388183594\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002878427505493164\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 853, 9767,  215,  ..., 2775, 5078,  232], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.050666093826293945\n",
            "Loss: 0.2987148258965702\n",
            "rotation error:  tensor(322.8461, dtype=torch.float64)\n",
            "translation error:  tensor(1.0675, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.64867639541626 seconds ---\n",
            "Processing file: bed_0404\n",
            "feature extraction time:  8.479270696640015\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.032817840576171875\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0027239322662353516\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 454, 3233, 1078,  ..., 6542, 5436,  426], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.047756195068359375\n",
            "Loss: 0.29387857154818686\n",
            "rotation error:  tensor(9.5596, dtype=torch.float64)\n",
            "translation error:  tensor(0.9899, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.95667791366577 seconds ---\n",
            "Epoch: [0/10], Batch: 509, Loss: 0.29387857154818686\n",
            "Processing file: bed_0405\n",
            "feature extraction time:  9.270439863204956\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025815963745117188\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003168821334838867\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4284, 4252,    4,  ..., 1451, 1192, 5264], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05569171905517578\n",
            "Loss: 0.26361548275480573\n",
            "rotation error:  tensor(342.1462, dtype=torch.float64)\n",
            "translation error:  tensor(0.6592, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.862380027770996 seconds ---\n",
            "Processing file: bed_0406\n",
            "feature extraction time:  8.60434341430664\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.027853965759277344\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0038869380950927734\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9600, 1288, 5338,  ..., 6301, 2646, 1260], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05139040946960449\n",
            "Loss: 0.21407516222489775\n",
            "rotation error:  tensor(342.4181, dtype=torch.float64)\n",
            "translation error:  tensor(0.6463, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.2659170627594 seconds ---\n",
            "Processing file: bed_0407\n",
            "feature extraction time:  9.22130036354065\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025568246841430664\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030057430267333984\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9913, 2622,   13,  ...,  118, 6205, 5676], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04782438278198242\n",
            "Loss: 0.3013290709823208\n",
            "rotation error:  tensor(7.2061, dtype=torch.float64)\n",
            "translation error:  tensor(1.0252, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.79128670692444 seconds ---\n",
            "Processing file: bed_0408\n",
            "feature extraction time:  8.540326595306396\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.031316518783569336\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003429412841796875\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9942, 9559, 5764,  ..., 1427, 7749, 4428], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05261540412902832\n",
            "Loss: 0.16679816073672626\n",
            "rotation error:  tensor(93.2140, dtype=torch.float64)\n",
            "translation error:  tensor(0.5156, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.053972482681274 seconds ---\n",
            "Processing file: bed_0409\n",
            "feature extraction time:  9.331254005432129\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.026383399963378906\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0032291412353515625\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3143, 6262, 7858,  ..., 9749, 8010,  959], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05139756202697754\n",
            "Loss: 0.09172793109353375\n",
            "rotation error:  tensor(6.7608, dtype=torch.float64)\n",
            "translation error:  tensor(0.3250, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.846880435943604 seconds ---\n",
            "Epoch: [0/10], Batch: 514, Loss: 0.09172793109353375\n",
            "Processing file: bed_0410\n",
            "feature extraction time:  8.695489406585693\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03037881851196289\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.004047393798828125\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2196, 9764, 7148,  ..., 9047,  535, 8654], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05665159225463867\n",
            "Loss: 0.14942871964900956\n",
            "rotation error:  tensor(15.2286, dtype=torch.float64)\n",
            "translation error:  tensor(0.3258, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.2167067527771 seconds ---\n",
            "Processing file: bed_0411\n",
            "feature extraction time:  9.334123611450195\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025281429290771484\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003064393997192383\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   5, 1728, 6221,  ...,  554, 2870, 5537], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04853200912475586\n",
            "Loss: 0.3287330750697906\n",
            "rotation error:  tensor(175.4557, dtype=torch.float64)\n",
            "translation error:  tensor(1.0743, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.990001916885376 seconds ---\n",
            "Processing file: bed_0412\n",
            "feature extraction time:  8.717605590820312\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.027878284454345703\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030879974365234375\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7313,  754, 4567,  ..., 1971, 3377, 9670], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.06345963478088379\n",
            "Loss: 0.14985316971645019\n",
            "rotation error:  tensor(148.0001, dtype=torch.float64)\n",
            "translation error:  tensor(0.1108, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.100687980651855 seconds ---\n",
            "Processing file: bed_0413\n",
            "feature extraction time:  9.272111177444458\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0240018367767334\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028808116912841797\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7454, 5689,  126,  ..., 1810, 4781, 9595], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04931902885437012\n",
            "Loss: 0.3408909299764843\n",
            "rotation error:  tensor(236.8142, dtype=torch.float64)\n",
            "translation error:  tensor(0.5239, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.806759119033813 seconds ---\n",
            "Processing file: bed_0414\n",
            "feature extraction time:  8.744611024856567\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02990555763244629\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0034935474395751953\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5013, 9844, 5902,  ..., 5567, 3896, 4298], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05322670936584473\n",
            "Loss: 0.30678378262913775\n",
            "rotation error:  tensor(85.2392, dtype=torch.float64)\n",
            "translation error:  tensor(0.9056, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.12155055999756 seconds ---\n",
            "Epoch: [0/10], Batch: 519, Loss: 0.30678378262913775\n",
            "Processing file: bed_0415\n",
            "feature extraction time:  9.25722622871399\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024677753448486328\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029392242431640625\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 372, 4068, 7133,  ..., 6111, 5767,  959], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05515289306640625\n",
            "Loss: 0.27579301662176475\n",
            "rotation error:  tensor(244.4946, dtype=torch.float64)\n",
            "translation error:  tensor(0.4337, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.808534145355225 seconds ---\n",
            "Processing file: bed_0416\n",
            "feature extraction time:  8.827236890792847\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030798673629760742\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0033435821533203125\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6579,  497, 6938,  ..., 5549, 9034, 7876], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04893612861633301\n",
            "Loss: 0.3932246475274275\n",
            "rotation error:  tensor(41.6600, dtype=torch.float64)\n",
            "translation error:  tensor(1.2775, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.101839542388916 seconds ---\n",
            "Processing file: bed_0417\n",
            "feature extraction time:  9.194103240966797\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024950504302978516\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031473636627197266\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6310, 1871, 7364,  ..., 6835, 3327, 9518], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05436515808105469\n",
            "Loss: 0.189996921408275\n",
            "rotation error:  tensor(143.5368, dtype=torch.float64)\n",
            "translation error:  tensor(0.2342, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.6492657661438 seconds ---\n",
            "Processing file: bed_0418\n",
            "feature extraction time:  8.775728464126587\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030396461486816406\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0032880306243896484\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   4, 9994, 7366,  ..., 8626, 3261, 7290], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.051006317138671875\n",
            "Loss: 0.21319728147506622\n",
            "rotation error:  tensor(25.2715, dtype=torch.float64)\n",
            "translation error:  tensor(0.4309, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.12542414665222 seconds ---\n",
            "Processing file: bed_0419\n",
            "feature extraction time:  9.210764169692993\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024835586547851562\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002966642379760742\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5334,    1, 2409,  ..., 7414, 5309, 5057], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.054042816162109375\n",
            "Loss: 0.23001143562804527\n",
            "rotation error:  tensor(43.8726, dtype=torch.float64)\n",
            "translation error:  tensor(0.8684, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.86401104927063 seconds ---\n",
            "Epoch: [0/10], Batch: 524, Loss: 0.23001143562804527\n",
            "Processing file: bed_0420\n",
            "feature extraction time:  8.612769365310669\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.031045198440551758\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.004388093948364258\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   4, 8023, 9725,  ...,  585, 9769, 3181], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.053739070892333984\n",
            "Loss: 0.351710480227423\n",
            "rotation error:  tensor(17.1158, dtype=torch.float64)\n",
            "translation error:  tensor(1.1181, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.07191300392151 seconds ---\n",
            "Processing file: bed_0421\n",
            "feature extraction time:  9.258263111114502\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.026253461837768555\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003453493118286133\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1861, 9978, 9851,  ..., 4781, 7059, 3422], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.055974721908569336\n",
            "Loss: 0.10870101247731599\n",
            "rotation error:  tensor(96.1328, dtype=torch.float64)\n",
            "translation error:  tensor(0.2600, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.787854433059692 seconds ---\n",
            "Processing file: bed_0422\n",
            "feature extraction time:  8.508575916290283\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03312826156616211\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0020694732666015625\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1849, 6956, 5317,  ..., 8309, 9767, 1143], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05588483810424805\n",
            "Loss: 0.21906879760213605\n",
            "rotation error:  tensor(24.3268, dtype=torch.float64)\n",
            "translation error:  tensor(0.5814, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.061270713806152 seconds ---\n",
            "Processing file: bed_0423\n",
            "feature extraction time:  9.174495458602905\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03139138221740723\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003206968307495117\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([8388,  301, 8155,  ..., 7893, 2929, 2606], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05662226676940918\n",
            "Loss: 0.3036514385609329\n",
            "rotation error:  tensor(11.6602, dtype=torch.float64)\n",
            "translation error:  tensor(0.6205, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.660534143447876 seconds ---\n",
            "Processing file: bed_0424\n",
            "feature extraction time:  8.706839561462402\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.029597759246826172\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031604766845703125\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5371, 9518, 2390,  ..., 8197, 6310, 7545], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05717754364013672\n",
            "Loss: 0.3169675380261643\n",
            "rotation error:  tensor(300.8723, dtype=torch.float64)\n",
            "translation error:  tensor(0.7258, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.162518978118896 seconds ---\n",
            "Epoch: [0/10], Batch: 529, Loss: 0.3169675380261643\n",
            "Processing file: bed_0425\n",
            "feature extraction time:  9.158048629760742\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024655580520629883\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0027761459350585938\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 422, 3089, 8264,  ..., 9734, 3934, 8323], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.055643320083618164\n",
            "Loss: 0.1539039620493624\n",
            "rotation error:  tensor(26.3730, dtype=torch.float64)\n",
            "translation error:  tensor(0.4368, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.557270765304565 seconds ---\n",
            "Processing file: bed_0426\n",
            "feature extraction time:  8.60319209098816\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.029148340225219727\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030508041381835938\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9457, 9991, 9725,  ..., 7485, 2633, 6787], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05546975135803223\n",
            "Loss: 0.18708174959613955\n",
            "rotation error:  tensor(16.1842, dtype=torch.float64)\n",
            "translation error:  tensor(0.4824, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.048487186431885 seconds ---\n",
            "Processing file: bed_0427\n",
            "feature extraction time:  9.237732887268066\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024729251861572266\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031049251556396484\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9955,   38, 9801,  ..., 9131, 9777, 6422], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04476761817932129\n",
            "Loss: 0.04510850681400607\n",
            "rotation error:  tensor(6.3599, dtype=torch.float64)\n",
            "translation error:  tensor(0.1426, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.76019597053528 seconds ---\n",
            "Processing file: bed_0428\n",
            "feature extraction time:  8.593984365463257\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030104398727416992\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003255605697631836\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3997, 7543, 1089,  ..., 9000, 7344, 2906], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0530400276184082\n",
            "Loss: 0.1963990717739027\n",
            "rotation error:  tensor(214.8689, dtype=torch.float64)\n",
            "translation error:  tensor(0.1859, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.09860587120056 seconds ---\n",
            "Processing file: bed_0429\n",
            "feature extraction time:  9.303349494934082\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02464008331298828\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028390884399414062\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7483, 9889, 9981,  ..., 3043, 3110, 9897], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04989933967590332\n",
            "Loss: 0.22168930819886046\n",
            "rotation error:  tensor(13.0657, dtype=torch.float64)\n",
            "translation error:  tensor(0.5621, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.9129376411438 seconds ---\n",
            "Epoch: [0/10], Batch: 534, Loss: 0.22168930819886046\n",
            "Processing file: bed_0430\n",
            "feature extraction time:  8.750307083129883\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03223776817321777\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030748844146728516\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5546, 2379,    5,  ..., 3562, 9456, 2930], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.050127267837524414\n",
            "Loss: 0.099957415626286\n",
            "rotation error:  tensor(2.6559, dtype=torch.float64)\n",
            "translation error:  tensor(0.3472, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.17325210571289 seconds ---\n",
            "Processing file: bed_0431\n",
            "feature extraction time:  9.303924322128296\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02502727508544922\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002814054489135742\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1776, 9247,    2,  ..., 1503,  409, 8315], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04909157752990723\n",
            "Loss: 0.284732146522405\n",
            "rotation error:  tensor(214.7246, dtype=torch.float64)\n",
            "translation error:  tensor(0.8177, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.7178692817688 seconds ---\n",
            "Processing file: bed_0432\n",
            "feature extraction time:  8.815674543380737\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03226423263549805\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003382444381713867\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1264, 9001, 4521,  ..., 6767, 3782, 5810], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.050977468490600586\n",
            "Loss: 0.21994795661685762\n",
            "rotation error:  tensor(27.0625, dtype=torch.float64)\n",
            "translation error:  tensor(0.5216, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.1961510181427 seconds ---\n",
            "Processing file: bed_0433\n",
            "feature extraction time:  9.197261810302734\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02480006217956543\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030918121337890625\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2354, 5852, 9668,  ..., 2382, 3779, 7655], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.053211212158203125\n",
            "Loss: 0.23223854256251492\n",
            "rotation error:  tensor(280.6180, dtype=torch.float64)\n",
            "translation error:  tensor(0.3618, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.710359811782837 seconds ---\n",
            "Processing file: bed_0434\n",
            "feature extraction time:  8.746729612350464\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.04561734199523926\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0033321380615234375\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   2, 5543,  648,  ..., 9455, 2197, 6406], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04468274116516113\n",
            "Loss: 0.3234663010616001\n",
            "rotation error:  tensor(8.2418, dtype=torch.float64)\n",
            "translation error:  tensor(0.6067, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.927922010421753 seconds ---\n",
            "Epoch: [0/10], Batch: 539, Loss: 0.3234663010616001\n",
            "Processing file: bed_0435\n",
            "feature extraction time:  9.286977529525757\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0269010066986084\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0033004283905029297\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2233, 6502,   46,  ..., 1748, 6184, 2128], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04420661926269531\n",
            "Loss: 0.4219287281812255\n",
            "rotation error:  tensor(229.7017, dtype=torch.float64)\n",
            "translation error:  tensor(1.2215, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.748035430908203 seconds ---\n",
            "Processing file: bed_0436\n",
            "feature extraction time:  8.708804845809937\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03101038932800293\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0032434463500976562\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 280, 2019, 9957,  ..., 1215, 1940, 4123], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.049948930740356445\n",
            "Loss: 0.28178019737817267\n",
            "rotation error:  tensor(28.7381, dtype=torch.float64)\n",
            "translation error:  tensor(0.9630, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.029978275299072 seconds ---\n",
            "Processing file: bed_0437\n",
            "feature extraction time:  9.294349670410156\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.027252674102783203\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0007150173187255859\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3599, 8103, 6501,  ..., 8630, 8135, 7081], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0556950569152832\n",
            "Loss: 0.2838565089455972\n",
            "rotation error:  tensor(59.2375, dtype=torch.float64)\n",
            "translation error:  tensor(0.5372, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.812864303588867 seconds ---\n",
            "Processing file: bed_0438\n",
            "feature extraction time:  8.635202884674072\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.029347658157348633\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003459930419921875\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3458, 5507, 4416,  ..., 1262, 2930, 5654], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05231642723083496\n",
            "Loss: 0.19468756394337694\n",
            "rotation error:  tensor(19.3793, dtype=torch.float64)\n",
            "translation error:  tensor(0.7250, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.062632083892822 seconds ---\n",
            "Processing file: bed_0439\n",
            "feature extraction time:  9.22678828239441\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.027355432510375977\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.00072479248046875\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5236, 5702,  739,  ..., 3476, 8146, 1033], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.054772377014160156\n",
            "Loss: 0.12147834803859191\n",
            "rotation error:  tensor(7.9638, dtype=torch.float64)\n",
            "translation error:  tensor(0.3825, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.629794597625732 seconds ---\n",
            "Epoch: [0/10], Batch: 544, Loss: 0.12147834803859191\n",
            "Processing file: bed_0440\n",
            "feature extraction time:  8.508495807647705\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0320436954498291\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0034389495849609375\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2748, 9129, 5496,  ..., 1988, 7443, 9236], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.050424814224243164\n",
            "Loss: 0.29116486454551876\n",
            "rotation error:  tensor(23.4815, dtype=torch.float64)\n",
            "translation error:  tensor(1.0405, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.054868936538696 seconds ---\n",
            "Processing file: bed_0441\n",
            "feature extraction time:  9.204196453094482\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025867223739624023\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0032181739807128906\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6272, 3425, 7976,  ...,  157, 3545, 9536], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04722332954406738\n",
            "Loss: 0.2062355056668759\n",
            "rotation error:  tensor(39.7623, dtype=torch.float64)\n",
            "translation error:  tensor(0.5262, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.734679460525513 seconds ---\n",
            "Processing file: bed_0442\n",
            "feature extraction time:  8.581521272659302\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.034493446350097656\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002672910690307617\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9598,   19, 2900,  ..., 9840, 2860, 5291], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04879879951477051\n",
            "Loss: 0.366398289359411\n",
            "rotation error:  tensor(13.0127, dtype=torch.float64)\n",
            "translation error:  tensor(0.9873, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.108471870422363 seconds ---\n",
            "Processing file: bed_0443\n",
            "feature extraction time:  9.230367183685303\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02565932273864746\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003049135208129883\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6106, 8668,   10,  ..., 9873, 1278, 4307], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05474686622619629\n",
            "Loss: 0.4325988088527668\n",
            "rotation error:  tensor(292.5910, dtype=torch.float64)\n",
            "translation error:  tensor(0.7534, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.750091552734375 seconds ---\n",
            "Processing file: bed_0444\n",
            "feature extraction time:  8.559787034988403\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03259754180908203\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003157377243041992\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5830,  924, 2874,  ..., 7060, 1723, 5821], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05498194694519043\n",
            "Loss: 0.18845474219813735\n",
            "rotation error:  tensor(19.3811, dtype=torch.float64)\n",
            "translation error:  tensor(0.4741, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.05627417564392 seconds ---\n",
            "Epoch: [0/10], Batch: 549, Loss: 0.18845474219813735\n",
            "Processing file: bed_0445\n",
            "feature extraction time:  9.180514097213745\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025135278701782227\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0033562183380126953\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([8967, 5556, 6000,  ..., 6039, 9668, 9818], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.052198171615600586\n",
            "Loss: 0.22161146447054617\n",
            "rotation error:  tensor(10.9345, dtype=torch.float64)\n",
            "translation error:  tensor(0.6569, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.733973503112793 seconds ---\n",
            "Processing file: bed_0446\n",
            "feature extraction time:  8.526695728302002\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030660629272460938\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003093719482421875\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([  16, 7503, 9991,  ..., 3745, 5433, 3332], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05076241493225098\n",
            "Loss: 0.1685225581676269\n",
            "rotation error:  tensor(9.0256, dtype=torch.float64)\n",
            "translation error:  tensor(0.5403, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.08653211593628 seconds ---\n",
            "Processing file: bed_0447\n",
            "feature extraction time:  9.282836437225342\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02590489387512207\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0032072067260742188\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9752, 6130, 7715,  ..., 7611, 2798, 1674], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05657196044921875\n",
            "Loss: 0.23802205681636968\n",
            "rotation error:  tensor(237.7801, dtype=torch.float64)\n",
            "translation error:  tensor(0.5842, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.714677572250366 seconds ---\n",
            "Processing file: bed_0448\n",
            "feature extraction time:  8.522966861724854\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.029465198516845703\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002225160598754883\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9599, 5874, 5860,  ..., 9732, 6936, 7708], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05367612838745117\n",
            "Loss: 0.25099141193169217\n",
            "rotation error:  tensor(21.6019, dtype=torch.float64)\n",
            "translation error:  tensor(0.4718, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.912310123443604 seconds ---\n",
            "Processing file: bed_0449\n",
            "feature extraction time:  9.252224445343018\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025556087493896484\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003130197525024414\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2635, 6780, 7320,  ..., 4574, 3087, 9032], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05654430389404297\n",
            "Loss: 0.24489481955827752\n",
            "rotation error:  tensor(52.3133, dtype=torch.float64)\n",
            "translation error:  tensor(0.3800, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.771446704864502 seconds ---\n",
            "Epoch: [0/10], Batch: 554, Loss: 0.24489481955827752\n",
            "Processing file: bed_0450\n",
            "feature extraction time:  8.577991247177124\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03514218330383301\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003640890121459961\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7657, 1324, 8199,  ..., 6305, 2944, 8348], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05238652229309082\n",
            "Loss: 0.07372707103851683\n",
            "rotation error:  tensor(20.2840, dtype=torch.float64)\n",
            "translation error:  tensor(0.2294, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.097320795059204 seconds ---\n",
            "Processing file: bed_0451\n",
            "feature extraction time:  9.13821816444397\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02497124671936035\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030972957611083984\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([  85, 2459, 5076,  ..., 6296, 6649, 7082], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.054851531982421875\n",
            "Loss: 0.2350046325513434\n",
            "rotation error:  tensor(216.4634, dtype=torch.float64)\n",
            "translation error:  tensor(0.2821, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.659499168395996 seconds ---\n",
            "Processing file: bed_0452\n",
            "feature extraction time:  8.511137008666992\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.031552791595458984\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002523183822631836\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6390, 3292, 3031,  ..., 9480, 4657, 7050], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.052863359451293945\n",
            "Loss: 0.030731255134472977\n",
            "rotation error:  tensor(4.0327, dtype=torch.float64)\n",
            "translation error:  tensor(0.0892, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.046175241470337 seconds ---\n",
            "Processing file: bed_0453\n",
            "feature extraction time:  9.301650524139404\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025526046752929688\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003009319305419922\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7146, 1883, 8390,  ...,  555, 1752, 3228], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05244112014770508\n",
            "Loss: 0.1482619754552016\n",
            "rotation error:  tensor(3.6722, dtype=torch.float64)\n",
            "translation error:  tensor(0.5031, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.825647354125977 seconds ---\n",
            "Processing file: bed_0454\n",
            "feature extraction time:  8.663682460784912\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03590559959411621\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0036478042602539062\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5547,    1, 9829,  ..., 1072, 7851, 3272], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05120086669921875\n",
            "Loss: 0.16915461745219476\n",
            "rotation error:  tensor(23.9640, dtype=torch.float64)\n",
            "translation error:  tensor(0.4309, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.213682413101196 seconds ---\n",
            "Epoch: [0/10], Batch: 559, Loss: 0.16915461745219476\n",
            "Processing file: bed_0455\n",
            "feature extraction time:  9.25004768371582\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.026160001754760742\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003390073776245117\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5827, 9706, 9702,  ..., 4519, 8549, 8240], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.051470279693603516\n",
            "Loss: 0.16351930357990818\n",
            "rotation error:  tensor(17.0899, dtype=torch.float64)\n",
            "translation error:  tensor(0.5116, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.73944354057312 seconds ---\n",
            "Processing file: bed_0456\n",
            "feature extraction time:  8.5123929977417\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03384804725646973\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0036211013793945312\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7665,  155, 5410,  ..., 6376, 2482, 1685], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05286717414855957\n",
            "Loss: 0.14391154854119478\n",
            "rotation error:  tensor(11.6179, dtype=torch.float64)\n",
            "translation error:  tensor(0.3118, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.061238765716553 seconds ---\n",
            "Processing file: bed_0457\n",
            "feature extraction time:  9.30980658531189\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02493453025817871\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003111600875854492\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1153,  984, 2032,  ..., 6066, 3850, 5942], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05783677101135254\n",
            "Loss: 0.1651249415123119\n",
            "rotation error:  tensor(2.0218, dtype=torch.float64)\n",
            "translation error:  tensor(0.4125, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.825596809387207 seconds ---\n",
            "Processing file: bed_0458\n",
            "feature extraction time:  8.58668828010559\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03263998031616211\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0006079673767089844\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9815,    1, 6213,  ..., 3094, 8257, 2626], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05000782012939453\n",
            "Loss: 0.21099804090408786\n",
            "rotation error:  tensor(10.4582, dtype=torch.float64)\n",
            "translation error:  tensor(0.5246, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.947238445281982 seconds ---\n",
            "Processing file: bed_0459\n",
            "feature extraction time:  9.306628942489624\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024448394775390625\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0033462047576904297\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2312, 6806, 4658,  ..., 5295, 6159, 3258], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0530240535736084\n",
            "Loss: 0.1961409041309819\n",
            "rotation error:  tensor(12.0262, dtype=torch.float64)\n",
            "translation error:  tensor(0.4393, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.80895733833313 seconds ---\n",
            "Epoch: [0/10], Batch: 564, Loss: 0.1961409041309819\n",
            "Processing file: bed_0460\n",
            "feature extraction time:  8.61482834815979\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03111886978149414\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0007810592651367188\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2807, 6665,  607,  ..., 5718,  937, 2126], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.050878286361694336\n",
            "Loss: 0.22971897965380117\n",
            "rotation error:  tensor(17.0656, dtype=torch.float64)\n",
            "translation error:  tensor(0.7360, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.13001799583435 seconds ---\n",
            "Processing file: bed_0461\n",
            "feature extraction time:  9.317954540252686\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025954723358154297\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0007512569427490234\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6263, 1052, 2852,  ..., 1578, 6842, 2999], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05295610427856445\n",
            "Loss: 0.17637230539326132\n",
            "rotation error:  tensor(14.0250, dtype=torch.float64)\n",
            "translation error:  tensor(0.3280, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.77160143852234 seconds ---\n",
            "Processing file: bed_0462\n",
            "feature extraction time:  8.547009229660034\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.032753705978393555\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0006539821624755859\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9440, 1057, 7593,  ..., 9759, 6040, 7555], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.049147844314575195\n",
            "Loss: 0.2166558980435221\n",
            "rotation error:  tensor(254.4157, dtype=torch.float64)\n",
            "translation error:  tensor(0.4299, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.986868143081665 seconds ---\n",
            "Processing file: bed_0463\n",
            "feature extraction time:  9.16624927520752\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024749040603637695\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003528594970703125\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   1, 7428,  828,  ..., 8168, 6128,  190], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04567384719848633\n",
            "Loss: 0.36286524781985197\n",
            "rotation error:  tensor(169.3098, dtype=torch.float64)\n",
            "translation error:  tensor(1.0717, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.695181369781494 seconds ---\n",
            "Processing file: bed_0464\n",
            "feature extraction time:  8.673140048980713\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03635287284851074\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003715991973876953\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 314, 7002, 9957,  ..., 9364, 7174, 5630], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05432462692260742\n",
            "Loss: 0.4064411451552594\n",
            "rotation error:  tensor(280.4117, dtype=torch.float64)\n",
            "translation error:  tensor(1.0348, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.173238277435303 seconds ---\n",
            "Epoch: [0/10], Batch: 569, Loss: 0.4064411451552594\n",
            "Processing file: bed_0465\n",
            "feature extraction time:  9.162321090698242\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02611517906188965\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003267526626586914\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5535, 3030, 9547,  ..., 7368, 7299, 7371], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05095553398132324\n",
            "Loss: 0.2047985426792188\n",
            "rotation error:  tensor(102.4154, dtype=torch.float64)\n",
            "translation error:  tensor(0.3213, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.650460243225098 seconds ---\n",
            "Processing file: bed_0466\n",
            "feature extraction time:  8.509240627288818\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.027168989181518555\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031557083129882812\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3274,   40,  943,  ..., 4945, 7483, 1538], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0586237907409668\n",
            "Loss: 0.2567022092581724\n",
            "rotation error:  tensor(312.3728, dtype=torch.float64)\n",
            "translation error:  tensor(0.4989, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.122822046279907 seconds ---\n",
            "Processing file: bed_0467\n",
            "feature extraction time:  9.251980781555176\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02546405792236328\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029544830322265625\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 550, 3335, 2634,  ..., 3624, 8761, 1364], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05151844024658203\n",
            "Loss: 0.09705434405051336\n",
            "rotation error:  tensor(10.0255, dtype=torch.float64)\n",
            "translation error:  tensor(0.2693, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.74457597732544 seconds ---\n",
            "Processing file: bed_0468\n",
            "feature extraction time:  8.569550037384033\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030738353729248047\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002026796340942383\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9654, 8010, 9972,  ..., 8322,  908, 3945], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05345559120178223\n",
            "Loss: 0.2346611494094452\n",
            "rotation error:  tensor(25.5302, dtype=torch.float64)\n",
            "translation error:  tensor(0.8387, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.04274034500122 seconds ---\n",
            "Processing file: bed_0469\n",
            "feature extraction time:  9.179968118667603\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024753332138061523\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030069351196289062\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   1, 9179, 4528,  ..., 3818, 9727, 5085], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04862332344055176\n",
            "Loss: 0.3828852518274716\n",
            "rotation error:  tensor(29.5388, dtype=torch.float64)\n",
            "translation error:  tensor(1.0411, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.632972717285156 seconds ---\n",
            "Epoch: [0/10], Batch: 574, Loss: 0.3828852518274716\n",
            "Processing file: bed_0470\n",
            "feature extraction time:  8.467237710952759\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.031316518783569336\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.004010438919067383\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1023, 3727, 5013,  ..., 7431,  391,  994], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0534517765045166\n",
            "Loss: 0.21770277183879214\n",
            "rotation error:  tensor(93.3442, dtype=torch.float64)\n",
            "translation error:  tensor(0.3659, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.941648960113525 seconds ---\n",
            "Processing file: bed_0471\n",
            "feature extraction time:  9.324327230453491\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02612924575805664\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0033164024353027344\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9916,  599, 8021,  ..., 5076, 5326, 6122], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.059227705001831055\n",
            "Loss: 0.1924772070256904\n",
            "rotation error:  tensor(12.7902, dtype=torch.float64)\n",
            "translation error:  tensor(0.5501, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.837223529815674 seconds ---\n",
            "Processing file: bed_0472\n",
            "feature extraction time:  8.636077880859375\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.033989667892456055\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003206968307495117\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4680, 5735, 7268,  ..., 8306, 6465, 2381], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05469346046447754\n",
            "Loss: 0.5224746744718942\n",
            "rotation error:  tensor(190.3385, dtype=torch.float64)\n",
            "translation error:  tensor(1.5951, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.178632974624634 seconds ---\n",
            "Processing file: bed_0473\n",
            "feature extraction time:  9.251145362854004\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0249021053314209\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030994415283203125\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4594, 9702, 8629,  ..., 7960, 4088, 3436], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04429030418395996\n",
            "Loss: 0.33218788154992646\n",
            "rotation error:  tensor(9.6011, dtype=torch.float64)\n",
            "translation error:  tensor(0.8393, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.77361273765564 seconds ---\n",
            "Processing file: bed_0474\n",
            "feature extraction time:  8.514983177185059\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03039073944091797\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.00347137451171875\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9362, 5610,  511,  ..., 3195, 6568, 7008], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05514860153198242\n",
            "Loss: 0.26257464657179774\n",
            "rotation error:  tensor(10.1430, dtype=torch.float64)\n",
            "translation error:  tensor(0.9719, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.024184942245483 seconds ---\n",
            "Epoch: [0/10], Batch: 579, Loss: 0.26257464657179774\n",
            "Processing file: bed_0475\n",
            "feature extraction time:  9.208892345428467\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02547764778137207\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030088424682617188\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 977, 2886, 7471,  ..., 1502, 1554, 1396], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04976034164428711\n",
            "Loss: 0.13812205952245557\n",
            "rotation error:  tensor(6.9833, dtype=torch.float64)\n",
            "translation error:  tensor(0.2630, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.649874448776245 seconds ---\n",
            "Processing file: bed_0476\n",
            "feature extraction time:  8.583239316940308\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03349900245666504\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0034613609313964844\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6733, 1296, 8296,  ..., 3046, 9628,  252], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05555987358093262\n",
            "Loss: 0.2573174410207524\n",
            "rotation error:  tensor(14.0426, dtype=torch.float64)\n",
            "translation error:  tensor(0.7996, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.060980558395386 seconds ---\n",
            "Processing file: bed_0477\n",
            "feature extraction time:  9.163024425506592\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025348901748657227\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003581523895263672\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9344, 9938,   21,  ..., 3385, 1492, 2392], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0462644100189209\n",
            "Loss: 0.30763486364656284\n",
            "rotation error:  tensor(9.5146, dtype=torch.float64)\n",
            "translation error:  tensor(0.9489, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.74050259590149 seconds ---\n",
            "Processing file: bed_0478\n",
            "feature extraction time:  8.459943056106567\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.030614376068115234\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0035839080810546875\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9804, 9744,    2,  ...,  998, 5549, 9369], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05797386169433594\n",
            "Loss: 0.23481640300876855\n",
            "rotation error:  tensor(313.8445, dtype=torch.float64)\n",
            "translation error:  tensor(0.3792, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.817234754562378 seconds ---\n",
            "Processing file: bed_0479\n",
            "feature extraction time:  9.30630350112915\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025034427642822266\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031862258911132812\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2299, 3310, 8240,  ..., 5541, 9416, 9172], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05369281768798828\n",
            "Loss: 0.20242272409679898\n",
            "rotation error:  tensor(37.0372, dtype=torch.float64)\n",
            "translation error:  tensor(0.4597, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.873450756072998 seconds ---\n",
            "Epoch: [0/10], Batch: 584, Loss: 0.20242272409679898\n",
            "Processing file: bed_0480\n",
            "feature extraction time:  8.494139194488525\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03333783149719238\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003675222396850586\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([8112, 6165,    1,  ..., 1403, 6562, 8797], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.047719717025756836\n",
            "Loss: 0.21695296838500397\n",
            "rotation error:  tensor(19.5202, dtype=torch.float64)\n",
            "translation error:  tensor(0.7959, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.03337025642395 seconds ---\n",
            "Processing file: bed_0481\n",
            "feature extraction time:  9.145540714263916\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024790525436401367\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003134489059448242\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 565, 6134, 4565,  ..., 9586, 8060, 8062], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05403017997741699\n",
            "Loss: 0.3200789351159211\n",
            "rotation error:  tensor(12.7417, dtype=torch.float64)\n",
            "translation error:  tensor(0.8756, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.28902792930603 seconds ---\n",
            "Processing file: bed_0482\n",
            "feature extraction time:  8.802122831344604\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03211665153503418\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0033452510833740234\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   6, 7056, 3256,  ..., 4506, 6919, 4223], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.049947261810302734\n",
            "Loss: 0.38885718639214106\n",
            "rotation error:  tensor(271.9432, dtype=torch.float64)\n",
            "translation error:  tensor(1.0845, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.66335964202881 seconds ---\n",
            "Processing file: bed_0483\n",
            "feature extraction time:  8.94962453842163\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02552056312561035\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0032286643981933594\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([5717, 2274, 2137,  ..., 4447, 1991, 9713], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04694008827209473\n",
            "Loss: 0.06919682876884282\n",
            "rotation error:  tensor(16.8881, dtype=torch.float64)\n",
            "translation error:  tensor(0.1335, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.175212144851685 seconds ---\n",
            "Processing file: bed_0484\n",
            "feature extraction time:  9.034213781356812\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024602651596069336\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031380653381347656\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2719, 8342, 4336,  ..., 5089, 6469, 5815], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05158805847167969\n",
            "Loss: 0.18250351483490002\n",
            "rotation error:  tensor(9.4730, dtype=torch.float64)\n",
            "translation error:  tensor(0.6155, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.624112606048584 seconds ---\n",
            "Epoch: [0/10], Batch: 589, Loss: 0.18250351483490002\n",
            "Processing file: bed_0485\n",
            "feature extraction time:  8.948715209960938\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024893999099731445\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029497146606445312\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1791, 4055, 9708,  ..., 7435, 4962, 4689], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05584144592285156\n",
            "Loss: 0.32059756615696794\n",
            "rotation error:  tensor(179.1745, dtype=torch.float64)\n",
            "translation error:  tensor(0.9192, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.22061848640442 seconds ---\n",
            "Processing file: bed_0486\n",
            "feature extraction time:  8.940824270248413\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025965452194213867\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029239654541015625\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2490, 9410, 8120,  ..., 7635, 6480, 8277], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05342364311218262\n",
            "Loss: 0.21803714383078213\n",
            "rotation error:  tensor(154.8505, dtype=torch.float64)\n",
            "translation error:  tensor(0.3756, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.700376749038696 seconds ---\n",
            "Processing file: bed_0487\n",
            "feature extraction time:  8.826992273330688\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025738000869750977\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030286312103271484\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1173, 4627, 3933,  ..., 4659, 1563, 5157], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.049288034439086914\n",
            "Loss: 0.07345146010320155\n",
            "rotation error:  tensor(14.2849, dtype=torch.float64)\n",
            "translation error:  tensor(0.1897, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.071472883224487 seconds ---\n",
            "Processing file: bed_0488\n",
            "feature extraction time:  8.828504085540771\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024990558624267578\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029876232147216797\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([6963, 3170, 9916,  ..., 1835, 3744, 8727], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05168008804321289\n",
            "Loss: 0.2262276674181175\n",
            "rotation error:  tensor(7.4276, dtype=torch.float64)\n",
            "translation error:  tensor(0.4479, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.99303960800171 seconds ---\n",
            "Processing file: bed_0489\n",
            "feature extraction time:  8.482332229614258\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02469921112060547\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003085613250732422\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([8213, 5544, 6659,  ..., 6878, 5794,  466], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04583001136779785\n",
            "Loss: 0.43497937081732185\n",
            "rotation error:  tensor(83.7535, dtype=torch.float64)\n",
            "translation error:  tensor(1.2944, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.682186126708984 seconds ---\n",
            "Epoch: [0/10], Batch: 594, Loss: 0.43497937081732185\n",
            "Processing file: bed_0490\n",
            "feature extraction time:  9.132801294326782\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02452564239501953\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002787351608276367\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([2285,   99, 8216,  ..., 2253, 4248, 6332], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04378390312194824\n",
            "Loss: 0.24886366552359182\n",
            "rotation error:  tensor(7.9731, dtype=torch.float64)\n",
            "translation error:  tensor(0.5064, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.417692184448242 seconds ---\n",
            "Processing file: bed_0491\n",
            "feature extraction time:  8.566865921020508\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02390575408935547\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002841472625732422\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9788, 7061,   68,  ..., 5750, 9441, 4157], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.052788496017456055\n",
            "Loss: 0.17338320921741662\n",
            "rotation error:  tensor(8.2619, dtype=torch.float64)\n",
            "translation error:  tensor(0.3633, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.761595249176025 seconds ---\n",
            "Processing file: bed_0492\n",
            "feature extraction time:  8.935899019241333\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0262601375579834\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0030808448791503906\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([7910, 9870, 2446,  ..., 4582, 1905, 3172], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.047842979431152344\n",
            "Loss: 0.3224932037739869\n",
            "rotation error:  tensor(131.5622, dtype=torch.float64)\n",
            "translation error:  tensor(0.8766, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.128939867019653 seconds ---\n",
            "Processing file: bed_0493\n",
            "feature extraction time:  8.72005558013916\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.03066086769104004\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003404855728149414\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 686, 4051, 2221,  ..., 7297, 4579, 5747], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.047628164291381836\n",
            "Loss: 0.38216087369987345\n",
            "rotation error:  tensor(237.2321, dtype=torch.float64)\n",
            "translation error:  tensor(0.7745, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.600632190704346 seconds ---\n",
            "Processing file: bed_0494\n",
            "feature extraction time:  8.86852216720581\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025847911834716797\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003319263458251953\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9656,   16, 4691,  ..., 5741, 3785,  865], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0496668815612793\n",
            "Loss: 0.23778742335916345\n",
            "rotation error:  tensor(244.7595, dtype=torch.float64)\n",
            "translation error:  tensor(0.8189, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.16116762161255 seconds ---\n",
            "Epoch: [0/10], Batch: 599, Loss: 0.23778742335916345\n",
            "Processing file: bed_0495\n",
            "feature extraction time:  9.096367120742798\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02443528175354004\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003432750701904297\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9579, 2620, 6701,  ..., 8132, 8456, 1651], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0514981746673584\n",
            "Loss: 0.1877699323858824\n",
            "rotation error:  tensor(5.3211, dtype=torch.float64)\n",
            "translation error:  tensor(0.6371, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.891008853912354 seconds ---\n",
            "Processing file: bed_0496\n",
            "feature extraction time:  9.048887252807617\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02478623390197754\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002992868423461914\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 257, 7644, 5474,  ..., 4605, 9714, 7719], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04897618293762207\n",
            "Loss: 0.17773182212430183\n",
            "rotation error:  tensor(3.1685, dtype=torch.float64)\n",
            "translation error:  tensor(0.3306, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.27933645248413 seconds ---\n",
            "Processing file: bed_0497\n",
            "feature extraction time:  8.893037557601929\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02429485321044922\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029022693634033203\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([  22, 2809, 9854,  ..., 1413, 2850, 5671], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04839015007019043\n",
            "Loss: 0.2936381828236819\n",
            "rotation error:  tensor(123.4950, dtype=torch.float64)\n",
            "translation error:  tensor(1.0400, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.744851112365723 seconds ---\n",
            "Processing file: bed_0498\n",
            "feature extraction time:  8.784273386001587\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025233030319213867\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003006458282470703\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3312,    4, 7375,  ..., 7045, 4542, 8407], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.058855295181274414\n",
            "Loss: 0.22346605020807153\n",
            "rotation error:  tensor(45.4036, dtype=torch.float64)\n",
            "translation error:  tensor(0.3531, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.91810894012451 seconds ---\n",
            "Processing file: bed_0499\n",
            "feature extraction time:  8.825339555740356\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.0249173641204834\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0029914379119873047\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([ 931, 1235, 4983,  ..., 5679, 6984, 3799], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05132555961608887\n",
            "Loss: 0.1920137494060968\n",
            "rotation error:  tensor(14.9664, dtype=torch.float64)\n",
            "translation error:  tensor(0.6864, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.19256067276001 seconds ---\n",
            "Epoch: [0/10], Batch: 604, Loss: 0.1920137494060968\n",
            "Processing file: bed_0500\n",
            "feature extraction time:  8.492865562438965\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02470541000366211\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.002984762191772461\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([8320, 6540, 4592,  ..., 3873, 2359, 7048], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05186152458190918\n",
            "Loss: 0.06052596596049718\n",
            "rotation error:  tensor(13.6700, dtype=torch.float64)\n",
            "translation error:  tensor(0.1644, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.63666319847107 seconds ---\n",
            "Processing file: bed_0501\n",
            "feature extraction time:  8.957029581069946\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024353981018066406\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031588077545166016\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([8164,    3, 9943,  ..., 2082, 4576, 8175], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05963754653930664\n",
            "Loss: 0.21290982755486446\n",
            "rotation error:  tensor(10.0007, dtype=torch.float64)\n",
            "translation error:  tensor(0.4451, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.190208911895752 seconds ---\n",
            "Processing file: bed_0502\n",
            "feature extraction time:  8.462611198425293\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025220632553100586\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0028641223907470703\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([3596, 5349, 5527,  ..., 8989, 6037, 2811], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05279064178466797\n",
            "Loss: 0.21310239721370136\n",
            "rotation error:  tensor(31.0177, dtype=torch.float64)\n",
            "translation error:  tensor(0.7033, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.710686683654785 seconds ---\n",
            "Processing file: bed_0503\n",
            "feature extraction time:  8.816659688949585\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.025200605392456055\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.003173828125\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([4032, 3153,   12,  ..., 5558, 9242, 5696], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.054009437561035156\n",
            "Loss: 0.2923965837894512\n",
            "rotation error:  tensor(241.6813, dtype=torch.float64)\n",
            "translation error:  tensor(0.5621, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.152148962020874 seconds ---\n",
            "Processing file: bed_0504\n",
            "feature extraction time:  8.651070356369019\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.034146785736083984\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0038623809814453125\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([1540, 5800, 5862,  ..., 7956, 7050, 1098], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.05325484275817871\n",
            "Loss: 0.1971565846999236\n",
            "rotation error:  tensor(338.8939, dtype=torch.float64)\n",
            "translation error:  tensor(0.7053, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.61901330947876 seconds ---\n",
            "Epoch: [0/10], Batch: 609, Loss: 0.1971565846999236\n",
            "Processing file: bed_0505\n",
            "feature extraction time:  8.938846826553345\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.024950265884399414\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0027265548706054688\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([9965, 9851, 3185,  ..., 4177, 6713, 2158], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.04904007911682129\n",
            "Loss: 0.2467364660364256\n",
            "rotation error:  tensor(97.8590, dtype=torch.float64)\n",
            "translation error:  tensor(0.7319, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 18.149493932724 seconds ---\n",
            "Processing file: bed_0506\n",
            "feature extraction time:  9.017613649368286\n",
            "src_keypts_idx_unsqueezed:  torch.Size([1, 6, 64])\n",
            "src_keypts:  torch.Size([1, 64, 6])\n",
            "Grouping keypoints time:  0.02470707893371582\n",
            "B:  1\n",
            "K_topk:  64\n",
            "nsample:  32\n",
            "num_feat:  32\n",
            "get_cat_feat_src time:  0.0031614303588867188\n",
            "tgt_pts_xyz:  torch.Size([1, 10000, 3])\n",
            "ref_pts:  torch.Size([1, 10000, 3])\n",
            "dist_normalize:  torch.Size([1, 13824, 32])\n",
            "feat_weight_map:  torch.Size([1, 13824, 32, 32])\n",
            "idx_1_mask:  tensor([[0]])\n",
            "idx_1_mask_flatten:  tensor([0])\n",
            "idx_2_mask:  tensor([   2, 8880, 6171,  ..., 9386, 9843, 7146], device='cuda:0')\n",
            "get_cat_feat_tgt time:  0.0525357723236084\n",
            "Loss: 0.2114144300219636\n",
            "rotation error:  tensor(111.6137, dtype=torch.float64)\n",
            "translation error:  tensor(0.4831, device='cuda:0', dtype=torch.float64, grad_fn=<MeanBackward0>)\n",
            "--- 17.812089204788208 seconds ---\n",
            "Processing file: bed_0507\n"
          ]
        }
      ],
      "source": [
        "optim = Adam(model.parameters(), lr=lr)\n",
        "\n",
        "# begin train \n",
        "model.train()\n",
        "loss_epoch_avg = []\n",
        "for epoch in range(num_epochs):\n",
        "    print(f\"epoch #{epoch}\")\n",
        "    loss_epoch = []\n",
        "    running_loss = 0.0\n",
        "\n",
        "    for n_batch, (src, target, R_gt, t_gt, ) in enumerate(train_loader):\n",
        "        start_time = time.time()\n",
        "        # mini batch\n",
        "        src, target, R_gt, t_gt = src.to(device), target.to(device), R_gt.to(device), t_gt.to(device)\n",
        "        t_init = torch.randn((1, 3))\n",
        "        R_prior = R_gt.clone()\n",
        "        corruption = torch.FloatTensor(batch_size, 3, 3).uniform_(0.9, 1.1).cuda()\n",
        "        R_prior = R_prior*corruption\n",
        "        src_keypts, target_vcp = model(src, target, R_prior, t_init)\n",
        "        # print('src_keypts shape', src_keypts.shape)\n",
        "        # print('target_vcp shape', target_vcp.shape)\n",
        "        # zero gradient \n",
        "        optim.zero_grad()\n",
        "        loss, R_pred, t_pred = deepVCP_loss(src_keypts, target_vcp, R_gt, t_gt, alpha=0.5)\n",
        "\n",
        "\n",
        "        # error metric for rigid body transformation\n",
        "        r_pred = R.from_matrix(R_pred.squeeze(0).cpu().detach().numpy())\n",
        "        r_pred_arr = torch.tensor(r_pred.as_euler('xyz', degrees=True)).reshape(1, 3)\n",
        "        r_gt = R.from_matrix(R_gt.squeeze(0).cpu().detach().numpy())\n",
        "        r_gt_arr = torch.tensor(r_gt.as_euler('xyz', degrees=True)).reshape(1, 3)\n",
        "        pdist = nn.PairwiseDistance(p = 2)\n",
        "        t_pred = t_pred.squeeze(-1)\n",
        "        t_gt = t_gt.squeeze(-1)\n",
        "\n",
        "        print(\"rotation error: \", pdist(r_pred_arr, r_gt_arr).mean())\n",
        "        print(\"translation error: \", pdist(t_pred, t_gt).mean())\n",
        "\n",
        "        # backward pass\n",
        "        loss.backward()\n",
        "        # update parameters \n",
        "        optim.step()\n",
        "\n",
        "        running_loss += loss.item()\n",
        "        loss_epoch += [loss.item()]\n",
        "        print(\"--- %s seconds ---\" % (time.time() - start_time))\n",
        "        if (n_batch + 1) % 5 == 0:\n",
        "            print(\"Epoch: [{}/{}], Batch: {}, Loss: {}\".format(\n",
        "                epoch, num_epochs, n_batch, loss.item()))\n",
        "            running_loss = 0.0\n",
        "\n",
        "    torch.save(model.state_dict(), \"epoch_\" + str(epoch) + \"_model.pt\")\n",
        "    loss_epoch_avg += [sum(loss_epoch) / len(loss_epoch)]\n",
        "    print(\"loss epoch\", loss_epoch)\n",
        "    with open(\"training_loss_\" + str(epoch) + \".txt\", \"wb\") as fp:   #Pickling\n",
        "        pickle.dump(loss_epoch, fp)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0Cym5SGxlJ0K",
        "outputId": "4a6965ca-21ac-48af-a509-6692870d8468"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "tensor([[[0.9160, 1.1861, 0.9869],\n",
              "         [0.8915, 0.9294, 1.0784],\n",
              "         [1.1967, 0.8959, 0.9514]],\n",
              "\n",
              "        [[0.8575, 0.8269, 0.8158],\n",
              "         [1.0765, 0.9359, 0.8825],\n",
              "         [1.0584, 0.8722, 0.8100]],\n",
              "\n",
              "        [[1.0827, 1.0449, 1.1016],\n",
              "         [0.8251, 0.9746, 1.1939],\n",
              "         [1.0324, 0.8375, 1.0622]]])"
            ]
          },
          "execution_count": 12,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "torch.FloatTensor(3, 3, 3).uniform_(0.8, 1.2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2SL28L_2da7f"
      },
      "outputs": [],
      "source": [
        "# save\n",
        "print(\"Finished Training\")\n",
        "torch.save(model.state_dict(), model_path)"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3.10.10 ('InverseCW4')",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.10.10"
    },
    "vscode": {
      "interpreter": {
        "hash": "709ee83345f2a73a4810e0cadc6e4be23f2f850d1ef51d9b25c99df9df8dd8b0"
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}