feat: update naic20 1-st solution

README.md

@@ -4,6 +4,7 @@ FastReID is a research platform that implements state-of-the-art re-identificati
 
 ## What's New
 
+- [Jan 2021] NAIC20(reid track) [1-st solution](https://github.com/JDAI-CV/fast-reid/tree/master/projects/NAIC20) based on fastreid has been released！
 - [Jan 2021] FastReID V1.0 has been released！🎉 
 Support many tasks beyond reid, such image retrieval and face recognition. See [release notes](https://github.com/JDAI-CV/fast-reid/releases/tag/v1.0.0).
 - [Oct 2020] Added the [Hyper-Parameter Optimization](https://github.com/JDAI-CV/fast-reid/tree/master/projects/FastTune) based on fastreid. See `projects/FastTune`.

projects/NAIC20/README.md

@@ -0,0 +1,64 @@
+# NAIC20 Competition (ReID Track) 
+
+This repository contains the 1-st place solution of ReID Competition of NAIC. We got the first place in the final stage. 
+
+## Introduction
+
+Detailed information about the NAIC competition can be found [here](https://naic.pcl.ac.cn/homepage/index.html).
+
+## Useful Tricks
+
+- [x] DataAugmentation (RandomErasing + ColorJitter + Augmix + RandomAffine + RandomHorizontallyFilp + Padding + RandomCrop)
+- [x] LR Scheduler (Warmup + CosineAnnealing)
+- [x] Optimizer (Adam)
+- [x] FP16 mixed precision training
+- [x] CircleSoftmax
+- [x] Pairwise Cosface
+- [x] GeM pooling
+- [x] Remove Long Tail Data (pid with single image)
+- [x] Channel Shuffle
+- [x] Distmat Ensemble
+
+1. Due to the competition's rule, pseudo label is not allowed in the preliminary and semi-finals, but can be used in finals.
+2. We combine naic19, naic20r1 and naic20r2 datasets, but there are overlap and noise between these datasets. So we
+use an automatic data clean strategy for data clean. The cleaned txt files are put here. Sorry that this part cannot ben open sourced.
+3. Due to the characteristics of the encrypted dataset, we found **channel shuffle** very helpful.
+It's an offline data augmentation method. Specifically, for each id, random choice an order of channel, 
+such as `(2, 1, 0)`, then apply this order for all images of this id, and make it a new id.
+With this method, you can enlarge the scale of identities. Theoretically, each id can be enlarged to 5 times.
+Considering computational efficiency and marginal effect, we just enlarge each id once.
+But this trick is no effect in normal dataset.
+4. Due to the distribution of dataset, we found pairwise cosface can greatly boost model performance.
+5. The performance of `resnest` is far better than `ibn`. 
+We choose `resnest101`, `resnest200` with different resolution (192x256, 192x384) to ensemble. 
+
+## Training & Submission in Command Line
+
+Before starting, please see [GETTING_STARTED.md](https://github.com/JDAI-CV/fast-reid/blob/master/GETTING_STARTED.md) for the basic setup of FastReID.
+All configs are made for 2-GPU training.
+
+1. To train a model, first set up the corresponding datasets following [datasets/README.md](https://github.com/JDAI-CV/fast-reid/tree/master/datasets), then run:
+
+```bash
+python3 projects/NAIC20/train_net.py --config-file projects/NAIC20/configs/r34-ibn.yml --num-gpus 2 
+```
+
+2. After the model is trained, you can start to generate submission file. First, modify the content of `MODEL` in `submit.yml` to 
+adapt your trained model, and set `MODEL.WEIGHTS` to the path of your trained model, then run:
+
+```bash
+python3 projects/NAIC20/train_net.py --config-file projects/NAIC20/configs/submit.yml --eval-only --commit --num-gpus 2
+```
+
+You can find `submit.json` and `distmat.npy` in `OUTPUT_DIR` of `submit.yml`.
+
+## Ablation Study
+
+To quickly verify the results, we use resnet34-ibn as backbone to conduct ablation study.
+The datasets are `naic19`, `naic20r1` and `naic20r2`.
+
+| Setting | Rank-1 | mAP |
+| ------  | ------ | --- |
+| Baseline | 70.11 | 63.29 |
+| w/ tripletx10 | 73.79 | 67.01 |
+| w/ cosface | 75.61 | 70.07 |

projects/NAIC20/configs/Base-naic.yml

@@ -0,0 +1,96 @@
+MODEL:
+  META_ARCHITECTURE: Baseline
+
+  FREEZE_LAYERS: [ backbone ]
+
+  HEADS:
+    NAME: EmbeddingHead
+    NORM: BN
+    EMBEDDING_DIM: 0
+    NECK_FEAT: after
+    POOL_LAYER: gempool
+    CLS_LAYER: circleSoftmax
+    SCALE: 64
+    MARGIN: 0.35
+
+  LOSSES:
+    NAME: ("CrossEntropyLoss", "Cosface",)
+
+    CE:
+      EPSILON: 0.
+      SCALE: 1.
+
+    TRI:
+      MARGIN: 0.
+      HARD_MINING: True
+      NORM_FEAT: True
+      SCALE: 1.
+
+    COSFACE:
+      MARGIN: 0.35
+      GAMMA: 64
+      SCALE: 1.
+
+INPUT:
+  SIZE_TRAIN: [ 256, 128 ]
+  SIZE_TEST: [ 256, 128 ]
+
+  DO_AUGMIX: True
+  AUGMIX_PROB: 0.5
+
+  DO_AFFINE: True
+
+  REA:
+    ENABLED: True
+    VALUE: [ 0., 0., 0. ]
+
+  CJ:
+    ENABLED: True
+    BRIGHTNESS: 0.15
+    CONTRAST: 0.1
+    SATURATION: 0.
+    HUE: 0.
+
+DATALOADER:
+  PK_SAMPLER: True
+  NAIVE_WAY: True
+  NUM_INSTANCE: 2
+  NUM_WORKERS: 8
+
+SOLVER:
+  FP16_ENABLED: False
+  OPT: Adam
+  SCHED: CosineAnnealingLR
+  MAX_EPOCH: 30
+  BASE_LR: 0.0007
+  BIAS_LR_FACTOR: 1.
+  WEIGHT_DECAY: 0.0005
+  WEIGHT_DECAY_BIAS: 0.0005
+  IMS_PER_BATCH: 256
+
+  DELAY_EPOCHS: 5
+  ETA_MIN_LR: 0.0000007
+
+  FREEZE_ITERS: 1000
+  FREEZE_FC_ITERS: 0
+
+  WARMUP_FACTOR: 0.1
+  WARMUP_ITERS: 4000
+
+  CHECKPOINT_PERIOD: 3
+
+DATASETS:
+  NAMES: ("NAIC20_R2", "NAIC20_R1", "NAIC19",)
+  TESTS: ("NAIC20_R2",)
+  RM_LT: True
+
+TEST:
+  EVAL_PERIOD: 3
+  IMS_PER_BATCH: 256
+  RERANK:
+    ENABLED: False
+    K1: 20
+    K2: 3
+    LAMBDA: 0.5
+
+CUDNN_BENCHMARK: True

projects/NAIC20/configs/nest101-base.yml

@@ -0,0 +1,10 @@
+_BASE_: Base-naic.yml
+
+MODEL:
+  BACKBONE:
+    NAME: build_resnest_backbone
+    DEPTH: 101x
+    WITH_IBN: False
+    PRETRAIN: True
+
+OUTPUT_DIR: projects/NAIC20/logs/nest101-128x256

projects/NAIC20/configs/r34-ibn.yml

@@ -0,0 +1,11 @@
+_BASE_: Base-naic.yml
+
+MODEL:
+  BACKBONE:
+    NAME: build_resnet_backbone
+    DEPTH: 34x
+    FEAT_DIM: 512
+    WITH_IBN: True
+    PRETRAIN: True
+
+OUTPUT_DIR: projects/NAIC20/logs/r34_ibn-128x256

projects/NAIC20/configs/submit.yml

@@ -0,0 +1,24 @@
+_BASE_: Base-naic.yml
+
+MODEL:
+  BACKBONE:
+    NAME: build_resnet_backbone
+    DEPTH: 34x
+    FEAT_DIM: 512
+    WITH_IBN: True
+
+  WEIGHTS: projects/NAIC20/logs/reproduce/r34-tripletx10/model_best.pth
+
+DATASETS:
+  TESTS: ("NAIC20_R2A",)
+
+TEST:
+  RERANK:
+    ENABLED: True
+    K1: 20
+    K2: 3
+    LAMBDA: 0.8
+
+  SAVE_DISTMAT: True
+
+OUTPUT_DIR: projects/NAIC20/logs/r34_ibn-128x256-submit