GCN_stockMarket.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "view-in-github"
   },
   "source": [
    "<a href=\"https://colab.research.google.com/github/Le-Triumvirat/CADS-Internship/blob/main/Anomaly_detection_in_financial_services_using_knowledge_graphs_and_machine_learning.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "aDtk1HTUXK5C"
   },
   "source": [
    "<h1>Installing DGL and DGL-KE</h1>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "zoqMIp_noYPx",
    "outputId": "759c67b6-4740-4524-c336-53d96f6707eb"
   },
   "source": [
    "> pip install dgl\n",
    "\n",
    "> pip install dglke"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "lF-HrZy0VsMi"
   },
   "source": [
    "# Main\n",
    ">    Our cave to brainstorm and invent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "lXFox_QxVroH"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using backend: pytorch\n"
     ]
    }
   ],
   "source": [
    "import dgl\n",
    "import numpy as np\n",
    "import networkx as nx\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# unique nodes are 34 and edge connections are 156 (might include either directed or undirected)\n",
    "def build_karate_club_graph():\n",
    "    # source node\n",
    "    src = np.array([1, 2, 2, 3, 3, 3, 4, 5, 6, 6, 6, 7, 7, 7, 7, 8, 8, 9, 10, 10,\n",
    "        10, 11, 12, 12, 13, 13, 13, 13, 16, 16, 17, 17, 19, 19, 21, 21,\n",
    "        25, 25, 27, 27, 27, 28, 29, 29, 30, 30, 31, 31, 31, 31, 32, 32,\n",
    "        32, 32, 32, 32, 32, 32, 32, 32, 32, 33, 33, 33, 33, 33, 33, 33,\n",
    "        33, 33, 33, 33, 33, 33, 33, 33, 33, 33])\n",
    "    # destination node\n",
    "    dst = np.array([0, 0, 1, 0, 1, 2, 0, 0, 0, 4, 5, 0, 1, 2, 3, 0, 2, 2, 0, 4,\n",
    "        5, 0, 0, 3, 0, 1, 2, 3, 5, 6, 0, 1, 0, 1, 0, 1, 23, 24, 2, 23,\n",
    "        24, 2, 23, 26, 1, 8, 0, 24, 25, 28, 2, 8, 14, 15, 18, 20, 22, 23,\n",
    "        29, 30, 31, 8, 9, 13, 14, 15, 18, 19, 20, 22, 23, 26, 27, 28, 29, 30,\n",
    "        31, 32])\n",
    "    # concatenating source and destination to form connections represented as directed, as DGL by default doesn't include directed.\n",
    "    u = np.concatenate([src,dst])\n",
    "    v = np.concatenate([dst,src])\n",
    "    return dgl.graph((u,v))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "9kdcKlCtV1DC",
    "outputId": "e5f0491d-f4bc-4868-fce0-f512b9ca68f3"
   },
   "outputs": [],
   "source": [
    "g = build_karate_club_graph()\n",
    "# g.ndata['features'] = torch.zeros(34,1)\n",
    "# print(g)\n",
    "# print(nx)\n",
    "nx_g = g.to_networkx().to_undirected()\n",
    "# print(type(nx_g))\n",
    "'''A layout function for returning coordinates of edges'''\n",
    "pos = nx.kamada_kawai_layout(nx_g)\n",
    "# print(pos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 265
    },
    "id": "JdVX0h88WMe3",
    "outputId": "cb5ab5aa-2f16-4328-c401-16d54dd350a9"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "'''nothing interesting here'''\n",
    "for i in range(34):\n",
    "  plt.scatter(pos[i][0],pos[i][1],color='red')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 319
    },
    "id": "hbCtUuw0Wcvz",
    "outputId": "9f8e0b0e-59c8-452f-bf8a-7810d18cf486"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "'''The real deal'''\n",
    "nx.draw(nx_g, pos, with_labels=True, node_color=[[1,.4,.5]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "_8bc132ddJed",
    "outputId": "cbf5d098-67b7-492a-9e01-ab46bac4899a"
   },
   "outputs": [],
   "source": [
    "'''Creation of GCN starts'''\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "\n",
    "# a=torch.tensor([[1,2,3],[4,5,6]])\n",
    "embed = nn.Embedding(34,5)\n",
    "# print(embed(a))\n",
    "# print(type(embed))\n",
    "# print(embed.weight)\n",
    "g.ndata['feat'] = embed.weight\n",
    "# g.ndata['feat'][[0,1]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "6qFVksnMnQ5t"
   },
   "outputs": [],
   "source": [
    "from dgl.nn.pytorch import GraphConv\n",
    "\n",
    "class GCN(nn.Module):\n",
    "    def __init__(self, in_feats, hidden_size, num_classes):\n",
    "        # super(GCN, self).__init__()\n",
    "        super().__init__()\n",
    "        self.conv1 = GraphConv(in_feats, hidden_size)\n",
    "        self.conv2 = GraphConv(hidden_size, num_classes)\n",
    "\n",
    "    def forward(self, g, inputs):\n",
    "        h = self.conv1(g, inputs)\n",
    "        h = torch.relu(h)\n",
    "        h = self.conv2(g, h)\n",
    "        return h\n",
    "net = GCN(5, 5, 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "VCMpfdBXXCPm"
   },
   "outputs": [],
   "source": [
    "inputs = embed.weight\n",
    "# print(inputs.size())\n",
    "labeled_nodes = torch.tensor([0, 33])  # only the instructor and the president nodes are labeled\n",
    "labels = torch.tensor([0, 1])  # their labels are different"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "collapsed": true,
    "id": "U65PVwv5XvND",
    "jupyter": {
     "outputs_hidden": true
    },
    "outputId": "85ab4a09-3a6b-4dd1-9ecc-b5cc03d903ef",
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0 | Loss: 0.6507\n",
      "Epoch 1 | Loss: 0.6288\n",
      "Epoch 2 | Loss: 0.6068\n",
      "Epoch 3 | Loss: 0.5842\n",
      "Epoch 4 | Loss: 0.5616\n",
      "Epoch 5 | Loss: 0.5388\n",
      "Epoch 6 | Loss: 0.5154\n",
      "Epoch 7 | Loss: 0.4912\n",
      "Epoch 8 | Loss: 0.4669\n",
      "Epoch 9 | Loss: 0.4428\n",
      "Epoch 10 | Loss: 0.4183\n",
      "Epoch 11 | Loss: 0.3929\n",
      "Epoch 12 | Loss: 0.3672\n",
      "Epoch 13 | Loss: 0.3412\n",
      "Epoch 14 | Loss: 0.3158\n",
      "Epoch 15 | Loss: 0.2914\n",
      "Epoch 16 | Loss: 0.2677\n",
      "Epoch 17 | Loss: 0.2444\n",
      "Epoch 18 | Loss: 0.2220\n",
      "Epoch 19 | Loss: 0.2005\n",
      "Epoch 20 | Loss: 0.1796\n",
      "Epoch 21 | Loss: 0.1602\n",
      "Epoch 22 | Loss: 0.1421\n",
      "Epoch 23 | Loss: 0.1252\n",
      "Epoch 24 | Loss: 0.1099\n",
      "Epoch 25 | Loss: 0.0960\n",
      "Epoch 26 | Loss: 0.0836\n",
      "Epoch 27 | Loss: 0.0726\n",
      "Epoch 28 | Loss: 0.0629\n",
      "Epoch 29 | Loss: 0.0544\n",
      "Epoch 30 | Loss: 0.0470\n",
      "Epoch 31 | Loss: 0.0407\n",
      "Epoch 32 | Loss: 0.0352\n",
      "Epoch 33 | Loss: 0.0306\n",
      "Epoch 34 | Loss: 0.0266\n",
      "Epoch 35 | Loss: 0.0232\n",
      "Epoch 36 | Loss: 0.0203\n",
      "Epoch 37 | Loss: 0.0178\n",
      "Epoch 38 | Loss: 0.0157\n",
      "Epoch 39 | Loss: 0.0139\n",
      "Epoch 40 | Loss: 0.0123\n",
      "Epoch 41 | Loss: 0.0110\n",
      "Epoch 42 | Loss: 0.0098\n",
      "Epoch 43 | Loss: 0.0089\n",
      "Epoch 44 | Loss: 0.0080\n",
      "Epoch 45 | Loss: 0.0073\n",
      "Epoch 46 | Loss: 0.0066\n",
      "Epoch 47 | Loss: 0.0060\n",
      "Epoch 48 | Loss: 0.0056\n",
      "Epoch 49 | Loss: 0.0051\n"
     ]
    }
   ],
   "source": [
    "import itertools\n",
    "\n",
    "optimizer = torch.optim.Adam(itertools.chain(net.parameters(), embed.parameters()), lr=0.01)\n",
    "all_logits = []\n",
    "for epoch in range(50):\n",
    "    logits = net(g, inputs)\n",
    "    # we save the logits for visualization later\n",
    "    all_logits.append(logits.detach())\n",
    "    logp = F.log_softmax(logits, 1)\n",
    "    # we only compute loss for labeled nodes\n",
    "    loss = F.nll_loss(logp[labeled_nodes], labels)\n",
    "\n",
    "    optimizer.zero_grad()\n",
    "    loss.backward()\n",
    "    optimizer.step()\n",
    "\n",
    "    print('Epoch %d | Loss: %.4f' % (epoch, loss.item()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "2JXMKOBDmixT",
    "outputId": "69072133-8d1a-4226-c762-8b4abea45073",
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total epochs : 50\n",
      "Size of each graph in an epoch : torch.Size([34, 2])\n"
     ]
    }
   ],
   "source": [
    "print(f'Total epochs : {len(all_logits)}')\n",
    "print(f'Size of each graph in an epoch : {all_logits[0].size()}')\n",
    "\n",
    "# a=[1,2,3,4.5]\n",
    "# max(a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "id": "zNshBy55lxPz"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.animation as animation\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "def draw(i):\n",
    "    '''Unique colors for node with feature 0 and 1'''\n",
    "    cls1color = '#00FFFF'\n",
    "    cls2color = '#FF00FF'\n",
    "    '''Feature embeddings are taken as coordinates or vice-versa. Its bidirectional, you know'''\n",
    "    pos = {}\n",
    "    '''Colors for every node is stored in a list for passing argument to networkx'''\n",
    "    colors = []\n",
    "    '''For every epoch(i), the feature embeddings are taken for the respective node.'''\n",
    "    for v in range(34):\n",
    "        pos[v] = all_logits[i][v].numpy()\n",
    "        cls = pos[v].argmax() #argmax() returns the position of the max value present in the list. Either 0 or 1 in this case.\n",
    "        colors.append(cls1color if cls else cls2color)\n",
    "    ax.cla() #clears the axes\n",
    "    ax.axis('off') #axes are set to off\n",
    "    ax.set_title('Epoch: %d' % i)\n",
    "    \n",
    "#     nx.draw_networkx(nx_g.to_undirected(), pos, node_color=colors, with_labels=True, node_size=300, ax=ax)\n",
    "    nx.draw(nx_g.to_undirected(), pos, node_color=colors, with_labels=True, node_size=300, ax=ax)\n",
    "\n",
    "# fig = plt.figure(dpi=100)\n",
    "fig = plt.figure() # by default, dpi is 100\n",
    "fig.clf() # to clear the figure, just as a best practice\n",
    "ax = fig.subplots() # single axis\n",
    "draw(0)  # draw the prediction of the first epoch\n",
    "# plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6ZPJs64apK7D",
    "outputId": "c094cab6-442d-4f3b-fa6d-f626984cfce5"
   },
   "outputs": [],
   "source": [
    "'''Not working.'''\n",
    "ani = animation.FuncAnimation(fig, func=draw, frames=np.arange(0,len(all_logits)), interval=200)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 531
    },
    "id": "xYiE3ae-aSaL",
    "outputId": "f827ff19-5bbe-4305-b5cd-47ebf4a27577"
   },
   "source": [
    "<h1>DGL with CoraDataset</h1>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "import dgl\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  NumNodes: 2708\n",
      "  NumEdges: 10556\n",
      "  NumFeats: 1433\n",
      "  NumClasses: 7\n",
      "  NumTrainingSamples: 140\n",
      "  NumValidationSamples: 500\n",
      "  NumTestSamples: 1000\n",
      "Done loading data from cached files.\n",
      "Number of categories: 7\n"
     ]
    }
   ],
   "source": [
    "import dgl.data\n",
    "\n",
    "dataset = dgl.data.CoraGraphDataset()\n",
    "print('Number of categories:', dataset.num_classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Node features\n",
      "Feature space dimension : torch.Size([2708, 1433])\n",
      "{'train_mask': tensor([ True,  True,  True,  ..., False, False, False]), 'label': tensor([3, 4, 4,  ..., 3, 3, 3]), 'val_mask': tensor([False, False, False,  ..., False, False, False]), 'test_mask': tensor([False, False, False,  ...,  True,  True,  True]), 'feat': tensor([[0., 0., 0.,  ..., 0., 0., 0.],\n",
      "        [0., 0., 0.,  ..., 0., 0., 0.],\n",
      "        [0., 0., 0.,  ..., 0., 0., 0.],\n",
      "        ...,\n",
      "        [0., 0., 0.,  ..., 0., 0., 0.],\n",
      "        [0., 0., 0.,  ..., 0., 0., 0.],\n",
      "        [0., 0., 0.,  ..., 0., 0., 0.]])}\n",
      "Edge features\n",
      "{}\n"
     ]
    }
   ],
   "source": [
    "g = dataset[0]\n",
    "print('Node features')\n",
    "print(f\"Feature space dimension : {g.ndata['feat'].size()}\")\n",
    "print(g.ndata)\n",
    "print('Edge features')\n",
    "print(g.edata)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "from dgl.nn import GraphConv\n",
    "\n",
    "class GCN_cora(nn.Module):\n",
    "    def __init__(self, in_feats, h_feats, num_classes):\n",
    "        super().__init__()\n",
    "        self.conv1 = GraphConv(in_feats, h_feats)\n",
    "        self.conv2 = GraphConv(h_feats, num_classes)\n",
    "\n",
    "    def forward(self, g, in_feat):\n",
    "        h = self.conv1(g, in_feat)\n",
    "        h = F.relu(h)\n",
    "        h = self.conv2(g, h)\n",
    "        return h\n",
    "\n",
    "# Create the model with given dimensions\n",
    "model = GCN_cora(g.ndata['feat'].shape[1], 16, dataset.num_classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In epoch 0, loss: 1.945, val acc: 0.154 (best 0.154), test acc: 0.168 (best 0.168)\n",
      "In epoch 10, loss: 1.787, val acc: 0.482 (best 0.518), test acc: 0.473 (best 0.533)\n",
      "In epoch 20, loss: 1.545, val acc: 0.576 (best 0.576), test acc: 0.582 (best 0.582)\n",
      "In epoch 30, loss: 1.244, val acc: 0.628 (best 0.628), test acc: 0.639 (best 0.639)\n",
      "In epoch 40, loss: 0.925, val acc: 0.656 (best 0.658), test acc: 0.670 (best 0.663)\n",
      "In epoch 50, loss: 0.646, val acc: 0.708 (best 0.708), test acc: 0.715 (best 0.715)\n",
      "In epoch 60, loss: 0.435, val acc: 0.722 (best 0.722), test acc: 0.733 (best 0.733)\n",
      "In epoch 70, loss: 0.291, val acc: 0.742 (best 0.742), test acc: 0.755 (best 0.755)\n",
      "In epoch 80, loss: 0.197, val acc: 0.756 (best 0.756), test acc: 0.764 (best 0.764)\n",
      "In epoch 90, loss: 0.138, val acc: 0.760 (best 0.762), test acc: 0.755 (best 0.755)\n"
     ]
    }
   ],
   "source": [
    "def train(g, model):\n",
    "    optimizer = torch.optim.Adam(model.parameters(), lr=0.01)\n",
    "    best_val_acc = 0\n",
    "    best_test_acc = 0\n",
    "\n",
    "    features = g.ndata['feat']\n",
    "    labels = g.ndata['label']\n",
    "    train_mask = g.ndata['train_mask']\n",
    "    val_mask = g.ndata['val_mask']\n",
    "    test_mask = g.ndata['test_mask']\n",
    "    for e in range(100):\n",
    "        # Forward\n",
    "        logits = model(g, features)\n",
    "\n",
    "        # Compute prediction\n",
    "        pred = logits.argmax(1)\n",
    "\n",
    "        # Compute loss\n",
    "        # Note that you should only compute the losses of the nodes in the training set.\n",
    "        '''CrossEntropyLoss is used as it is a categorical set.'''\n",
    "        loss = F.cross_entropy(logits[train_mask], labels[train_mask])\n",
    "\n",
    "        # Compute accuracy on training/validation/test\n",
    "        train_acc = (pred[train_mask] == labels[train_mask]).float().mean()\n",
    "        val_acc = (pred[val_mask] == labels[val_mask]).float().mean()\n",
    "        test_acc = (pred[test_mask] == labels[test_mask]).float().mean()\n",
    "\n",
    "        # Save the best validation accuracy and the corresponding test accuracy.\n",
    "        if best_val_acc < val_acc:\n",
    "            best_val_acc = val_acc\n",
    "            best_test_acc = test_acc\n",
    "\n",
    "        # Backward\n",
    "        optimizer.zero_grad()\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "        if e % 10 == 0:\n",
    "            print('In epoch {}, loss: {:.3f}, val acc: {:.3f} (best {:.3f}), test acc: {:.3f} (best {:.3f})'.format(\n",
    "                e, loss, val_acc, best_val_acc, test_acc, best_test_acc))\n",
    "model = GCN(g.ndata['feat'].shape[1], 16, dataset.num_classes)\n",
    "train(g, model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([1433, 16])\n",
      "torch.Size([16])\n",
      "torch.Size([16, 7])\n",
      "torch.Size([7])\n"
     ]
    }
   ],
   "source": [
    "damn=model.parameters()\n",
    "i=0\n",
    "while(i<4):\n",
    "    print(next(damn).size())\n",
    "    i+=1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([3, 4, 4,  ..., 3, 3, 3])\n",
      "torch.Size([2708])\n",
      "torch.Size([2708, 7])\n",
      "tensor(3)\n"
     ]
    }
   ],
   "source": [
    "features = g.ndata['feat']\n",
    "labels = g.ndata['label']\n",
    "train_mask = g.ndata['train_mask']\n",
    "val_mask = g.ndata['val_mask']\n",
    "test_mask = g.ndata['test_mask']\n",
    "\n",
    "logits = model(g, features)\n",
    "\n",
    "# Compute prediction\n",
    "pred = logits.argmax(1)\n",
    "print(pred)\n",
    "print(pred.size())\n",
    "print(logits.size())\n",
    "print(logits[0].argmax(0))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1>Knowledge graph</h1>\n",
    "<p>Headline files:</p>\n",
    "<ol>\n",
    "    CNN\n",
    "    <li>boeing.csv</li>\n",
    "    <li>google.csv</li>\n",
    "    <li>walmart.csv</li>\n",
    "    <br>Reuters\n",
    "    <li>apple.csv</li>\n",
    "    <li>microsoft.csv</li>\n",
    "    <li>samsung.csv</li>\n",
    "</ol>\n",
    "\n",
    "Headlines dataset are present in the directory \"headlines\"\n",
    "To install pandas, perform the following command in the terminal\n",
    "\n",
    "> **pip install pandas**\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "directory = './headlines/'\n",
    "headline_filenames = ['boeing.csv','google.csv','walmart.csv','apple.csv','microsoft.csv','samsung.csv']\n",
    "stock_filenames = {'apple':'AAPL.csv','boeing':'BA.csv','google':'GOOG.csv','microsoft':'MSFT.csv','samsung':'SSNLF.csv','walmart':'WMT.csv'}\n",
    "companies = ['boeing','google','walmart','apple','microsoft','samsung']\n",
    "months = ['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep','Oct','Nov','Dec']\n",
    "# headline_filenames[0][:-4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'filenames' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-3-eece1e4604d9>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[1;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mfilenames\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m     \u001b[0ma\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mpd\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdirectory\u001b[0m\u001b[1;33m+\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0ma\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mhead\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'filenames' is not defined"
     ]
    }
   ],
   "source": [
    "for i in filenames:\n",
    "    a=pd.read_csv(directory+i)\n",
    "    print(a.head(2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since Apple and Microsoft don't have headers in their csv files, the following code includes them. Just execute it once."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original dataset of apple.csv\n",
      "              Date                                              Title\n",
      "0  2011/10/18 0:00                 Apple iPhone 4S: Believe the hype?\n",
      "1  2011/10/18 0:00  Tech wrap: Apple misses, Intel beats quarterly...\n"
     ]
    }
   ],
   "source": [
    "print('Original dataset of apple.csv')\n",
    "apple=pd.read_csv(directory+'apple.csv')\n",
    "print(apple.head(2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Updated dataset of apple.csv\n",
      "              Date                                              Title\n",
      "0  2011/10/18 0:00                 Apple iPhone 4S: Believe the hype?\n",
      "1  2011/10/18 0:00  Tech wrap: Apple misses, Intel beats quarterly...\n"
     ]
    }
   ],
   "source": [
    "print('Updated dataset of apple.csv')\n",
    "apple.to_csv(directory+'apple.csv',header=[\"Date\",\"Title\"],index=False)\n",
    "apple=pd.read_csv(directory+'apple.csv')\n",
    "print(apple.head(2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original dataset of microsoft.csv\n",
      "                 Date                                              Title\n",
      "0  2011-10-26 0:00:00  Tech wrap: RIM delays tablet update, Nokia unv...\n",
      "1  2011-11-01 0:00:00  Tech wrap: Yahoo finds interclick, pays $270 m...\n"
     ]
    }
   ],
   "source": [
    "print('Original dataset of microsoft.csv')\n",
    "microsoft=pd.read_csv(directory+'microsoft.csv')\n",
    "print(microsoft.head(2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Updated dataset of microsoft.csv\n",
      "                 Date                                              Title\n",
      "0  2011-10-26 0:00:00  Tech wrap: RIM delays tablet update, Nokia unv...\n",
      "1  2011-11-01 0:00:00  Tech wrap: Yahoo finds interclick, pays $270 m...\n"
     ]
    }
   ],
   "source": [
    "print('Updated dataset of microsoft.csv')\n",
    "microsoft.to_csv(directory+'microsoft.csv',header=[\"Date\",\"Title\"],index=False)\n",
    "microsoft=pd.read_csv(directory+'microsoft.csv')\n",
    "print(microsoft.head(2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3>Changing all the date category's format to yyyy/m/d and headlines to smallercase</h3>\n",
    "1. Google <br>\n",
    "2. Boeing <br>\n",
    "3. Walmart <br>\n",
    "4. Samsung <br>\n",
    "5. Microsoft"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "For Google, Boeing and Walmart\n",
    "'''\n",
    "def datePreProcess(company):\n",
    "    df=pd.read_csv('./headlines/'+company+'.csv')\n",
    "    df_len = len(df.index) # no of rows\n",
    "    if(company in ['google','boeing','walmart']):\n",
    "        for i in range(df_len):\n",
    "            temp = df['Date'][i].split('-')\n",
    "            temp[2] = '20'+temp[2]\n",
    "            df['Date'][i] = temp[2]+'/'+str(months.index(temp[1])+1)+'/'+temp[0]\n",
    "            df['Title'][i] = df['Title'][i].lower()\n",
    "        df.to_csv('./headlines/Pre-processed/'+company+'.csv',header=[\"Date\",\"Title\"],index=False)\n",
    "        \n",
    "    if(company=='samsung'):\n",
    "        for i in range(df_len):\n",
    "            temp = df['Date'][i].split('-')\n",
    "            if(temp[2][0]=='0'): # for day\n",
    "                temp[2]=temp[2][-1]\n",
    "            if(temp[1][0]=='0'): # for month\n",
    "                temp[1]=temp[1][-1]\n",
    "            df['Date'][i] = '/'.join(temp)\n",
    "            df['Title'][i] = df['Title'][i].lower()\n",
    "        df.to_csv('./headlines/Pre-processed/'+company+'.csv',header=[\"Date\",\"Title\"],index=False)\n",
    "        \n",
    "    if(company=='microsoft'):\n",
    "        for i in range(df_len):\n",
    "            temp = df['Date'][i][:-8].split('-')\n",
    "            if(temp[2][0]=='0'):\n",
    "                temp[2]=temp[2][-1]\n",
    "            if(temp[1][0]=='0'): # for month\n",
    "                temp[1]=temp[1][-1]\n",
    "            df['Date'][i] = '/'.join(temp)\n",
    "            df['Title'][i] = df['Title'][i].lower()\n",
    "        df.to_csv('./headlines/Pre-processed/'+company+'.csv',header=[\"Date\",\"Title\"],index=False)\n",
    "    \n",
    "    if(company=='apple'):\n",
    "        for i in range(df_len):\n",
    "            df['Date'][i] = df['Date'][i][:-5]\n",
    "            df['Title'][i] = df['Title'][i].lower()\n",
    "        df.to_csv('./headlines/Pre-processed/'+company+'.csv',header=[\"Date\",\"Title\"],index=False)\n",
    "    print(f'Successfully converted headlines to lowercase and updated the date format of {company}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully converted headlines to lowercase and updated the date format of google\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "Google headline date preprocessing\n",
    "'''\n",
    "datePreProcess('google')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully converted headlines to lowercase and updated the date format of boeing\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "Boeing headline date preprocessing\n",
    "'''\n",
    "datePreProcess('boeing')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully converted headlines to lowercase and updated the date format of walmart\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "Walmart headline date preprocessing\n",
    "'''\n",
    "datePreProcess('walmart')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully converted headlines to lowercase and updated the date format of samsung\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "Samsung headline date preprocessing\n",
    "'''\n",
    "datePreProcess('samsung')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully converted headlines to lowercase and updated the date format of microsoft\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "Microsoft headline date preprocessing\n",
    "'''\n",
    "datePreProcess('microsoft')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully converted headlines to lowercase and updated the date format of apple\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "Apple headline date preprocessing\n",
    "'''\n",
    "datePreProcess('apple')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting dgl\n",
      "  Downloading dgl-0.6.1-cp39-cp39-win_amd64.whl (3.1 MB)\n",
      "Requirement already satisfied: numpy>=1.14.0 in c:\\users\\balaji\\appdata\\local\\programs\\python\\python39\\lib\\site-packages (from dgl) (1.20.2)\n",
      "Collecting requests>=2.19.0\n",
      "  Downloading requests-2.26.0-py2.py3-none-any.whl (62 kB)\n",
      "Collecting networkx>=2.1\n",
      "  Downloading networkx-2.6.2-py3-none-any.whl (1.9 MB)\n",
      "Requirement already satisfied: scipy>=1.1.0 in c:\\users\\balaji\\appdata\\local\\programs\\python\\python39\\lib\\site-packages (from dgl) (1.6.3)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.21.1 in c:\\users\\balaji\\appdata\\local\\programs\\python\\python39\\lib\\site-packages (from requests>=2.19.0->dgl) (1.26.5)\n",
      "Collecting idna<4,>=2.5; python_version >= \"3\"\n",
      "  Downloading idna-3.2-py3-none-any.whl (59 kB)\n",
      "Collecting certifi>=2017.4.17\n",
      "  Downloading certifi-2021.5.30-py2.py3-none-any.whl (145 kB)\n",
      "Collecting charset-normalizer~=2.0.0; python_version >= \"3\"\n",
      "  Downloading charset_normalizer-2.0.3-py3-none-any.whl (35 kB)\n",
      "Installing collected packages: idna, certifi, charset-normalizer, requests, networkx, dgl\n",
      "Successfully installed certifi-2021.5.30 charset-normalizer-2.0.3 dgl-0.6.1 idna-3.2 networkx-2.6.2 requests-2.26.0\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: You are using pip version 20.2.3; however, version 21.2.1 is available.\n",
      "You should consider upgrading via the 'c:\\users\\balaji\\appdata\\local\\programs\\python\\python39\\python.exe -m pip install --upgrade pip' command.\n"
     ]
    }
   ],
   "source": [
    "pip install dgl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting torch\n",
      "  Downloading torch-1.9.0-cp39-cp39-win_amd64.whl (222.0 MB)\n",
      "Collecting typing-extensions\n",
      "  Downloading typing_extensions-3.10.0.0-py3-none-any.whl (26 kB)\n",
      "Installing collected packages: typing-extensions, torch\n",
      "Successfully installed torch-1.9.0 typing-extensions-3.10.0.0\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: You are using pip version 20.2.3; however, version 21.2.1 is available.\n",
      "You should consider upgrading via the 'c:\\users\\balaji\\appdata\\local\\programs\\python\\python39\\python.exe -m pip install --upgrade pip' command.\n"
     ]
    }
   ],
   "source": [
    "pip install torch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using backend: pytorch\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import dgl\n",
    "import numpy as np\n",
    "import networkx as nx\n",
    "\n",
    "G = nx.Graph()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# colors=[]\n",
    "# company_index=0\n",
    "# for file in headline_filenames:\n",
    "#     company_name = companies[company_index]\n",
    "#     comp_news = pd.read_csv('./headlines/Pre-processed/'+file)\n",
    "#     length = len(comp_news.index)\n",
    "#     for index in range(length):\n",
    "#         news = comp_news['Title'][index]\n",
    "#         G.add_edges_from([(company_name,news)])\n",
    "#         for comp in companies:\n",
    "#             if comp in news:\n",
    "#                 G.add_edges_from([(comp,news)])\n",
    "#             else:\n",
    "#                 continue\n",
    "# #     print(length)\n",
    "# nx.draw(G,with_labels=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "'''\n",
    "Google KG complete\n",
    "'''\n",
    "Google_G = nx.Graph()\n",
    "Google_G.add_node('Google')\n",
    "Google = pd.read_csv('./headlines/Pre-processed/google.csv')\n",
    "length_google = len(Google.index)\n",
    "\n",
    "for i in range(length_google):\n",
    "    Google_G.add_edges_from([('Google',Google['Title'][i])])\n",
    "pos = nx.kamada_kawai_layout(Google_G)\n",
    "nx.draw(Google_G,pos,with_labels=False)\n",
    "# 1,7,2,3,"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "ename": "IndexError",
     "evalue": "index 0 is out of bounds for axis 0 with size 0",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mIndexError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-5-8e9d16fffe9e>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     11\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m10\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     12\u001b[0m \u001b[1;31m#     stock_value = str(Google_stock.loc[Google_stock['Date']==Google['Date'][i]]['Close'])\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 13\u001b[1;33m     \u001b[0mstock_value\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mGoogle_stock\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mGoogle_stock\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0meq\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mGoogle\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m'Date'\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0many\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mto_numpy\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m4\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     14\u001b[0m \u001b[1;31m#     print(stock_value if stock_value else print('no'))\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     15\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mstock_value\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mIndexError\u001b[0m: index 0 is out of bounds for axis 0 with size 0"
     ]
    }
   ],
   "source": [
    "'''\n",
    "Google KG minimal\n",
    "'''\n",
    "Google_G_min = nx.DiGraph()\n",
    "Google_G_min.add_node('Google')\n",
    "Google = pd.read_csv('./headlines/Pre-processed/google.csv')\n",
    "Google_stock = pd.read_csv('./Stock Price/GOOG.csv')\n",
    "\n",
    "# length_google = len(Google.index)\n",
    "colors=['red']\n",
    "for i in range(10):\n",
    "#     stock_value = str(Google_stock.loc[Google_stock['Date']==Google['Date'][i]]['Close'])\n",
    "    stock_value = Google_stock[Google_stock.eq(Google['Date'][i]).any(1)].to_numpy()[0][4]\n",
    "#     print(stock_value if stock_value else print('no'))\n",
    "    print(stock_value)\n",
    "    \n",
    "#     Google_G_min.add_edge('Google',Google['Title'][i])\n",
    "#     Google_G_min.add_edge(Google['Title'][i],Google['Date'][i])\n",
    "#     Google_G_min.add_edge(Google['Date'][i],'Google')\n",
    "    \n",
    "#     Google_G_min.add_edge(Google['Date'][i],stock_value)\n",
    "\n",
    "# pos = nx.kamada_kawai_layout(Google_G_min)\n",
    "pos = nx.random_layout(Google_G_min,center=None,dim=2)\n",
    "# nx.draw(Google_G_min,pos,with_labels=True,node_color='orange')\n",
    "nx.draw(Google_G_min,pos,with_labels=True,node_color='orange')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       False\n",
       "1       False\n",
       "2       False\n",
       "3       False\n",
       "4       False\n",
       "        ...  \n",
       "2702    False\n",
       "2703    False\n",
       "2704    False\n",
       "2705     True\n",
       "2706    False\n",
       "Name: Date, Length: 2707, dtype: bool"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stock_value = Google_stock.loc[Google_stock['Date']==Google['Date'][4]]['Close'].iloc[0]\n",
    "stock_value\n",
    "type(stock_value)\n",
    "\n",
    "# 2017/2/23\n",
    "Google_stock['Date']=='2017/2/23'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "831.330017"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# GG = nx.lollipop_graph(4, 10)\n",
    "GG = nx.DiGraph()\n",
    "GG.add_edges_from([(1,2,{'close':'yep'})])\n",
    "GG.add_edge(2,3)\n",
    "pos = nx.random_layout(GG,center=None,dim=2)\n",
    "nx.draw(GG,pos,with_labels=True)\n",
    "\n",
    "a=Google_stock[Google_stock.eq('2017/2/23').any(1)].to_numpy()[0][4]\n",
    "# b=a.to_numpy()\n",
    "# b[0][4]\n",
    "a\n",
    "# print(a['Close'].value())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KDJ (Min and Max) 0.0\t 1.0\n",
      "W%R (Min and Max) -1.0\t 0.0\n",
      "RSI (Min and Max) -92.080017\t 41.369995\n",
      "\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "KDJ  |  W%R  |  RSI\n",
    "'''\n",
    "import pandas as pd\n",
    "goog = pd.read_csv('Stock Price\\GOOG.csv')\n",
    "print(f\"KDJ (Min and Max) {goog['KDJ'].min()}\\t {goog['KDJ'].max()}\")\n",
    "print(f\"W%R (Min and Max) {goog['W%R'].min()}\\t {goog['W%R'].max()}\")\n",
    "print(f\"RSI (Min and Max) {goog['RSI'].min()}\\t {goog['RSI'].max()}\")\n",
    "row_count = len(goog)\n",
    "'''\n",
    "Labels\n",
    "0 - overbought > 0.8 (sell)\n",
    "1 - oversold < 0.2 (buy)\n",
    "2 - nil trend [0.2,0.8]\n",
    "Note : Only AAPL has a column called as label\n",
    "'''\n",
    "# goog['KDJ'].sum()/len(goog['KDJ'].index)\n",
    "# goog['KDJ'].max()\n",
    "print('')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h4>KDJ | W%R</h4>\n",
    "If it is between 20 and 80 the market is neutral, if it is above 80 it is bullish/overbought and if it is below 20 it is bearish/oversold. If it is below 0 or above 100 it is very bearish or very bullish, but also very-oversold and very-overbought.\n",
    "\n",
    "Overbought --> overvalued that its intrinsic(bullish)<br>\n",
    "Oversold --> undervalued that its intrinsic(bearish)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.plot(goog['High'])\n",
    "plt.show()\n",
    "plt.plot(goog['Close'],color='red')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "label=[]\n",
    "for i in range(row_count):\n",
    "    if(goog['KDJ'][i] < 0.2):\n",
    "        label.append(1)\n",
    "    elif(goog['KDJ'][i] > 0.8):\n",
    "        label.append(0)\n",
    "    else:\n",
    "        status = ((goog['Close'][i]-goog['Open'][i])/goog['Open'][i])*100\n",
    "        if(status < -0.66):\n",
    "            label.append(0)\n",
    "        elif(status > 0.66):\n",
    "            label.append(1)\n",
    "        else:\n",
    "            label.append(2)\n",
    "# print(label)\n",
    "plt.hist(label)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "950\n",
      "841\n",
      "916\n"
     ]
    }
   ],
   "source": [
    "print(label.count(0))\n",
    "print(label.count(1))\n",
    "print(label.count(2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "x<20...80<x\n",
    "$100\n",
    "(+$70)\n",
    "$30\n",
    "((closing-opening)/(opening))*100\n",
    "<-0.75 sell\n",
    ">0.75 buy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Volume</th>\n",
       "      <th>KDJ</th>\n",
       "      <th>W%R</th>\n",
       "      <th>RSI</th>\n",
       "      <th>Label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2006/5/25</td>\n",
       "      <td>188.315186</td>\n",
       "      <td>190.262527</td>\n",
       "      <td>184.952072</td>\n",
       "      <td>190.257553</td>\n",
       "      <td>16495700</td>\n",
       "      <td>0.999063</td>\n",
       "      <td>-0.000937</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2006/5/26</td>\n",
       "      <td>191.032516</td>\n",
       "      <td>191.693222</td>\n",
       "      <td>188.787125</td>\n",
       "      <td>189.442856</td>\n",
       "      <td>7381600</td>\n",
       "      <td>0.225640</td>\n",
       "      <td>-0.774360</td>\n",
       "      <td>1.814697</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2006/5/30</td>\n",
       "      <td>187.917770</td>\n",
       "      <td>189.268982</td>\n",
       "      <td>184.524841</td>\n",
       "      <td>184.768265</td>\n",
       "      <td>8688100</td>\n",
       "      <td>0.051310</td>\n",
       "      <td>-0.948690</td>\n",
       "      <td>5.674591</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2006/5/31</td>\n",
       "      <td>185.692245</td>\n",
       "      <td>187.902878</td>\n",
       "      <td>182.204926</td>\n",
       "      <td>184.708649</td>\n",
       "      <td>16066300</td>\n",
       "      <td>0.439408</td>\n",
       "      <td>-0.560592</td>\n",
       "      <td>1.059616</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2006/6/1</td>\n",
       "      <td>185.563095</td>\n",
       "      <td>190.257553</td>\n",
       "      <td>184.599365</td>\n",
       "      <td>190.073761</td>\n",
       "      <td>12637600</td>\n",
       "      <td>0.967518</td>\n",
       "      <td>-0.032482</td>\n",
       "      <td>-4.365112</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2702</th>\n",
       "      <td>2017/2/17</td>\n",
       "      <td>823.020020</td>\n",
       "      <td>828.070007</td>\n",
       "      <td>821.655029</td>\n",
       "      <td>828.070007</td>\n",
       "      <td>1611000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-2.910034</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2703</th>\n",
       "      <td>2017/2/21</td>\n",
       "      <td>828.659973</td>\n",
       "      <td>833.450012</td>\n",
       "      <td>828.349976</td>\n",
       "      <td>831.659973</td>\n",
       "      <td>1262300</td>\n",
       "      <td>0.649014</td>\n",
       "      <td>-0.350986</td>\n",
       "      <td>-2.589966</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2704</th>\n",
       "      <td>2017/2/22</td>\n",
       "      <td>828.659973</td>\n",
       "      <td>833.250000</td>\n",
       "      <td>828.640015</td>\n",
       "      <td>830.760010</td>\n",
       "      <td>982900</td>\n",
       "      <td>0.459870</td>\n",
       "      <td>-0.540130</td>\n",
       "      <td>1.899963</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2705</th>\n",
       "      <td>2017/2/23</td>\n",
       "      <td>830.119995</td>\n",
       "      <td>832.460022</td>\n",
       "      <td>822.880005</td>\n",
       "      <td>831.330017</td>\n",
       "      <td>1472800</td>\n",
       "      <td>0.882046</td>\n",
       "      <td>-0.117954</td>\n",
       "      <td>0.429993</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2706</th>\n",
       "      <td>2017/2/24</td>\n",
       "      <td>827.729980</td>\n",
       "      <td>829.000000</td>\n",
       "      <td>824.200012</td>\n",
       "      <td>828.640015</td>\n",
       "      <td>1392200</td>\n",
       "      <td>0.925003</td>\n",
       "      <td>-0.074997</td>\n",
       "      <td>3.690002</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2707 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           Date        Open        High         Low       Close    Volume  \\\n",
       "0     2006/5/25  188.315186  190.262527  184.952072  190.257553  16495700   \n",
       "1     2006/5/26  191.032516  191.693222  188.787125  189.442856   7381600   \n",
       "2     2006/5/30  187.917770  189.268982  184.524841  184.768265   8688100   \n",
       "3     2006/5/31  185.692245  187.902878  182.204926  184.708649  16066300   \n",
       "4      2006/6/1  185.563095  190.257553  184.599365  190.073761  12637600   \n",
       "...         ...         ...         ...         ...         ...       ...   \n",
       "2702  2017/2/17  823.020020  828.070007  821.655029  828.070007   1611000   \n",
       "2703  2017/2/21  828.659973  833.450012  828.349976  831.659973   1262300   \n",
       "2704  2017/2/22  828.659973  833.250000  828.640015  830.760010    982900   \n",
       "2705  2017/2/23  830.119995  832.460022  822.880005  831.330017   1472800   \n",
       "2706  2017/2/24  827.729980  829.000000  824.200012  828.640015   1392200   \n",
       "\n",
       "           KDJ       W%R       RSI  Label  \n",
       "0     0.999063 -0.000937  0.000000      0  \n",
       "1     0.225640 -0.774360  1.814697      0  \n",
       "2     0.051310 -0.948690  5.674591      1  \n",
       "3     0.439408 -0.560592  1.059616      2  \n",
       "4     0.967518 -0.032482 -4.365112      0  \n",
       "...        ...       ...       ...    ...  \n",
       "2702  1.000000  0.000000 -2.910034      0  \n",
       "2703  0.649014 -0.350986 -2.589966      2  \n",
       "2704  0.459870 -0.540130  1.899963      2  \n",
       "2705  0.882046 -0.117954  0.429993      0  \n",
       "2706  0.925003 -0.074997  3.690002      0  \n",
       "\n",
       "[2707 rows x 10 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "goog['Label']=label\n",
    "goog"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = goog.iloc[:,1:9]\n",
    "y = goog.Label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.feature_selection import SelectKBest\n",
    "from sklearn.feature_selection import f_classif"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train (2165, 8) (2165,)\n",
      "Test (542, 8) (542,)\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1)\n",
    "print('Train', X_train.shape, y_train.shape)\n",
    "print('Test', X_test.shape, y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "fs = SelectKBest(score_func=f_classif, k='all')\n",
    "fs.fit(X_train, y_train)\n",
    "X_train_fs = fs.transform(X_train)\n",
    "X_test_fs = fs.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature 0: 5.372446\n",
      "Feature 1: 4.855520\n",
      "Feature 2: 5.973447\n",
      "Feature 3: 5.971765\n",
      "Feature 4: 41.722333\n",
      "Feature 5: 555.970817\n",
      "Feature 6: 555.970817\n",
      "Feature 7: 95.010051\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(fs.scores_)):\n",
    "    print('Feature %d: %f' % (i, fs.scores_[i]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = goog.iloc[:,6:8]\n",
    "y = goog.Label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7785977859778598\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state=1)\n",
    "knn = KNeighborsClassifier(n_neighbors=7)\n",
    "knn.fit(X_train, y_train)\n",
    "print(knn.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn import metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 74.35424354243543\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1)\n",
    "clf = DecisionTreeClassifier()\n",
    "clf = clf.fit(X_train,y_train)\n",
    "y_pred = clf.predict(X_test)\n",
    "print(\"Accuracy:\",metrics.accuracy_score(y_test, y_pred)*100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn import metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 66.78966789667896\n"
     ]
    }
   ],
   "source": [
    "x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1)\n",
    "model = GaussianNB()\n",
    "model.fit(x_train, y_train)\n",
    "y_pred = model.predict(x_test)\n",
    "print(\"Accuracy:\",metrics.accuracy_score(y_test, y_pred)*100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Maximum Accuracy: 75.83025830258303  at number of estimators =  1\n",
      "Minimum Accuracy: 71.95571955719558  at number of estimators =  2\n",
      "Average Accuracy: 73.6531365313652\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn import metrics\n",
    "xp =[]\n",
    "yp =[]\n",
    "maxp = 0\n",
    "minp = 100\n",
    "nM = 1\n",
    "nm = 1\n",
    "aa = 0\n",
    "x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1)\n",
    "for i in range (1,100):\n",
    "    clf=RandomForestClassifier(n_estimators=i)\n",
    "    clf.fit(x_train,y_train)\n",
    "    y_pred=clf.predict(x_test)\n",
    "    p = metrics.accuracy_score(y_test, y_pred)\n",
    "    if p*100 > maxp:\n",
    "        maxp = p*100\n",
    "        nM = i\n",
    "    if p*100 < minp:\n",
    "        minp = p*100\n",
    "        nm = i\n",
    "    xp.append(i)\n",
    "    yp.append(p*100)\n",
    "    aa = aa + p*100\n",
    "plt.plot(xp, yp)\n",
    "plt.ylim(0,100)\n",
    "plt.xlim(0,100)\n",
    "plt.xlabel('Number of estimators')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.title('Random Forest Classifier')\n",
    "plt.show()\n",
    "print(\"Maximum Accuracy:\",maxp,\" at number of estimators = \",nM)\n",
    "print(\"Minimum Accuracy:\",minp,\" at number of estimators = \",nm)\n",
    "print(\"Average Accuracy:\",aa/100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5277"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "TF-IDF trial\n",
    "'''\n",
    "import pandas as pd\n",
    "# from sklearn.feature_extraction.text import TfidfTransformer\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "googH = pd.read_csv('./headlines/Pre-processed/google.csv')\n",
    "type(googH['Title'])\n",
    "headlines_list=googH['Title'].values.tolist()\n",
    "len(headlines_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5277\n",
      "[[0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " ...\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]]\n"
     ]
    }
   ],
   "source": [
    "tfIdfVectorizer=TfidfVectorizer(use_idf=True)\n",
    "tfIdf = tfIdfVectorizer.fit_transform(headlines_list)\n",
    "print(tfIdf.shape[0])\n",
    "print(tfIdf.todense())\n",
    "df = pd.DataFrame(tfIdf[0].T.todense(), index=tfIdfVectorizer.get_feature_names(), columns=[\"TF-IDF\"])\n",
    "df = df.sort_values('TF-IDF', ascending=False)\n",
    "df\n",
    "df.to_csv('./TFIDF'+'.csv',header=[\"TF-IDF\"],index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'DataFrame' object has no attribute 'todense'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-9-82431c84b00d>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mdf\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mhead\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m20\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtodense\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;32mc:\\users\\balaji\\appdata\\local\\programs\\python\\python39\\lib\\site-packages\\pandas\\core\\generic.py\u001b[0m in \u001b[0;36m__getattr__\u001b[1;34m(self, name)\u001b[0m\n\u001b[0;32m   5463\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_info_axis\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_can_hold_identifiers_and_holds_name\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mname\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   5464\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mname\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 5465\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mobject\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m__getattribute__\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mname\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   5466\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   5467\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m__setattr__\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mname\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0mstr\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m->\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mAttributeError\u001b[0m: 'DataFrame' object has no attribute 'todense'"
     ]
    }
   ],
   "source": [
    "df.head(20).todense()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Rough work"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<4x52 sparse matrix of type '<class 'numpy.float64'>'\n",
       "\twith 61 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "corpus = [\n",
    "    'Hardest choice in the lives requires the strongest wills',\n",
    "    'When I am done, half of humanity will still exist. And I hope they remember you',\n",
    "    'Little one, its a simple calculus. This universe is finite, its resources finite. If life is left unchecked, life will cease to exist. It needs correction.',\n",
    "    'I used the stones to destroy the stones. It nearly killed me, but the work is done. It always will be. I am...inevitable'\n",
    "]\n",
    "\n",
    "vectorizer = TfidfVectorizer()\n",
    "X = vectorizer.fit_transform(corpus)\n",
    "# print(X[1])\n",
    "X\n",
    "#print(vectorizer.get_feature_names())\n",
    "# print(len(vectorizer.get_feature_names()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['calculus', 'cease', 'choice', 'correction', 'destroy', 'exist', 'finite', 'half', 'hardest', 'hope', 'humanity', 'inevitable', 'killed', 'left', 'life', 'little', 'lives', 'nearly', 'needs', 'remember', 'requires', 'resources', 'simple', 'stones', 'strongest', 'unchecked', 'universe', 'used', 'wills', 'work']\n",
      "30\n"
     ]
    }
   ],
   "source": [
    "vectorizer1 = TfidfVectorizer(analyzer='word', stop_words='english', use_idf=True)\n",
    "X = vectorizer1.fit_transform(corpus)\n",
    "# print(X)\n",
    "print(vectorizer1.get_feature_names())\n",
    "print(len(vectorizer1.get_feature_names()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['historic oscar victories for espn, netflix and amazon',\n",
       "  'apple, microsoft, paypal join legal fight for transgender rights',\n",
       "  \"why you shouldn't freak out (yet) about the 'cloudbleed' security leak\",\n",
       "  \"google's waymo sues uber over self-driving car technology\",\n",
       "  'tech criticizes trump on transgender reversal',\n",
       "  'google commits $11.5 million to racial justice efforts',\n",
       "  'lyft expands into 54 new cities, while uber does damage control',\n",
       "  'apple criticizes trump over reversal of transgender bathroom guidelines',\n",
       "  'these 10 stocks dominate the market',\n",
       "  \"russia accuses western media of spreading 'fake news'\"]]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[headlines_list[:10]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['apple', 'bathroom', 'criticizes', 'guidelines', 'reversal', 'transgender', 'trump']\n",
      "7\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(1, 7)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vectorizer2 = TfidfVectorizer(analyzer='word', stop_words='english', use_idf=True)\n",
    "X = vectorizer2.fit_transform([headlines_list[7]])\n",
    "# print(X)\n",
    "print(vectorizer2.get_feature_names())\n",
    "print(len(vectorizer2.get_feature_names()))\n",
    "X.shape # --> (documents,features). Features are unique words in a single document"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "include_colab_link": true,
   "name": "Anomaly detection in financial services using knowledge graphs and machine learning",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.9.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}