cnn.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import datetime\n",
    "import torch\n",
    "import torch.nn.functional as F\n",
    "import torchtext.data as data\n",
    "from src import obj_dict, movie_reviews_dataset, model_cnn, train\n",
    "\n",
    "args = obj_dict.objdict({\n",
    "    'batch_size': 1024,\n",
    "    'cuda': True and torch.cuda.is_available(),\n",
    "    'device': 0,\n",
    "    'epochs': 250,\n",
    "    'save_dir': './snapshots/cnn/',\n",
    "    'static': False,\n",
    "    'kernel_sizes': [3, 4, 5],\n",
    "    'embed_dim': 300,\n",
    "    'kernel_num': 100,\n",
    "    'dropout': 0.5,\n",
    "    'lr': 0.001,\n",
    "    'log_interval': 100,\n",
    "    'test_interval': 100,\n",
    "    'save_interval': 500\n",
    "})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Setup Movie Reviews data set loader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'batch_size': 1024, 'cuda': False, 'device': 0, 'epochs': 250, 'save_dir': './snapshots/cnn/', 'static': False, 'kernel_sizes': [3, 4, 5], 'embed_dim': 300, 'kernel_num': 100, 'dropout': 0.5, 'lr': 0.001, 'log_interval': 100, 'test_interval': 100, 'save_interval': 500}\n\nLoading data...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 21109 samples\n"
     ]
    }
   ],
   "source": [
    "# load MR dataset\n",
    "def mr(text_field, label_field, **kargs):\n",
    "    train_data, dev_data = movie_reviews_dataset.MR.splits(text_field, label_field, root='./data')\n",
    "    text_field.build_vocab(train_data, dev_data)\n",
    "    label_field.build_vocab(train_data, dev_data)\n",
    "    train_iter, dev_iter = data.Iterator.splits(\n",
    "        (train_data, dev_data),\n",
    "        batch_sizes=(args.batch_size, len(dev_data)),\n",
    "        **kargs)\n",
    "    return train_iter, dev_iter\n",
    "\n",
    "print(args)\n",
    "\n",
    "# load data\n",
    "print(\"\\nLoading data...\")\n",
    "text_field = data.Field(lower=True)\n",
    "label_field = data.Field(sequential=False)\n",
    "train_iter, dev_iter = mr(text_field, label_field, device=args.device if args.cuda else -1, repeat=False)\n",
    "print(\"Loaded\", len(text_field.vocab), \"samples\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\rBatch[100] - loss: 0.676938  error: 40.0000%(228/380)"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\nEvaluation - loss: 0.681569  error: 42.2139%(616/1066) \n\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\rBatch[200] - loss: 0.671291  error: 41.5789%(222/380)"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\nEvaluation - loss: 0.674037  error: 39.5872%(644/1066) \n\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-3-eb89830ad7d0>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[0;31m# model\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0mcnn\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel_cnn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mCNN_Text\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 8\u001b[0;31m \u001b[0mtrain\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_iter\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdev_iter\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcnn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/home/val/PycharmProjects/rcnn/src/train.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(train_iter, dev_iter, model, args)\u001b[0m\n\u001b[1;32m     27\u001b[0m                 \u001b[0mlogit\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfeature\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     28\u001b[0m                 \u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mF\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcross_entropy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlogit\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 29\u001b[0;31m                 \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     30\u001b[0m                 \u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     31\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.6/site-packages/torch/autograd/variable.py\u001b[0m in \u001b[0;36mbackward\u001b[0;34m(self, gradient, retain_variables)\u001b[0m\n\u001b[1;32m    143\u001b[0m                     'or with gradient w.r.t. the variable')\n\u001b[1;32m    144\u001b[0m             \u001b[0mgradient\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnew\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresize_as_\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfill_\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 145\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_execution_engine\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun_backward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mgradient\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mretain_variables\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    146\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    147\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mregister_hook\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhook\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.6/site-packages/torch/nn/_functions/conv.py\u001b[0m in \u001b[0;36mbackward\u001b[0;34m(self, grad_output)\u001b[0m\n\u001b[1;32m     46\u001b[0m             \u001b[0mgrad_output\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweight\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_view4d\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgrad_output\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweight\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     47\u001b[0m         grad_input = (self._grad_input(input, weight, grad_output)\n\u001b[0;32m---> 48\u001b[0;31m                       if self.needs_input_grad[0] else None)\n\u001b[0m\u001b[1;32m     49\u001b[0m         grad_weight, grad_bias = (\n\u001b[1;32m     50\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_grad_params\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbias\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgrad_output\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.6/site-packages/torch/nn/_functions/conv.py\u001b[0m in \u001b[0;36m_grad_input\u001b[0;34m(self, input, weight, grad_output)\u001b[0m\n\u001b[1;32m    112\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mgrad_input\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    113\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 114\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_thnn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'grad_input'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgrad_output\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    115\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    116\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_grad_params\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbias\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgrad_output\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.6/site-packages/torch/nn/_functions/conv.py\u001b[0m in \u001b[0;36m_thnn\u001b[0;34m(self, fn_name, input, weight, *args)\u001b[0m\n\u001b[1;32m    154\u001b[0m         \u001b[0mimpl\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_thnn_convs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mthnn_class_name\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    155\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgroups\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 156\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mimpl\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mfn_name\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_bufs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    157\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    158\u001b[0m             \u001b[0mres\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/anaconda/lib/python3.6/site-packages/torch/nn/_functions/conv.py\u001b[0m in \u001b[0;36mcall_grad_input\u001b[0;34m(self, bufs, input, weight, grad_output)\u001b[0m\n\u001b[1;32m    244\u001b[0m         \u001b[0margs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mparse_arguments\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0marguments\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m5\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbufs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkernel_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    245\u001b[0m         getattr(backend, fn.name)(backend.library_state, input, grad_output,\n\u001b[0;32m--> 246\u001b[0;31m                                   grad_input, weight, *args)\n\u001b[0m\u001b[1;32m    247\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mgrad_input\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    248\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mcall_grad_input\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ],
     "output_type": "error"
    }
   ],
   "source": [
    "args.embed_num = len(text_field.vocab)\n",
    "args.class_num = len(label_field.vocab) - 1\n",
    "\n",
    "args.snapshot_save_dir = os.path.join(args.save_dir, datetime.datetime.now().strftime('%Y-%m-%d_%H-%M-%S'))\n",
    "\n",
    "# model\n",
    "cnn = model_cnn.CNN_Text(args)\n",
    "train.train(train_iter, dev_iter, cnn, args)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Final Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\nEvaluation - loss: 0.576529  acc: 73.4522%(783/1066) \n\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(0.5765285241447291, 73.45215759849906, 783, 1066)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.eval(dev_iter, cnn, args)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    ""
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2.0
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}