vgg16_bn.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Module Import"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "\n",
    "import torch\n",
    "from torch.utils.data import DataLoader\n",
    "\n",
    "import torchvision.models as models\n",
    "from torchvision import transforms\n",
    "from torchvision.datasets import ImageFolder\n",
    "\n",
    "from tqdm import tqdm_notebook\n",
    "\n",
    "save_outputs = True # whether to save intermediate outputs for later reuse.\n",
    "output_dir = 'outputs/vgg16_bn/'\n",
    "os.makedirs(output_dir, exist_ok=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Initialize vgg16_bn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "net = models.vgg16_bn(pretrained=True)\n",
    "net.eval()\n",
    "net.cuda();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## and set hooks to extract hidden activations across layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "features_12 activations are registered\n",
      "features_16 activations are registered\n",
      "features_19 activations are registered\n",
      "features_22 activations are registered\n",
      "features_26 activations are registered\n",
      "features_29 activations are registered\n",
      "features_32 activations are registered\n",
      "features_36 activations are registered\n",
      "features_39 activations are registered\n",
      "features_42 activations are registered\n",
      "classifier_1 activations are registered\n",
      "classifier_4 activations are registered\n"
     ]
    }
   ],
   "source": [
    "hidden_outputs = {}\n",
    "def named_hook(name):\n",
    "    def hook(module, input, output):\n",
    "        hidden_outputs[name] = output\n",
    "    return hook\n",
    "\n",
    "# activations = dict(features_2=[], features_5=[], features_9=[],\n",
    "#                    features_12=[], features_16=[], features_19=[],\n",
    "#                    features_22=[], features_26=[], features_29=[],\n",
    "#                    features_32=[], features_36=[], features_39=[],\n",
    "#                    features_42=[], classifier_1=[], classifier_4=[])\n",
    "activations = dict(features_12=[], features_16=[], features_19=[],\n",
    "                   features_22=[], features_26=[], features_29=[],\n",
    "                   features_32=[], features_36=[], features_39=[],\n",
    "                   features_42=[], classifier_1=[], classifier_4=[])\n",
    "\n",
    "for name in activations.keys():\n",
    "    meta_arch_name, layer_idx = name.split('_')\n",
    "    layer_idx = int(layer_idx)\n",
    "    if meta_arch_name == 'features':\n",
    "        net.features[layer_idx].register_forward_hook(named_hook(meta_arch_name+'_'+str(layer_idx)))\n",
    "    elif meta_arch_name == 'classifier':\n",
    "        net.classifier[layer_idx].register_forward_hook(named_hook(meta_arch_name+'_'+str(layer_idx)))\n",
    "            \n",
    "    print('{}_{} activations are registered'.format(meta_arch_name, layer_idx))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Initialize dataset & data loader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# of images : 50000\n"
     ]
    }
   ],
   "source": [
    "img_root = './data/ILSVRC2012_img_val/'\n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize((224, 224)),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize(mean=(0.485, 0.456, 0.406),\n",
    "                         std=(0.229, 0.224, 0.225))\n",
    "])\n",
    "\n",
    "num_class = 1000\n",
    "imgs_per_class = 50\n",
    "batch_size = 25\n",
    "iters_per_class = imgs_per_class//batch_size\n",
    "assert imgs_per_class/batch_size == float(iters_per_class)\n",
    "\n",
    "num_workers = 6\n",
    "dset = ImageFolder(img_root, transform=transform)\n",
    "dloader = DataLoader(dset, batch_size=batch_size, drop_last=False, shuffle=False, num_workers=num_workers)\n",
    "print('# of images :', len(dset))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Extract hidden activations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start extracting class-conditional hidden activations.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "abffba3f586e4590a4d39c27fafd3a87",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=1000), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "merge activations. this takes time if save_outputs==True\n",
      "outputs/vgg16_bn/features_12_activations.pth is saved\n",
      "outputs/vgg16_bn/features_16_activations.pth is saved\n",
      "outputs/vgg16_bn/features_19_activations.pth is saved\n",
      "outputs/vgg16_bn/features_22_activations.pth is saved\n",
      "outputs/vgg16_bn/features_26_activations.pth is saved\n",
      "outputs/vgg16_bn/features_29_activations.pth is saved\n",
      "outputs/vgg16_bn/features_32_activations.pth is saved\n",
      "outputs/vgg16_bn/features_36_activations.pth is saved\n",
      "outputs/vgg16_bn/features_39_activations.pth is saved\n",
      "outputs/vgg16_bn/features_42_activations.pth is saved\n",
      "outputs/vgg16_bn/classifier_1_activations.pth is saved\n",
      "outputs/vgg16_bn/classifier_4_activations.pth is saved\n"
     ]
    }
   ],
   "source": [
    "print('start extracting class-conditional hidden activations.')\n",
    "loader = iter(dloader)\n",
    "for i in tqdm_notebook(range(num_class)):\n",
    "    \n",
    "    class_conditional_activations = dict()\n",
    "    for name in activations:\n",
    "        class_conditional_activations[name] = 0.\n",
    "    \n",
    "    for k in range(iters_per_class):\n",
    "        img, label = loader.next()\n",
    "        img = img.cuda()\n",
    "        net(img)\n",
    "\n",
    "        for name in activations:\n",
    "            meta_arch_name, layer_idx = name.split('_')\n",
    "            layer_idx = int(layer_idx)\n",
    "            output_name = meta_arch_name+'_'+str(layer_idx)\n",
    "            class_conditional_activations[name] += hidden_outputs[output_name].view(batch_size, -1).sum(0).div(imgs_per_class).data.cpu()\n",
    "            \n",
    "    for name in activations:\n",
    "        activations[name].append(class_conditional_activations[name])\n",
    "        \n",
    "print('merge activations. this takes time if save_outputs==True')\n",
    "for name in activations:\n",
    "    activations[name] = torch.stack(activations[name], dim=0)\n",
    "    \n",
    "    if save_outputs:\n",
    "        activations_path = os.path.join(output_dir, name+'_activations.pth')\n",
    "        with open(activations_path, 'wb+') as f:\n",
    "            torch.save(activations[name], activations_path)\n",
    "        print('{} is saved'.format(activations_path))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculate Class Selectivity Index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start calculating class selectivity index\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a5979a015bda4e16b5fdcdc68ca811be",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=1605632), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/features_12_selectivity_results.pth is saved\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fccd06c4e8c24b479d3eb893b3e9ee48",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=802816), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/features_16_selectivity_results.pth is saved\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a2cb95d7e3154576a053c22a5dbd4f91",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=802816), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/features_19_selectivity_results.pth is saved\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "35664918616c4fd8b79becaf14fe74cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=802816), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/features_22_selectivity_results.pth is saved\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0534b4338f2441a5a44cb9ce830d19fc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=401408), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/features_26_selectivity_results.pth is saved\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c9a200193f2948b9ac4ccb8f89a69690",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=401408), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/features_29_selectivity_results.pth is saved\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7c5085ce4b664e1d93213637d458d7cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=401408), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/features_32_selectivity_results.pth is saved\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "25bbb7622f144c52a8e5830498282564",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=100352), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/features_36_selectivity_results.pth is saved\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "83c22f506c3749579c4e9f950533a6b5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=100352), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/features_39_selectivity_results.pth is saved\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bb604f46cad74d289d0b358d087aa538",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=100352), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/features_42_selectivity_results.pth is saved\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8e15fd3328b64cce8d90e18a7e63f4d5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=4096), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/classifier_1_selectivity_results.pth is saved\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7e9d055f79bb489dbc3c66fd84db7f1d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=4096), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "outputs/vgg16_bn/classifier_4_selectivity_results.pth is saved\n"
     ]
    }
   ],
   "source": [
    "print('start calculating class selectivity index')\n",
    "result = {}\n",
    "for name in activations:\n",
    "    layer_act = activations[name]\n",
    "\n",
    "    num_class, num_neuron = layer_act.size()\n",
    "    dead_neuron_class = torch.tensor(num_class)\n",
    "    dead_neuron_confidence = torch.tensor(0.)\n",
    "\n",
    "    selected_class = []\n",
    "    selectivity_index = []\n",
    "    for neuron_idx in tqdm_notebook(range(num_neuron)):\n",
    "        neuron_act = layer_act[:, neuron_idx]\n",
    "\n",
    "        # In the case of mean activations of a neuron are all zero across whole classes\n",
    "        # Simply consider that neuron as dead neuron.\n",
    "        if neuron_act.nonzero().size(0) == 0: \n",
    "            class_selected = dead_neuron_class\n",
    "            class_confidence = dead_neuron_confidence\n",
    "        else:\n",
    "            class_selected = neuron_act.argmax()\n",
    "            mu_max = neuron_act[class_selected]\n",
    "            mu_mmax = (neuron_act.sum()-mu_max).div(num_class-1)\n",
    "            class_confidence = (mu_max-mu_mmax).div(mu_max+mu_mmax)\n",
    "\n",
    "        selected_class.append(class_selected)   \n",
    "        selectivity_index.append(class_confidence)\n",
    "\n",
    "    selected_class = torch.stack(selected_class, 0)\n",
    "    selectivity_index = torch.stack(selectivity_index, 0)\n",
    "    \n",
    "    result[name] = dict()\n",
    "    result[name]['selected_class'] = selected_class\n",
    "    result[name]['selectivity_index'] = selectivity_index\n",
    "    \n",
    "    if save_outputs:\n",
    "        result_path = os.path.join(output_dir, name+'_selectivity_results.pth')\n",
    "        with open(result_path, 'wb+') as f:\n",
    "            torch.save(result[name], result_path)\n",
    "        print('{} is saved'.format(result_path))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot class selectivity distributions as a function of depth"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/wonkonge/anaconda3/envs/kongda_py36_torch40/lib/python3.6/site-packages/ipykernel_launcher.py:7: MatplotlibDeprecationWarning: The set_color_cycle function was deprecated in version 1.5. Use `.set_prop_cycle` instead.\n",
      "  import sys\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "num_plots = len(result)\n",
    "colormap = plt.cm.jet\n",
    "plt.figure(figsize=(10,10))\n",
    "plt.gca().set_color_cycle([colormap(i) for i in np.linspace(0, 0.9, num_plots)]);\n",
    "\n",
    "labels = []\n",
    "for name in result:\n",
    "    selectivity_index = result[name]['selectivity_index']\n",
    "    selectivity_index_hist = np.histogram(selectivity_index*100, bins=100, normed=True)\n",
    "    x = np.arange(len(selectivity_index_hist[0]))/len(selectivity_index_hist[0])\n",
    "    y = selectivity_index_hist[0]\n",
    "    plt.fill_between(x, y, step=\"pre\", alpha=0.6)\n",
    "    plt.plot(x, y)\n",
    "    labels.append(name)\n",
    "\n",
    "plt.legend(labels, ncol=1, loc='upper right', \n",
    "           columnspacing=2.0, labelspacing=1,\n",
    "           handletextpad=0.5, handlelength=1.5,\n",
    "           fancybox=True, shadow=True)\n",
    "plt.ylabel('PDF', fontsize=15, labelpad=15)\n",
    "plt.xlabel('Selectivity Index', fontsize=15, labelpad=15)\n",
    "\n",
    "figname = os.path.join(output_dir, 'histogram.png')\n",
    "plt.savefig(figname)\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}