pyTorch-models-Accuracies-WeightedAlphas.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  pretrained models in PyTorch\n",
    "\n",
    "### Generate data for Paper; does not use WeightWatcher tool here\n",
    "\n",
    "This code is the basis for the tool.\n",
    "\n",
    "\n",
    "This notebook only has the default pyTorch models\n",
    "\n",
    "More models are available in the osmr and/or cadene repos\n",
    "\n",
    "https://github.com/Cadene/pretrained-models.pytorch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:29:11.918002Z",
     "start_time": "2018-10-22T21:29:06.826556Z"
    }
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "import torchvision.models as models\n",
    "import torchvision.transforms as transforms\n",
    "import torch.nn as nn\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "import numpy as np\n",
    "import powerlaw\n",
    "\n",
    "from sklearn.decomposition import TruncatedSVD\n",
    " \n",
    "\n",
    "from tqdm import tqdm_notebook as tqdm\n",
    "\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Models in Torch\n",
    "\n",
    "Accuracies file is taken from the Cadene website\n",
    "\n",
    "https://github.com/Cadene/pretrained-models.pytorch\n",
    "\n",
    "Although I discovered later that this package does not work consistantly so I just use the prebuilt pytorch models here\n",
    "\n",
    "### <font color='red'> Check this file to ensure the numbers are correct...I am sure the trends are correct"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:29:21.548282Z",
     "start_time": "2018-10-22T21:29:21.506963Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "62"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracies = pd.read_csv('accuracies', delimiter=\"\\t\")\n",
    "len(accuracies)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:29:22.194786Z",
     "start_time": "2018-10-22T21:29:22.173139Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Tensorflow', 'ourporting', 'Caffe', 'Torch7', 'Keras', 'Pytorch'],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracies['platform'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:29:22.635859Z",
     "start_time": "2018-10-22T21:29:22.584119Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(24,                   model    platform    acc1    acc5\n",
       " 1       PNASNet-5-Large  ourporting  82.736  95.992\n",
       " 3        NASNet-A-Large  ourporting  82.566  96.086\n",
       " 5              SENet154  ourporting  81.304  95.498\n",
       " 7               PolyNet  ourporting  81.002  95.624\n",
       " 10  SE-ResNeXt101_32x4d  ourporting  80.236  95.028\n",
       " 12    InceptionResNetV2  ourporting  80.170  95.234\n",
       " 13          InceptionV4  ourporting  80.062  94.926\n",
       " 14    DualPathNet107_5k  ourporting  79.746  94.684\n",
       " 16       DualPathNet131  ourporting  79.432  94.574\n",
       " 17     DualPathNet92_5k  ourporting  79.400  94.620\n",
       " 18        DualPathNet98  ourporting  79.224  94.488\n",
       " 19   SE-ResNeXt50_32x4d  ourporting  79.076  94.434\n",
       " 22     ResNeXt101_64x4d  ourporting  78.956  94.252\n",
       " 23             Xception  ourporting  78.888  94.292\n",
       " 26         SE-ResNet152  ourporting  78.658  94.374\n",
       " 28         SE-ResNet101  ourporting  78.396  94.258\n",
       " 30     ResNeXt101_32x4d  ourporting  78.188  93.886\n",
       " 33          SE-ResNet50  ourporting  77.636  93.752\n",
       " 36          FBResNet152  ourporting  77.386  93.594\n",
       " 39    DualPathNet68b_5k  ourporting  77.034  93.590\n",
       " 41       CaffeResnet101  ourporting  76.200  92.766\n",
       " 44        DualPathNet68  ourporting  75.868  92.774\n",
       " 48      NASNet-A-Mobile  ourporting  74.080  91.740\n",
       " 50          BNInception  ourporting  73.522  91.560)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "our_acc = accuracies[accuracies['platform']=='ourporting']\n",
    "len(our_acc), our_acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:29:23.152825Z",
     "start_time": "2018-10-22T21:29:23.109944Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(21,             model platform    acc1    acc5\n",
       " 27      ResNet152  Pytorch  78.428  94.110\n",
       " 34    DenseNet161  Pytorch  77.560  93.798\n",
       " 35      ResNet101  Pytorch  77.438  93.672\n",
       " 37    InceptionV3  Pytorch  77.294  93.454\n",
       " 38    DenseNet201  Pytorch  77.152  93.548\n",
       " 42    DenseNet169  Pytorch  76.026  92.992\n",
       " 43       ResNet50  Pytorch  76.002  92.980\n",
       " 45    DenseNet121  Pytorch  74.646  92.136\n",
       " 46       VGG19_BN  Pytorch  74.266  92.066\n",
       " 49       ResNet34  Pytorch  73.554  91.456\n",
       " 51       VGG16_BN  Pytorch  73.518  91.608\n",
       " 52          VGG19  Pytorch  72.080  90.822\n",
       " 53          VGG16  Pytorch  71.636  90.354\n",
       " 54       VGG13_BN  Pytorch  71.508  90.494\n",
       " 55       VGG11_BN  Pytorch  70.452  89.818\n",
       " 56       ResNet18  Pytorch  70.142  89.274\n",
       " 57          VGG13  Pytorch  69.662  89.264\n",
       " 58          VGG11  Pytorch  68.970  88.746\n",
       " 59  SqueezeNet1_1  Pytorch  58.250  80.800\n",
       " 60  SqueezeNet1_0  Pytorch  58.108  80.428\n",
       " 61        Alexnet  Pytorch  56.432  79.194)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "py_acc = accuracies[accuracies['platform']=='Pytorch']\n",
    "len(py_acc), py_acc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- merge dataframes\n",
    "- get model names\n",
    "- load pretrained\n",
    "- compute lognorm\n",
    "- store in data frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:29:24.126421Z",
     "start_time": "2018-10-22T21:29:24.087431Z"
    }
   },
   "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>model</th>\n",
       "      <th>platform</th>\n",
       "      <th>acc1</th>\n",
       "      <th>acc5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>ResNet152</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>78.428</td>\n",
       "      <td>94.110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>DenseNet161</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.560</td>\n",
       "      <td>93.798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>ResNet101</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.438</td>\n",
       "      <td>93.672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>InceptionV3</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.294</td>\n",
       "      <td>93.454</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>DenseNet201</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.152</td>\n",
       "      <td>93.548</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          model platform    acc1    acc5\n",
       "27    ResNet152  Pytorch  78.428  94.110\n",
       "34  DenseNet161  Pytorch  77.560  93.798\n",
       "35    ResNet101  Pytorch  77.438  93.672\n",
       "37  InceptionV3  Pytorch  77.294  93.454\n",
       "38  DenseNet201  Pytorch  77.152  93.548"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df =  pd.concat([py_acc, our_acc])\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Minimum dimension for W"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:29:25.271404Z",
     "start_time": "2018-10-22T21:29:25.262224Z"
    }
   },
   "outputs": [],
   "source": [
    "MIN_DIM = 50"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### All Linear Models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_conv2D_Wmats(Wtensor):\n",
    "    \"\"\"Extract W slices from a 4 index conv2D tensor of shape: (N,M,i,j) or (M,N,i,j).  Return ij (N x M) matrices\"\"\"\n",
    "    Wmats = []\n",
    "    s = Wtensor.shape\n",
    "    N, M, imax, jmax = s[0],s[1],s[2],s[3]\n",
    "    #print(\"tensor shape\", N,M,imax,jmax)\n",
    "    for i in range(imax):\n",
    "        for j in range(jmax):\n",
    "            W = Wtensor[:,:,i,j]\n",
    "            if N < M:\n",
    "                W = W.T\n",
    "            Wmats.append(W)\n",
    "    return Wmats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:29:26.344742Z",
     "start_time": "2018-10-22T21:29:26.314026Z"
    }
   },
   "outputs": [],
   "source": [
    "def count_layers(model):\n",
    "    \"\"\"Loops over a pytorch model and counts all the weight matrices, linear, conv2d, and conv2d-slices\"\"\"\n",
    "    num_linear, num_conv2d, num_conv2d_slices = 0, 0, 0\n",
    "    \n",
    "    for im, m in enumerate(model.modules()):\n",
    "        #print(im,m)\n",
    "        if isinstance(m, nn.Linear):    \n",
    "            #print(\"linear\")\n",
    "            W = np.array(m.weight.data.clone().cpu())\n",
    "            if W is not None:\n",
    "                #print(im,m, W.shape)\n",
    "                num_linear += 1\n",
    "        elif isinstance(m, nn.Conv2d):\n",
    "            #print(\"conv2d\")\n",
    "            Wtensor = np.array(m.weight.data.clone().cpu())\n",
    "            #print(Wtensor.shape)\n",
    "            Wmats = get_conv2D_Wmats(Wtensor)\n",
    "            #print(im,m)\n",
    "            num_conv2d += 1\n",
    "            for W in Wmats:\n",
    "                #print(W.shape)\n",
    "                if (np.min(W.shape) >= MIN_DIM):\n",
    "                    num_conv2d_slices += 1\n",
    "                \n",
    "    return num_linear, num_conv2d, num_conv2d_slices\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:29:27.441606Z",
     "start_time": "2018-10-22T21:29:27.411156Z"
    }
   },
   "outputs": [],
   "source": [
    "def iter_pytorch_layers(model, min_dim=MIN_DIM):\n",
    "    \"\"\"Iterate over a pytorch layers and  weight matrix.  \n",
    "    Supports linear, conv2d, and conv2d-slices\"\"\"\n",
    "    for im, m in enumerate(model.modules()):\n",
    "        if isinstance(m, nn.Linear):     \n",
    "            W = np.array(m.weight.data.clone().cpu())\n",
    "            if W is not None:\n",
    "                yield W\n",
    "        elif isinstance(m, nn.Conv2d):\n",
    "            Wtensor = np.array(m.weight.data.clone().cpu())\n",
    "            Wmats = get_conv2D_Wmats(Wtensor)\n",
    "            for W in Wmats:\n",
    "                if (np.min(W.shape) >= min_dim):\n",
    "                    yield W"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:30:00.199636Z",
     "start_time": "2018-10-22T21:30:00.179436Z"
    }
   },
   "outputs": [],
   "source": [
    "def calc_model_weighted_alphas(model):\n",
    "    try:\n",
    "        w_alphas = []\n",
    "\n",
    "        for W in tqdm(iter_pytorch_layers(model)):\n",
    "            M, N = np.min(W.shape), np.max(W.shape)\n",
    "            \n",
    "            # we do NOT normalize by N here...\n",
    "            # probably should uses tructaed ZSVD for speed\n",
    "            #X=np.dot(W.T,W)#/N\n",
    "            #evals = np.linalg.eigvals(X)\n",
    "            \n",
    "            # orders of magnitude faster, even if we are missing very small 1 SV\n",
    "            svd = TruncatedSVD(n_components=M-1, n_iter=7, random_state=42)\n",
    "            svd.fit(W) \n",
    "            sv = svd.singular_values_\n",
    "            evals = sv*sv\n",
    "\n",
    "            l_max, l_min = np.max(evals), np.min(evals)\n",
    "        \n",
    "            fit = powerlaw.Fit(evals, xmax=l_max, verbose=False)\n",
    "            w_alphas.append(fit.alpha*np.log10(l_max))\n",
    "            \n",
    "\n",
    "    except Exception as e: \n",
    "\n",
    "        print(\"skipping \", model_name)\n",
    "        print(e)\n",
    "\n",
    "        pass\n",
    "    \n",
    "    return w_alphas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:32:36.320118Z",
     "start_time": "2018-10-22T21:30:02.517846Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "vgg11\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "11d381ebcbc141e3a7de986c22ab3b53",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/powerlaw.py:700: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  (Theoretical_CDF * (1 - Theoretical_CDF))\n",
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/powerlaw.py:700: RuntimeWarning: invalid value encountered in true_divide\n",
      "  (Theoretical_CDF * (1 - Theoretical_CDF))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "vgg11_bn\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5289669f7f4741c7a28722a4cd480473",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "vgg13\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7afe2f1cc9ff46a1865c00faf3705858",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "vgg13_bn\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e891d4f0fd74468daddf73fedde9d61e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "vgg16\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bc41206af8a34655b2db32785bd60fcf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "vgg16_bn\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f3600e847638452088efb8d7da0e20ab",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "vgg19\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "07ffbc4247524ffc98b37221b22db1ef",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "vgg19_bn\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b5983c4cad284b48b4fc25f7b21e1c6f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "squeezenet1_0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/torchvision/models/squeezenet.py:94: UserWarning: nn.init.kaiming_uniform is now deprecated in favor of nn.init.kaiming_uniform_.\n",
      "  init.kaiming_uniform(m.weight.data)\n",
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/torchvision/models/squeezenet.py:92: UserWarning: nn.init.normal is now deprecated in favor of nn.init.normal_.\n",
      "  init.normal(m.weight.data, mean=0.0, std=0.01)\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ecfba8a0a64c42859975eff1255342a7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "squeezenet1_1\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "39e8b959120a47839f6aa8c9049b3033",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/powerlaw.py:1178: RuntimeWarning: overflow encountered in double_scalars\n",
      "  return (self.alpha-1) * self.xmin**(self.alpha-1)\n",
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/powerlaw.py:825: RuntimeWarning: invalid value encountered in multiply\n",
      "  likelihoods = f*C\n",
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/scipy/optimize/optimize.py:563: RuntimeWarning: invalid value encountered in subtract\n",
      "  numpy.max(numpy.abs(fsim[0] - fsim[1:])) <= fatol):\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "densenet121\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/torchvision/models/densenet.py:212: UserWarning: nn.init.kaiming_normal is now deprecated in favor of nn.init.kaiming_normal_.\n",
      "  nn.init.kaiming_normal(m.weight.data)\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8c3ff79d8388426a8a6f7abf0147fd0e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "densenet161\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1722bf7ac9834e019659fe726aa6703c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "densenet169\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7b8d042288f7468b87f8a3c253d351e7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "densenet201\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cc1adc8bc50645e4927fef65ce9d283c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "resnet18\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "377dfd9b092644a8a4e11f12549f35f5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "resnet34\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8d92f5791c9a444f96771a876088b96d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "resnet50\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "70d2fe50d7a84717a7971f1dbef2be81",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "resnet101\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a19e5ca9b57845b9a639012e1c7c410b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "resnet152\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9f96b92d29d740ea938340b68b7b03e5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "w_alphas_4model = {}\n",
    "counts_4model = {}\n",
    "#models = [m.lower().replace('-','') for m in df['model'].values]\n",
    "model_names = []\n",
    "\n",
    "def update_model(model_name, model):\n",
    "    print(model_name)\n",
    "    w_alphas_4model[model_name] = calc_model_weighted_alphas(model)\n",
    "    counts_4model[model_name] = count_layers(model)\n",
    "    model_names.append(model_name)\n",
    "    \n",
    "    \n",
    "try:\n",
    "    model_name = 'vgg11'\n",
    "    model = models.vgg11(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "   \n",
    "    model_name = 'vgg11_bn'\n",
    "    model = models.vgg11_bn(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "    model_name = 'vgg13'\n",
    "    model = models.vgg13(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "   \n",
    "    model_name = 'vgg13_bn'\n",
    "    model = models.vgg13_bn(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "    model_name = 'vgg16'\n",
    "    model = models.vgg16(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "   \n",
    "    model_name = 'vgg16_bn'\n",
    "    model = models.vgg16_bn(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "    model_name = 'vgg19'\n",
    "    model = models.vgg19(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "   \n",
    "    model_name = 'vgg19_bn'\n",
    "    model = models.vgg19_bn(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "\n",
    "    model_name = 'squeezenet1_0'\n",
    "    model = models.squeezenet1_0(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "    model_name = 'squeezenet1_1'\n",
    "    model = models.squeezenet1_1(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "    model_name = 'densenet121'\n",
    "    model = models.densenet121(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "    model_name = 'densenet161'\n",
    "    model = models.densenet161(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "    model_name = 'densenet169'\n",
    "    model = models.densenet169(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "    model_name = 'densenet201'\n",
    "    model = models.densenet201(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "    model_name = 'resnet18'\n",
    "    model = models.resnet18(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "    model_name = 'resnet34'\n",
    "    model = models.resnet34(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "        \n",
    "    model_name = 'resnet50'\n",
    "    model = models.resnet50(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "\n",
    "    model_name = 'resnet101'\n",
    "    model = models.resnet101(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "\n",
    "    model_name = 'resnet152'\n",
    "    model = models.resnet152(pretrained=True)\n",
    "    update_model(model_name, model)\n",
    "    \n",
    "    \n",
    "except Exception as e: \n",
    "\n",
    "    print(\"skipping \", model_name)\n",
    "    print(e)\n",
    "    pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:32:46.115759Z",
     "start_time": "2018-10-22T21:32:46.076314Z"
    }
   },
   "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>model</th>\n",
       "      <th>platform</th>\n",
       "      <th>acc1</th>\n",
       "      <th>acc5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>ResNet152</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>78.428</td>\n",
       "      <td>94.110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>DenseNet161</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.560</td>\n",
       "      <td>93.798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>ResNet101</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.438</td>\n",
       "      <td>93.672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>InceptionV3</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.294</td>\n",
       "      <td>93.454</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>DenseNet201</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.152</td>\n",
       "      <td>93.548</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          model platform    acc1    acc5\n",
       "27    ResNet152  Pytorch  78.428  94.110\n",
       "34  DenseNet161  Pytorch  77.560  93.798\n",
       "35    ResNet101  Pytorch  77.438  93.672\n",
       "37  InceptionV3  Pytorch  77.294  93.454\n",
       "38  DenseNet201  Pytorch  77.152  93.548"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:32:48.706396Z",
     "start_time": "2018-10-22T21:32:48.624300Z"
    }
   },
   "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>model</th>\n",
       "      <th>platform</th>\n",
       "      <th>acc1</th>\n",
       "      <th>acc5</th>\n",
       "      <th>modelname</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>ResNet152</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>78.428</td>\n",
       "      <td>94.110</td>\n",
       "      <td>resnet152</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>DenseNet161</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.560</td>\n",
       "      <td>93.798</td>\n",
       "      <td>densenet161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>ResNet101</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.438</td>\n",
       "      <td>93.672</td>\n",
       "      <td>resnet101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>InceptionV3</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.294</td>\n",
       "      <td>93.454</td>\n",
       "      <td>inceptionv3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>DenseNet201</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.152</td>\n",
       "      <td>93.548</td>\n",
       "      <td>densenet201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>DenseNet169</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>76.026</td>\n",
       "      <td>92.992</td>\n",
       "      <td>densenet169</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>ResNet50</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>76.002</td>\n",
       "      <td>92.980</td>\n",
       "      <td>resnet50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>DenseNet121</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>74.646</td>\n",
       "      <td>92.136</td>\n",
       "      <td>densenet121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>VGG19_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>74.266</td>\n",
       "      <td>92.066</td>\n",
       "      <td>vgg19_bn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>ResNet34</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>73.554</td>\n",
       "      <td>91.456</td>\n",
       "      <td>resnet34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>VGG16_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>73.518</td>\n",
       "      <td>91.608</td>\n",
       "      <td>vgg16_bn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>VGG19</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>72.080</td>\n",
       "      <td>90.822</td>\n",
       "      <td>vgg19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>71.636</td>\n",
       "      <td>90.354</td>\n",
       "      <td>vgg16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>VGG13_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>71.508</td>\n",
       "      <td>90.494</td>\n",
       "      <td>vgg13_bn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>VGG11_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>70.452</td>\n",
       "      <td>89.818</td>\n",
       "      <td>vgg11_bn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>ResNet18</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>70.142</td>\n",
       "      <td>89.274</td>\n",
       "      <td>resnet18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>VGG13</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>69.662</td>\n",
       "      <td>89.264</td>\n",
       "      <td>vgg13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>VGG11</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>68.970</td>\n",
       "      <td>88.746</td>\n",
       "      <td>vgg11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>SqueezeNet1_1</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>58.250</td>\n",
       "      <td>80.800</td>\n",
       "      <td>squeezenet1_1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>SqueezeNet1_0</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>58.108</td>\n",
       "      <td>80.428</td>\n",
       "      <td>squeezenet1_0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>Alexnet</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>56.432</td>\n",
       "      <td>79.194</td>\n",
       "      <td>alexnet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PNASNet-5-Large</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>82.736</td>\n",
       "      <td>95.992</td>\n",
       "      <td>pnasnet5large</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NASNet-A-Large</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>82.566</td>\n",
       "      <td>96.086</td>\n",
       "      <td>nasnetalarge</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>SENet154</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>81.304</td>\n",
       "      <td>95.498</td>\n",
       "      <td>senet154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>PolyNet</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>81.002</td>\n",
       "      <td>95.624</td>\n",
       "      <td>polynet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>SE-ResNeXt101_32x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>80.236</td>\n",
       "      <td>95.028</td>\n",
       "      <td>seresnext101_32x4d</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>InceptionResNetV2</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>80.170</td>\n",
       "      <td>95.234</td>\n",
       "      <td>inceptionresnetv2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>InceptionV4</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>80.062</td>\n",
       "      <td>94.926</td>\n",
       "      <td>inceptionv4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>DualPathNet107_5k</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.746</td>\n",
       "      <td>94.684</td>\n",
       "      <td>dualpathnet107_5k</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>DualPathNet131</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.432</td>\n",
       "      <td>94.574</td>\n",
       "      <td>dualpathnet131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>DualPathNet92_5k</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.400</td>\n",
       "      <td>94.620</td>\n",
       "      <td>dualpathnet92_5k</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>DualPathNet98</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.224</td>\n",
       "      <td>94.488</td>\n",
       "      <td>dualpathnet98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>SE-ResNeXt50_32x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.076</td>\n",
       "      <td>94.434</td>\n",
       "      <td>seresnext50_32x4d</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>ResNeXt101_64x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.956</td>\n",
       "      <td>94.252</td>\n",
       "      <td>resnext101_64x4d</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Xception</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.888</td>\n",
       "      <td>94.292</td>\n",
       "      <td>xception</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>SE-ResNet152</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.658</td>\n",
       "      <td>94.374</td>\n",
       "      <td>seresnet152</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>SE-ResNet101</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.396</td>\n",
       "      <td>94.258</td>\n",
       "      <td>seresnet101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>ResNeXt101_32x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.188</td>\n",
       "      <td>93.886</td>\n",
       "      <td>resnext101_32x4d</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>SE-ResNet50</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>77.636</td>\n",
       "      <td>93.752</td>\n",
       "      <td>seresnet50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>FBResNet152</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>77.386</td>\n",
       "      <td>93.594</td>\n",
       "      <td>fbresnet152</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>DualPathNet68b_5k</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>77.034</td>\n",
       "      <td>93.590</td>\n",
       "      <td>dualpathnet68b_5k</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>CaffeResnet101</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>76.200</td>\n",
       "      <td>92.766</td>\n",
       "      <td>cafferesnet101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>DualPathNet68</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>75.868</td>\n",
       "      <td>92.774</td>\n",
       "      <td>dualpathnet68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>NASNet-A-Mobile</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>74.080</td>\n",
       "      <td>91.740</td>\n",
       "      <td>nasnetamobile</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>BNInception</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>73.522</td>\n",
       "      <td>91.560</td>\n",
       "      <td>bninception</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  model    platform    acc1    acc5           modelname\n",
       "27            ResNet152     Pytorch  78.428  94.110           resnet152\n",
       "34          DenseNet161     Pytorch  77.560  93.798         densenet161\n",
       "35            ResNet101     Pytorch  77.438  93.672           resnet101\n",
       "37          InceptionV3     Pytorch  77.294  93.454         inceptionv3\n",
       "38          DenseNet201     Pytorch  77.152  93.548         densenet201\n",
       "42          DenseNet169     Pytorch  76.026  92.992         densenet169\n",
       "43             ResNet50     Pytorch  76.002  92.980            resnet50\n",
       "45          DenseNet121     Pytorch  74.646  92.136         densenet121\n",
       "46             VGG19_BN     Pytorch  74.266  92.066            vgg19_bn\n",
       "49             ResNet34     Pytorch  73.554  91.456            resnet34\n",
       "51             VGG16_BN     Pytorch  73.518  91.608            vgg16_bn\n",
       "52                VGG19     Pytorch  72.080  90.822               vgg19\n",
       "53                VGG16     Pytorch  71.636  90.354               vgg16\n",
       "54             VGG13_BN     Pytorch  71.508  90.494            vgg13_bn\n",
       "55             VGG11_BN     Pytorch  70.452  89.818            vgg11_bn\n",
       "56             ResNet18     Pytorch  70.142  89.274            resnet18\n",
       "57                VGG13     Pytorch  69.662  89.264               vgg13\n",
       "58                VGG11     Pytorch  68.970  88.746               vgg11\n",
       "59        SqueezeNet1_1     Pytorch  58.250  80.800       squeezenet1_1\n",
       "60        SqueezeNet1_0     Pytorch  58.108  80.428       squeezenet1_0\n",
       "61              Alexnet     Pytorch  56.432  79.194             alexnet\n",
       "1       PNASNet-5-Large  ourporting  82.736  95.992       pnasnet5large\n",
       "3        NASNet-A-Large  ourporting  82.566  96.086        nasnetalarge\n",
       "5              SENet154  ourporting  81.304  95.498            senet154\n",
       "7               PolyNet  ourporting  81.002  95.624             polynet\n",
       "10  SE-ResNeXt101_32x4d  ourporting  80.236  95.028  seresnext101_32x4d\n",
       "12    InceptionResNetV2  ourporting  80.170  95.234   inceptionresnetv2\n",
       "13          InceptionV4  ourporting  80.062  94.926         inceptionv4\n",
       "14    DualPathNet107_5k  ourporting  79.746  94.684   dualpathnet107_5k\n",
       "16       DualPathNet131  ourporting  79.432  94.574      dualpathnet131\n",
       "17     DualPathNet92_5k  ourporting  79.400  94.620    dualpathnet92_5k\n",
       "18        DualPathNet98  ourporting  79.224  94.488       dualpathnet98\n",
       "19   SE-ResNeXt50_32x4d  ourporting  79.076  94.434   seresnext50_32x4d\n",
       "22     ResNeXt101_64x4d  ourporting  78.956  94.252    resnext101_64x4d\n",
       "23             Xception  ourporting  78.888  94.292            xception\n",
       "26         SE-ResNet152  ourporting  78.658  94.374         seresnet152\n",
       "28         SE-ResNet101  ourporting  78.396  94.258         seresnet101\n",
       "30     ResNeXt101_32x4d  ourporting  78.188  93.886    resnext101_32x4d\n",
       "33          SE-ResNet50  ourporting  77.636  93.752          seresnet50\n",
       "36          FBResNet152  ourporting  77.386  93.594         fbresnet152\n",
       "39    DualPathNet68b_5k  ourporting  77.034  93.590   dualpathnet68b_5k\n",
       "41       CaffeResnet101  ourporting  76.200  92.766      cafferesnet101\n",
       "44        DualPathNet68  ourporting  75.868  92.774       dualpathnet68\n",
       "48      NASNet-A-Mobile  ourporting  74.080  91.740       nasnetamobile\n",
       "50          BNInception  ourporting  73.522  91.560         bninception"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['modelname'] = df['model'].apply(lambda x: x.lower().replace('-',''))\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:32:52.793310Z",
     "start_time": "2018-10-22T21:32:52.769740Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "27             resnet152\n",
       "34           densenet161\n",
       "35             resnet101\n",
       "37           inceptionv3\n",
       "38           densenet201\n",
       "42           densenet169\n",
       "43              resnet50\n",
       "45           densenet121\n",
       "46              vgg19_bn\n",
       "49              resnet34\n",
       "51              vgg16_bn\n",
       "52                 vgg19\n",
       "53                 vgg16\n",
       "54              vgg13_bn\n",
       "55              vgg11_bn\n",
       "56              resnet18\n",
       "57                 vgg13\n",
       "58                 vgg11\n",
       "59         squeezenet1_1\n",
       "60         squeezenet1_0\n",
       "61               alexnet\n",
       "1          pnasnet5large\n",
       "3           nasnetalarge\n",
       "5               senet154\n",
       "7                polynet\n",
       "10    seresnext101_32x4d\n",
       "12     inceptionresnetv2\n",
       "13           inceptionv4\n",
       "14     dualpathnet107_5k\n",
       "16        dualpathnet131\n",
       "17      dualpathnet92_5k\n",
       "18         dualpathnet98\n",
       "19     seresnext50_32x4d\n",
       "22      resnext101_64x4d\n",
       "23              xception\n",
       "26           seresnet152\n",
       "28           seresnet101\n",
       "30      resnext101_32x4d\n",
       "33            seresnet50\n",
       "36           fbresnet152\n",
       "39     dualpathnet68b_5k\n",
       "41        cafferesnet101\n",
       "44         dualpathnet68\n",
       "48         nasnetamobile\n",
       "50           bninception\n",
       "Name: modelname, dtype: object"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['modelname']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:32:54.226821Z",
     "start_time": "2018-10-22T21:32:54.201505Z"
    }
   },
   "outputs": [],
   "source": [
    "avg_w_alphas = {}\n",
    "len_w_alphas = {}\n",
    "\n",
    "for model_name in df['modelname'].values:\n",
    "    avg_w_alphas[model_name] = 0.0\n",
    "    len_w_alphas[model_name] = 0.0\n",
    "    \n",
    "    if(model_name in w_alphas_4model):\n",
    "        avg_w_alphas[model_name] = np.average(w_alphas_4model[model_name])\n",
    "        len_w_alphas[model_name] = len(w_alphas_4model[model_name])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:33:28.433606Z",
     "start_time": "2018-10-22T21:33:28.410000Z"
    }
   },
   "outputs": [],
   "source": [
    "df['avg_w_alphas'] = df['modelname'].apply(lambda x: avg_w_alphas[x] )\n",
    "df['num_layers'] = df['modelname'].apply(lambda x: int(len_w_alphas[x]) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:33:30.176440Z",
     "start_time": "2018-10-22T21:33:30.058643Z"
    }
   },
   "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>model</th>\n",
       "      <th>platform</th>\n",
       "      <th>acc1</th>\n",
       "      <th>acc5</th>\n",
       "      <th>modelname</th>\n",
       "      <th>avg_w_alphas</th>\n",
       "      <th>num_layers</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>Alexnet</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>56.432</td>\n",
       "      <td>79.194</td>\n",
       "      <td>alexnet</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>BNInception</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>73.522</td>\n",
       "      <td>91.560</td>\n",
       "      <td>bninception</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>CaffeResnet101</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>76.200</td>\n",
       "      <td>92.766</td>\n",
       "      <td>cafferesnet101</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>DenseNet121</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>74.646</td>\n",
       "      <td>92.136</td>\n",
       "      <td>densenet121</td>\n",
       "      <td>1.248708</td>\n",
       "      <td>62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>DenseNet161</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.560</td>\n",
       "      <td>93.798</td>\n",
       "      <td>densenet161</td>\n",
       "      <td>0.838195</td>\n",
       "      <td>82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>DenseNet169</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>76.026</td>\n",
       "      <td>92.992</td>\n",
       "      <td>densenet169</td>\n",
       "      <td>0.678617</td>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>DenseNet201</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.152</td>\n",
       "      <td>93.548</td>\n",
       "      <td>densenet201</td>\n",
       "      <td>0.501668</td>\n",
       "      <td>102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>DualPathNet107_5k</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.746</td>\n",
       "      <td>94.684</td>\n",
       "      <td>dualpathnet107_5k</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>DualPathNet131</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.432</td>\n",
       "      <td>94.574</td>\n",
       "      <td>dualpathnet131</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>DualPathNet68</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>75.868</td>\n",
       "      <td>92.774</td>\n",
       "      <td>dualpathnet68</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>DualPathNet68b_5k</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>77.034</td>\n",
       "      <td>93.590</td>\n",
       "      <td>dualpathnet68b_5k</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>DualPathNet92_5k</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.400</td>\n",
       "      <td>94.620</td>\n",
       "      <td>dualpathnet92_5k</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>DualPathNet98</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.224</td>\n",
       "      <td>94.488</td>\n",
       "      <td>dualpathnet98</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>FBResNet152</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>77.386</td>\n",
       "      <td>93.594</td>\n",
       "      <td>fbresnet152</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>InceptionResNetV2</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>80.170</td>\n",
       "      <td>95.234</td>\n",
       "      <td>inceptionresnetv2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>InceptionV3</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.294</td>\n",
       "      <td>93.454</td>\n",
       "      <td>inceptionv3</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>InceptionV4</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>80.062</td>\n",
       "      <td>94.926</td>\n",
       "      <td>inceptionv4</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NASNet-A-Large</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>82.566</td>\n",
       "      <td>96.086</td>\n",
       "      <td>nasnetalarge</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>NASNet-A-Mobile</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>74.080</td>\n",
       "      <td>91.740</td>\n",
       "      <td>nasnetamobile</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PNASNet-5-Large</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>82.736</td>\n",
       "      <td>95.992</td>\n",
       "      <td>pnasnet5large</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>PolyNet</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>81.002</td>\n",
       "      <td>95.624</td>\n",
       "      <td>polynet</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>ResNeXt101_32x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.188</td>\n",
       "      <td>93.886</td>\n",
       "      <td>resnext101_32x4d</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>ResNeXt101_64x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.956</td>\n",
       "      <td>94.252</td>\n",
       "      <td>resnext101_64x4d</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>ResNet101</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.438</td>\n",
       "      <td>93.672</td>\n",
       "      <td>resnet101</td>\n",
       "      <td>-0.616471</td>\n",
       "      <td>368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>ResNet152</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>78.428</td>\n",
       "      <td>94.110</td>\n",
       "      <td>resnet152</td>\n",
       "      <td>-0.789808</td>\n",
       "      <td>555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>ResNet18</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>70.142</td>\n",
       "      <td>89.274</td>\n",
       "      <td>resnet18</td>\n",
       "      <td>0.610676</td>\n",
       "      <td>148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>ResNet34</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>73.554</td>\n",
       "      <td>91.456</td>\n",
       "      <td>resnet34</td>\n",
       "      <td>-0.393656</td>\n",
       "      <td>292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>ResNet50</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>76.002</td>\n",
       "      <td>92.980</td>\n",
       "      <td>resnet50</td>\n",
       "      <td>0.140391</td>\n",
       "      <td>181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>SE-ResNeXt101_32x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>80.236</td>\n",
       "      <td>95.028</td>\n",
       "      <td>seresnext101_32x4d</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>SE-ResNeXt50_32x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.076</td>\n",
       "      <td>94.434</td>\n",
       "      <td>seresnext50_32x4d</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>SE-ResNet101</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.396</td>\n",
       "      <td>94.258</td>\n",
       "      <td>seresnet101</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>SE-ResNet152</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.658</td>\n",
       "      <td>94.374</td>\n",
       "      <td>seresnet152</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>SE-ResNet50</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>77.636</td>\n",
       "      <td>93.752</td>\n",
       "      <td>seresnet50</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>SENet154</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>81.304</td>\n",
       "      <td>95.498</td>\n",
       "      <td>senet154</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>SqueezeNet1_0</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>58.108</td>\n",
       "      <td>80.428</td>\n",
       "      <td>squeezenet1_0</td>\n",
       "      <td>1.942516</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>SqueezeNet1_1</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>58.250</td>\n",
       "      <td>80.800</td>\n",
       "      <td>squeezenet1_1</td>\n",
       "      <td>1.560409</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>VGG11</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>68.970</td>\n",
       "      <td>88.746</td>\n",
       "      <td>vgg11</td>\n",
       "      <td>1.843995</td>\n",
       "      <td>66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>VGG11_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>70.452</td>\n",
       "      <td>89.818</td>\n",
       "      <td>vgg11_bn</td>\n",
       "      <td>1.597647</td>\n",
       "      <td>66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>VGG13</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>69.662</td>\n",
       "      <td>89.264</td>\n",
       "      <td>vgg13</td>\n",
       "      <td>1.645383</td>\n",
       "      <td>84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>VGG13_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>71.508</td>\n",
       "      <td>90.494</td>\n",
       "      <td>vgg13_bn</td>\n",
       "      <td>1.355124</td>\n",
       "      <td>84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>71.636</td>\n",
       "      <td>90.354</td>\n",
       "      <td>vgg16</td>\n",
       "      <td>1.405073</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>VGG16_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>73.518</td>\n",
       "      <td>91.608</td>\n",
       "      <td>vgg16_bn</td>\n",
       "      <td>1.082041</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>VGG19</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>72.080</td>\n",
       "      <td>90.822</td>\n",
       "      <td>vgg19</td>\n",
       "      <td>1.158888</td>\n",
       "      <td>138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>VGG19_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>74.266</td>\n",
       "      <td>92.066</td>\n",
       "      <td>vgg19_bn</td>\n",
       "      <td>0.811058</td>\n",
       "      <td>138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Xception</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.888</td>\n",
       "      <td>94.292</td>\n",
       "      <td>xception</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  model    platform    acc1    acc5           modelname  \\\n",
       "61              Alexnet     Pytorch  56.432  79.194             alexnet   \n",
       "50          BNInception  ourporting  73.522  91.560         bninception   \n",
       "41       CaffeResnet101  ourporting  76.200  92.766      cafferesnet101   \n",
       "45          DenseNet121     Pytorch  74.646  92.136         densenet121   \n",
       "34          DenseNet161     Pytorch  77.560  93.798         densenet161   \n",
       "42          DenseNet169     Pytorch  76.026  92.992         densenet169   \n",
       "38          DenseNet201     Pytorch  77.152  93.548         densenet201   \n",
       "14    DualPathNet107_5k  ourporting  79.746  94.684   dualpathnet107_5k   \n",
       "16       DualPathNet131  ourporting  79.432  94.574      dualpathnet131   \n",
       "44        DualPathNet68  ourporting  75.868  92.774       dualpathnet68   \n",
       "39    DualPathNet68b_5k  ourporting  77.034  93.590   dualpathnet68b_5k   \n",
       "17     DualPathNet92_5k  ourporting  79.400  94.620    dualpathnet92_5k   \n",
       "18        DualPathNet98  ourporting  79.224  94.488       dualpathnet98   \n",
       "36          FBResNet152  ourporting  77.386  93.594         fbresnet152   \n",
       "12    InceptionResNetV2  ourporting  80.170  95.234   inceptionresnetv2   \n",
       "37          InceptionV3     Pytorch  77.294  93.454         inceptionv3   \n",
       "13          InceptionV4  ourporting  80.062  94.926         inceptionv4   \n",
       "3        NASNet-A-Large  ourporting  82.566  96.086        nasnetalarge   \n",
       "48      NASNet-A-Mobile  ourporting  74.080  91.740       nasnetamobile   \n",
       "1       PNASNet-5-Large  ourporting  82.736  95.992       pnasnet5large   \n",
       "7               PolyNet  ourporting  81.002  95.624             polynet   \n",
       "30     ResNeXt101_32x4d  ourporting  78.188  93.886    resnext101_32x4d   \n",
       "22     ResNeXt101_64x4d  ourporting  78.956  94.252    resnext101_64x4d   \n",
       "35            ResNet101     Pytorch  77.438  93.672           resnet101   \n",
       "27            ResNet152     Pytorch  78.428  94.110           resnet152   \n",
       "56             ResNet18     Pytorch  70.142  89.274            resnet18   \n",
       "49             ResNet34     Pytorch  73.554  91.456            resnet34   \n",
       "43             ResNet50     Pytorch  76.002  92.980            resnet50   \n",
       "10  SE-ResNeXt101_32x4d  ourporting  80.236  95.028  seresnext101_32x4d   \n",
       "19   SE-ResNeXt50_32x4d  ourporting  79.076  94.434   seresnext50_32x4d   \n",
       "28         SE-ResNet101  ourporting  78.396  94.258         seresnet101   \n",
       "26         SE-ResNet152  ourporting  78.658  94.374         seresnet152   \n",
       "33          SE-ResNet50  ourporting  77.636  93.752          seresnet50   \n",
       "5              SENet154  ourporting  81.304  95.498            senet154   \n",
       "60        SqueezeNet1_0     Pytorch  58.108  80.428       squeezenet1_0   \n",
       "59        SqueezeNet1_1     Pytorch  58.250  80.800       squeezenet1_1   \n",
       "58                VGG11     Pytorch  68.970  88.746               vgg11   \n",
       "55             VGG11_BN     Pytorch  70.452  89.818            vgg11_bn   \n",
       "57                VGG13     Pytorch  69.662  89.264               vgg13   \n",
       "54             VGG13_BN     Pytorch  71.508  90.494            vgg13_bn   \n",
       "53                VGG16     Pytorch  71.636  90.354               vgg16   \n",
       "51             VGG16_BN     Pytorch  73.518  91.608            vgg16_bn   \n",
       "52                VGG19     Pytorch  72.080  90.822               vgg19   \n",
       "46             VGG19_BN     Pytorch  74.266  92.066            vgg19_bn   \n",
       "23             Xception  ourporting  78.888  94.292            xception   \n",
       "\n",
       "    avg_w_alphas  num_layers  \n",
       "61      0.000000           0  \n",
       "50      0.000000           0  \n",
       "41      0.000000           0  \n",
       "45      1.248708          62  \n",
       "34      0.838195          82  \n",
       "42      0.678617          86  \n",
       "38      0.501668         102  \n",
       "14      0.000000           0  \n",
       "16      0.000000           0  \n",
       "44      0.000000           0  \n",
       "39      0.000000           0  \n",
       "17      0.000000           0  \n",
       "18      0.000000           0  \n",
       "36      0.000000           0  \n",
       "12      0.000000           0  \n",
       "37      0.000000           0  \n",
       "13      0.000000           0  \n",
       "3       0.000000           0  \n",
       "48      0.000000           0  \n",
       "1       0.000000           0  \n",
       "7       0.000000           0  \n",
       "30      0.000000           0  \n",
       "22      0.000000           0  \n",
       "35     -0.616471         368  \n",
       "27     -0.789808         555  \n",
       "56      0.610676         148  \n",
       "49     -0.393656         292  \n",
       "43      0.140391         181  \n",
       "10      0.000000           0  \n",
       "19      0.000000           0  \n",
       "28      0.000000           0  \n",
       "26      0.000000           0  \n",
       "33      0.000000           0  \n",
       "5       0.000000           0  \n",
       "60      1.942516          23  \n",
       "59      1.560409          23  \n",
       "58      1.843995          66  \n",
       "55      1.597647          66  \n",
       "57      1.645383          84  \n",
       "54      1.355124          84  \n",
       "53      1.405073         111  \n",
       "51      1.082041         111  \n",
       "52      1.158888         138  \n",
       "46      0.811058         138  \n",
       "23      0.000000           0  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.sort_values('model')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:33:33.710273Z",
     "start_time": "2018-10-22T21:33:32.844105Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:2: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \n",
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:5: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \"\"\"\n",
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:8: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \n",
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:11: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  # This is added back by InteractiveShellApp.init_path()\n",
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:14: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \n",
      "/Users/charleshmartin/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:17: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n"
     ]
    }
   ],
   "source": [
    "df_vgg11 = df[(df['model']=='VGG11')| (df['model']=='VGG11_BN')]\n",
    "df_vgg11['legend']='VGG11'\n",
    "\n",
    "df_vgg13 = df[(df['model']=='VGG13')| (df['model']=='VGG13_BN')]\n",
    "df_vgg13['legend']='VGG13'\n",
    "\n",
    "df_vgg16 = df[(df['model']=='VGG16')| (df['model']=='VGG16_BN')]\n",
    "df_vgg16['legend']='VGG16'\n",
    "\n",
    "df_vgg19 = df[(df['model']=='VGG19')| (df['model']=='VGG19_BN')]\n",
    "df_vgg19['legend']='VGG19'\n",
    "\n",
    "df_squeezenet = df[(df['model']=='SqueezeNet1_0')| (df['model']=='SqueezeNet1_1')]\n",
    "df_squeezenet['legend']='SqueezeNet'\n",
    "\n",
    "df_resnxt = df[(df['model']=='ResNeXt101_32x4d')| (df['model']=='ResNeXt101_64x4d')]\n",
    "df_resnxt['legend']='ResNeXT101'\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.rcParams['figure.figsize'] = [10,10]\n",
    "matplotlib.rcParams.update({'font.size': 24})\n",
    "legend_prop={'size':20}\n",
    "marker_size = 75\n",
    "\n",
    "title = r\"Test Accuracy vs. Universal metric $\\hat{\\alpha}$\"+\"\\nPretrained \""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [],
   "source": [
    "from matplotlib.legend_handler import HandlerLineCollection, HandlerTuple"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:34:58.599857Z",
     "start_time": "2018-10-22T21:34:57.583864Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors=['blue', 'orange', 'green', 'red']\n",
    "\n",
    "plts = []\n",
    "labels = []\n",
    "for im, mf in enumerate([df_vgg11, df_vgg13, df_vgg16, df_vgg19]):\n",
    "#for mf in [df_vgg11, df_vgg13]:\n",
    "    x = mf['acc1'].values\n",
    "    y = mf['avg_w_alphas'].values\n",
    "    label = mf['legend'].values[0]\n",
    "    color = colors[im]\n",
    "    \n",
    "    p1 = plt.scatter(x[1],y[1], c=color, marker=\"o\", s=marker_size)\n",
    "    p2 = plt.scatter(x[0],y[0],  c=color,  marker=\"s\", s=marker_size)\n",
    "    \n",
    "    plts.append((p1,p2))\n",
    "    labels.append(label+\", BN\")\n",
    "\n",
    "plt.legend(plts, labels, scatterpoints=1,\n",
    "           numpoints=1, handler_map={tuple: HandlerTuple(ndivide=None)},prop=legend_prop)\n",
    "    \n",
    "plt.title(title+\"VGG and VGG_BN Models\")\n",
    "plt.xlabel(r\"top1 Test Accuracy\")\n",
    "plt.ylabel(r\"$\\hat{\\alpha}$\")\n",
    "plt.savefig(\"img/vgg-w_alphas.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ResNet\n",
    "\n",
    "Only exception is InceptionResNetv2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:35:06.389243Z",
     "start_time": "2018-10-22T21:35:05.094966Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[70.142] [0.61067626]\n",
      "[73.554] [-0.39365554]\n",
      "[76.002] [0.1403913]\n",
      "[77.438] [-0.61647061]\n",
      "[78.428] [-0.78980838]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "x, y = [], []\n",
    "\n",
    "modelnames = ['resnet18', 'resnet34', 'resnet50', 'resnet101', 'resnet152']\n",
    "\n",
    "for name in modelnames:\n",
    "    if 'resnet' in name:\n",
    "        mf = df[df.modelname==name]\n",
    "        x = mf['acc1'].values\n",
    "        y = mf['avg_w_alphas'].values\n",
    "        print(x,y)\n",
    "\n",
    "        if np.abs(y) > 0:\n",
    "            plt.scatter(x,y, label=name, s=marker_size)\n",
    "        \n",
    "plt.title(title+\"ResNet Models\")\n",
    "plt.xlabel(r\"top1 Test Accuracy\")\n",
    "plt.ylabel(r\"$\\hat{\\alpha}$\")\n",
    "plt.legend(prop=legend_prop)\n",
    "plt.savefig(\"img/resnet-pytorch-w_alphas.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:35:25.148462Z",
     "start_time": "2018-10-22T21:35:24.032914Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['figure.figsize'] = [8,8]\n",
    "\n",
    "x, y = [], []\n",
    "\n",
    "for name in df.modelname: \n",
    "    if 'densenet' in name and 'resnet' not in name:\n",
    "        #print(name)\n",
    "        mf = df[df.modelname==name]\n",
    "        x = mf['acc1'].values\n",
    "        y = mf['avg_w_alphas'].values\n",
    "        if np.abs(y) > 0:\n",
    "            plt.scatter(x,y, label=name, s=marker_size)\n",
    "        \n",
    "plt.title(title+\"DenseNet Models\")\n",
    "plt.xlabel(r\"top1 Test Accuracy\")\n",
    "plt.ylabel(r\"$\\hat{\\alpha}$\")\n",
    "plt.legend(prop=legend_prop)\n",
    "plt.savefig(\"img/densenet-pytorch-w_alphas.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ResNeXt, SqueezeNet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-10-22T21:35:31.084991Z",
     "start_time": "2018-10-22T21:35:29.394671Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "squeezenet1_0\n",
      "squeezenet1_1\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for mf in [df_squeezenet]:\n",
    "    for ix in [1,0]:\n",
    "        x = mf['acc1'].values[ix]\n",
    "        y = mf['avg_w_alphas'].values[ix]\n",
    "        label = mf.modelname.values[ix] # mf['legend'].values[ix]\n",
    "        print(label)\n",
    "        color =  'red' if ix==0  else 'orange' \n",
    "        plt.scatter(x,y,label=label, color=color,s=marker_size)\n",
    "        \n",
    "plt.legend(prop=legend_prop)\n",
    "plt.title(title+\"SqueezeNet Models\")\n",
    "plt.xlabel(r\"top1 Test Accuracy\")\n",
    "plt.ylabel(r\"$\\hat{\\alpha}$\")\n",
    "plt.savefig(\"img/squeezenet-pytorch-w_alphas.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "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>model</th>\n",
       "      <th>platform</th>\n",
       "      <th>acc1</th>\n",
       "      <th>acc5</th>\n",
       "      <th>modelname</th>\n",
       "      <th>avg_w_alphas</th>\n",
       "      <th>num_layers</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>ResNet152</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>78.428</td>\n",
       "      <td>94.110</td>\n",
       "      <td>resnet152</td>\n",
       "      <td>-0.789808</td>\n",
       "      <td>555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>DenseNet161</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.560</td>\n",
       "      <td>93.798</td>\n",
       "      <td>densenet161</td>\n",
       "      <td>0.838195</td>\n",
       "      <td>82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>ResNet101</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.438</td>\n",
       "      <td>93.672</td>\n",
       "      <td>resnet101</td>\n",
       "      <td>-0.616471</td>\n",
       "      <td>368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>InceptionV3</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.294</td>\n",
       "      <td>93.454</td>\n",
       "      <td>inceptionv3</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>DenseNet201</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>77.152</td>\n",
       "      <td>93.548</td>\n",
       "      <td>densenet201</td>\n",
       "      <td>0.501668</td>\n",
       "      <td>102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>DenseNet169</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>76.026</td>\n",
       "      <td>92.992</td>\n",
       "      <td>densenet169</td>\n",
       "      <td>0.678617</td>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>ResNet50</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>76.002</td>\n",
       "      <td>92.980</td>\n",
       "      <td>resnet50</td>\n",
       "      <td>0.140391</td>\n",
       "      <td>181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>DenseNet121</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>74.646</td>\n",
       "      <td>92.136</td>\n",
       "      <td>densenet121</td>\n",
       "      <td>1.248708</td>\n",
       "      <td>62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>VGG19_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>74.266</td>\n",
       "      <td>92.066</td>\n",
       "      <td>vgg19_bn</td>\n",
       "      <td>0.811058</td>\n",
       "      <td>138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>ResNet34</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>73.554</td>\n",
       "      <td>91.456</td>\n",
       "      <td>resnet34</td>\n",
       "      <td>-0.393656</td>\n",
       "      <td>292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>VGG16_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>73.518</td>\n",
       "      <td>91.608</td>\n",
       "      <td>vgg16_bn</td>\n",
       "      <td>1.082041</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>VGG19</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>72.080</td>\n",
       "      <td>90.822</td>\n",
       "      <td>vgg19</td>\n",
       "      <td>1.158888</td>\n",
       "      <td>138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>71.636</td>\n",
       "      <td>90.354</td>\n",
       "      <td>vgg16</td>\n",
       "      <td>1.405073</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>VGG13_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>71.508</td>\n",
       "      <td>90.494</td>\n",
       "      <td>vgg13_bn</td>\n",
       "      <td>1.355124</td>\n",
       "      <td>84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>VGG11_BN</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>70.452</td>\n",
       "      <td>89.818</td>\n",
       "      <td>vgg11_bn</td>\n",
       "      <td>1.597647</td>\n",
       "      <td>66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>ResNet18</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>70.142</td>\n",
       "      <td>89.274</td>\n",
       "      <td>resnet18</td>\n",
       "      <td>0.610676</td>\n",
       "      <td>148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>VGG13</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>69.662</td>\n",
       "      <td>89.264</td>\n",
       "      <td>vgg13</td>\n",
       "      <td>1.645383</td>\n",
       "      <td>84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>VGG11</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>68.970</td>\n",
       "      <td>88.746</td>\n",
       "      <td>vgg11</td>\n",
       "      <td>1.843995</td>\n",
       "      <td>66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>SqueezeNet1_1</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>58.250</td>\n",
       "      <td>80.800</td>\n",
       "      <td>squeezenet1_1</td>\n",
       "      <td>1.560409</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>SqueezeNet1_0</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>58.108</td>\n",
       "      <td>80.428</td>\n",
       "      <td>squeezenet1_0</td>\n",
       "      <td>1.942516</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>Alexnet</td>\n",
       "      <td>Pytorch</td>\n",
       "      <td>56.432</td>\n",
       "      <td>79.194</td>\n",
       "      <td>alexnet</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PNASNet-5-Large</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>82.736</td>\n",
       "      <td>95.992</td>\n",
       "      <td>pnasnet5large</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NASNet-A-Large</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>82.566</td>\n",
       "      <td>96.086</td>\n",
       "      <td>nasnetalarge</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>SENet154</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>81.304</td>\n",
       "      <td>95.498</td>\n",
       "      <td>senet154</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>PolyNet</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>81.002</td>\n",
       "      <td>95.624</td>\n",
       "      <td>polynet</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>SE-ResNeXt101_32x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>80.236</td>\n",
       "      <td>95.028</td>\n",
       "      <td>seresnext101_32x4d</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>InceptionResNetV2</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>80.170</td>\n",
       "      <td>95.234</td>\n",
       "      <td>inceptionresnetv2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>InceptionV4</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>80.062</td>\n",
       "      <td>94.926</td>\n",
       "      <td>inceptionv4</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>DualPathNet107_5k</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.746</td>\n",
       "      <td>94.684</td>\n",
       "      <td>dualpathnet107_5k</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>DualPathNet131</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.432</td>\n",
       "      <td>94.574</td>\n",
       "      <td>dualpathnet131</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>DualPathNet92_5k</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.400</td>\n",
       "      <td>94.620</td>\n",
       "      <td>dualpathnet92_5k</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>DualPathNet98</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.224</td>\n",
       "      <td>94.488</td>\n",
       "      <td>dualpathnet98</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>SE-ResNeXt50_32x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>79.076</td>\n",
       "      <td>94.434</td>\n",
       "      <td>seresnext50_32x4d</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>ResNeXt101_64x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.956</td>\n",
       "      <td>94.252</td>\n",
       "      <td>resnext101_64x4d</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Xception</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.888</td>\n",
       "      <td>94.292</td>\n",
       "      <td>xception</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>SE-ResNet152</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.658</td>\n",
       "      <td>94.374</td>\n",
       "      <td>seresnet152</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>SE-ResNet101</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.396</td>\n",
       "      <td>94.258</td>\n",
       "      <td>seresnet101</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>ResNeXt101_32x4d</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>78.188</td>\n",
       "      <td>93.886</td>\n",
       "      <td>resnext101_32x4d</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>SE-ResNet50</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>77.636</td>\n",
       "      <td>93.752</td>\n",
       "      <td>seresnet50</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>FBResNet152</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>77.386</td>\n",
       "      <td>93.594</td>\n",
       "      <td>fbresnet152</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>DualPathNet68b_5k</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>77.034</td>\n",
       "      <td>93.590</td>\n",
       "      <td>dualpathnet68b_5k</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>CaffeResnet101</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>76.200</td>\n",
       "      <td>92.766</td>\n",
       "      <td>cafferesnet101</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>DualPathNet68</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>75.868</td>\n",
       "      <td>92.774</td>\n",
       "      <td>dualpathnet68</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>NASNet-A-Mobile</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>74.080</td>\n",
       "      <td>91.740</td>\n",
       "      <td>nasnetamobile</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>BNInception</td>\n",
       "      <td>ourporting</td>\n",
       "      <td>73.522</td>\n",
       "      <td>91.560</td>\n",
       "      <td>bninception</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  model    platform    acc1    acc5           modelname  \\\n",
       "27            ResNet152     Pytorch  78.428  94.110           resnet152   \n",
       "34          DenseNet161     Pytorch  77.560  93.798         densenet161   \n",
       "35            ResNet101     Pytorch  77.438  93.672           resnet101   \n",
       "37          InceptionV3     Pytorch  77.294  93.454         inceptionv3   \n",
       "38          DenseNet201     Pytorch  77.152  93.548         densenet201   \n",
       "42          DenseNet169     Pytorch  76.026  92.992         densenet169   \n",
       "43             ResNet50     Pytorch  76.002  92.980            resnet50   \n",
       "45          DenseNet121     Pytorch  74.646  92.136         densenet121   \n",
       "46             VGG19_BN     Pytorch  74.266  92.066            vgg19_bn   \n",
       "49             ResNet34     Pytorch  73.554  91.456            resnet34   \n",
       "51             VGG16_BN     Pytorch  73.518  91.608            vgg16_bn   \n",
       "52                VGG19     Pytorch  72.080  90.822               vgg19   \n",
       "53                VGG16     Pytorch  71.636  90.354               vgg16   \n",
       "54             VGG13_BN     Pytorch  71.508  90.494            vgg13_bn   \n",
       "55             VGG11_BN     Pytorch  70.452  89.818            vgg11_bn   \n",
       "56             ResNet18     Pytorch  70.142  89.274            resnet18   \n",
       "57                VGG13     Pytorch  69.662  89.264               vgg13   \n",
       "58                VGG11     Pytorch  68.970  88.746               vgg11   \n",
       "59        SqueezeNet1_1     Pytorch  58.250  80.800       squeezenet1_1   \n",
       "60        SqueezeNet1_0     Pytorch  58.108  80.428       squeezenet1_0   \n",
       "61              Alexnet     Pytorch  56.432  79.194             alexnet   \n",
       "1       PNASNet-5-Large  ourporting  82.736  95.992       pnasnet5large   \n",
       "3        NASNet-A-Large  ourporting  82.566  96.086        nasnetalarge   \n",
       "5              SENet154  ourporting  81.304  95.498            senet154   \n",
       "7               PolyNet  ourporting  81.002  95.624             polynet   \n",
       "10  SE-ResNeXt101_32x4d  ourporting  80.236  95.028  seresnext101_32x4d   \n",
       "12    InceptionResNetV2  ourporting  80.170  95.234   inceptionresnetv2   \n",
       "13          InceptionV4  ourporting  80.062  94.926         inceptionv4   \n",
       "14    DualPathNet107_5k  ourporting  79.746  94.684   dualpathnet107_5k   \n",
       "16       DualPathNet131  ourporting  79.432  94.574      dualpathnet131   \n",
       "17     DualPathNet92_5k  ourporting  79.400  94.620    dualpathnet92_5k   \n",
       "18        DualPathNet98  ourporting  79.224  94.488       dualpathnet98   \n",
       "19   SE-ResNeXt50_32x4d  ourporting  79.076  94.434   seresnext50_32x4d   \n",
       "22     ResNeXt101_64x4d  ourporting  78.956  94.252    resnext101_64x4d   \n",
       "23             Xception  ourporting  78.888  94.292            xception   \n",
       "26         SE-ResNet152  ourporting  78.658  94.374         seresnet152   \n",
       "28         SE-ResNet101  ourporting  78.396  94.258         seresnet101   \n",
       "30     ResNeXt101_32x4d  ourporting  78.188  93.886    resnext101_32x4d   \n",
       "33          SE-ResNet50  ourporting  77.636  93.752          seresnet50   \n",
       "36          FBResNet152  ourporting  77.386  93.594         fbresnet152   \n",
       "39    DualPathNet68b_5k  ourporting  77.034  93.590   dualpathnet68b_5k   \n",
       "41       CaffeResnet101  ourporting  76.200  92.766      cafferesnet101   \n",
       "44        DualPathNet68  ourporting  75.868  92.774       dualpathnet68   \n",
       "48      NASNet-A-Mobile  ourporting  74.080  91.740       nasnetamobile   \n",
       "50          BNInception  ourporting  73.522  91.560         bninception   \n",
       "\n",
       "    avg_w_alphas  num_layers  \n",
       "27     -0.789808         555  \n",
       "34      0.838195          82  \n",
       "35     -0.616471         368  \n",
       "37      0.000000           0  \n",
       "38      0.501668         102  \n",
       "42      0.678617          86  \n",
       "43      0.140391         181  \n",
       "45      1.248708          62  \n",
       "46      0.811058         138  \n",
       "49     -0.393656         292  \n",
       "51      1.082041         111  \n",
       "52      1.158888         138  \n",
       "53      1.405073         111  \n",
       "54      1.355124          84  \n",
       "55      1.597647          66  \n",
       "56      0.610676         148  \n",
       "57      1.645383          84  \n",
       "58      1.843995          66  \n",
       "59      1.560409          23  \n",
       "60      1.942516          23  \n",
       "61      0.000000           0  \n",
       "1       0.000000           0  \n",
       "3       0.000000           0  \n",
       "5       0.000000           0  \n",
       "7       0.000000           0  \n",
       "10      0.000000           0  \n",
       "12      0.000000           0  \n",
       "13      0.000000           0  \n",
       "14      0.000000           0  \n",
       "16      0.000000           0  \n",
       "17      0.000000           0  \n",
       "18      0.000000           0  \n",
       "19      0.000000           0  \n",
       "22      0.000000           0  \n",
       "23      0.000000           0  \n",
       "26      0.000000           0  \n",
       "28      0.000000           0  \n",
       "30      0.000000           0  \n",
       "33      0.000000           0  \n",
       "36      0.000000           0  \n",
       "39      0.000000           0  \n",
       "41      0.000000           0  \n",
       "44      0.000000           0  \n",
       "48      0.000000           0  \n",
       "50      0.000000           0  "
      ]
     },
     "execution_count": 140,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "vgg19_bn\n",
      "vgg16_bn\n",
      "vgg19\n",
      "vgg16\n",
      "vgg13_bn\n",
      "vgg11_bn\n",
      "vgg13\n",
      "vgg11\n"
     ]
    }
   ],
   "source": [
    "for name in df.modelname: \n",
    "    if 'vgg' in name:\n",
    "        mf = df[df.modelname==name]\n",
    "        print(name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "VGG11 & 68.97 & 1.84 \\\\\n",
      "VGG11_BN & 70.45 & 1.60 \\\\\n",
      "\\hline\n",
      "VGG13 & 69.66 & 1.65 \\\\\n",
      "VGG13_BN & 71.51 & 1.36 \\\\\n",
      "\\hline\n",
      "VGG16 & 71.64 & 1.41 \\\\\n",
      "VGG16_BN & 73.52 & 1.08 \\\\\n",
      "\\hline\n",
      "VGG19 & 72.08 & 1.16 \\\\\n",
      "VGG19_BN & 74.27 & 0.81 \\\\\n",
      "\\hline\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "np.set_printoptions(precision=2)\n",
    "\n",
    "for name in ['vgg11', 'vgg11_bn' ,'hline', \n",
    "             'vgg13', 'vgg13_bn' ,'hline', \n",
    "             'vgg16', 'vgg16_bn' ,'hline', \n",
    "             'vgg19', 'vgg19_bn' ,'hline']:\n",
    "    if name=='hline':\n",
    "        print(\"\\\\hline\")\n",
    "    else:\n",
    "        mf = df[df.modelname==name]\n",
    "        modelname = mf.modelname.values[0].upper()\n",
    "        top1_acc = mf.acc1.values[0]\n",
    "        w_alpha = mf.avg_w_alphas.values[0]\n",
    "        print(\"{0} & {1:0.2f} & {2:0.2f} \\\\\\\\\".format(modelname, top1_acc, w_alpha))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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"
  },
  "widgets": {
   "state": {
    "52ecdae62c394a67868cd3ee83e19861": {
     "views": [
      {
       "cell_index": 14
      }
     ]
    },
    "e232317699e24554b5ed09fa6c78c662": {
     "views": [
      {
       "cell_index": 8
      }
     ]
    }
   },
   "version": "1.2.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}