show_within_network_RSMs.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "IPython.notebook.set_autosave_interval(5000)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Autosaving every 5 seconds\n"
     ]
    }
   ],
   "source": [
    "import hypertools as hyp\n",
    "from qmvpa import utils\n",
    "from data_loader import load_cifar\n",
    "from sklearn.model_selection import train_test_split\n",
    "import os\n",
    "import sys \n",
    "import numpy as np\n",
    "# import resnet\n",
    "import matplotlib.pyplot as plt \n",
    "import seaborn as sns \n",
    "from dep.read_acts_keras import get_activations\n",
    "from dep.utils import subset_units\n",
    "\n",
    "%matplotlib inline \n",
    "# %matplotlib notebook\n",
    "%autosave 5\n",
    "sns.set(style = 'white', font_scale = 1.2, rc={\"lines.linewidth\": 2.5})\n",
    "\n",
    "log_root = '/tigress/qlu/logs/keras-resnet/log'\n",
    "\n",
    "data_name = 'cifar10'\n",
    "model_name = 'resnet18'\n",
    "\n",
    "# load data \n",
    "_, X_test, _, Y_test, _, y_test, data_info = load_cifar(data_name)\n",
    "[n_classes, img_rows, img_cols, img_channels] = data_info\n",
    "n_test_egs = len(y_test)\n",
    "y_test_ids = np.argsort(np.ravel(y_test))\n",
    "\n",
    "# define params \n",
    "n_subjs = 5\n",
    "log_epochs = np.round(np.logspace(0,2,10))\n",
    "log_epochs = np.insert(log_epochs, 0, 0).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "layer_selected = [2, 10, 17, 25, 32, 40, 47, 55, 62, 66]\n",
    "window_size = 1000\n",
    "n_max_units = 500\n",
    "rdm_len = np.shape(y_test)[0] // window_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Layer 2: \n",
      "0 1 2 3 5 8 13 22 36 60 100 \n",
      "Layer 10: \n",
      "0 1 2 3 5 8 13 22 36 60 100 \n",
      "Layer 17: \n",
      "0 1 2 3 5 8 13 22 36 60 100 \n",
      "Layer 25: \n",
      "0 1 2 3 5 8 13 22 36 60 100 \n",
      "Layer 32: \n",
      "0 1 2 3 5 8 13 22 36 60 100 \n",
      "Layer 40: \n",
      "0 1 2 3 5 8 13 22 36 60 100 \n",
      "Layer 47: \n",
      "0 1 2 3 5 8 13 22 36 60 100 \n",
      "Layer 55: \n",
      "0 1 2 3 5 8 13 22 36 60 100 \n",
      "Layer 62: \n",
      "0 1 2 3 5 8 13 22 36 60 100 \n",
      "Layer 66: \n",
      "0 1 2 3 5 8 13 22 36 60 100 \n"
     ]
    }
   ],
   "source": [
    "# compute all RDMs\n",
    "rdm_le = []\n",
    "acts_le = []\n",
    "# loop over layers \n",
    "for l in layer_selected:\n",
    "    print('Layer %d: ' % l)\n",
    "    acts_e = []\n",
    "    rdm_e = []\n",
    "    # loop over epoches  \n",
    "    for e in log_epochs:\n",
    "        # activation log dir \n",
    "        print('%d ' % e, end='')\n",
    "        rdm_avg = np.zeros((rdm_len, rdm_len))\n",
    "        for subj_id in range(n_subjs):\n",
    "            log_dir = os.path.join(log_root, data_name, model_name, 'subj%.2d' % (subj_id))\n",
    "            acts_path = os.path.join(log_dir, 'epoch_%.3d' % e, 'activations')\n",
    "            acts_path_l = os.path.join(acts_path, 'layer_%.3d.npy' % (l))\n",
    "\n",
    "            # load activity \n",
    "            loaded_acts = np.load(acts_path_l)\n",
    "            loaded_acts = np.reshape(loaded_acts, [n_test_egs, -1])\n",
    "\n",
    "            # subset units \n",
    "            n_units = np.shape(loaded_acts)[1]\n",
    "            if n_units > n_max_units: \n",
    "                loaded_acts = subset_units(loaded_acts, n_max_units)\n",
    "\n",
    "            # use that to order X\n",
    "            sloaded_acts = loaded_acts[y_test_ids, :]\n",
    "            asloaded_acts = utils.window_avg(sloaded_acts, window_size)\n",
    "            \n",
    "            # accumulate RDM\n",
    "            rdm_avg += np.corrcoef(asloaded_acts) / n_subjs\n",
    "            \n",
    "        # done all subjects...\n",
    "        rdm_e.append(rdm_avg)\n",
    "#             acts_e.append(asloaded_acts)\n",
    "    # done all epoches...\n",
    "    print()\n",
    "    rdm_le.append(rdm_e)\n",
    "#     acts_le.append(acts_e)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,0,'Training epoches')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 792x720 with 110 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline \n",
    "sns.set(style = 'white', font_scale = 1.2, rc={\"lines.linewidth\": 2.5})\n",
    "\n",
    "n_layers = len(layer_selected)\n",
    "n_epoches = len(log_epochs)\n",
    "figsize = (n_epoches, n_layers)\n",
    "\n",
    "# n_epoches = 4\n",
    "# n_layers = 3\n",
    "\n",
    "f, axs = plt.subplots(n_layers, n_epoches, figsize = figsize, \n",
    "                      sharex=True,sharey=True)\n",
    "for l in range(n_layers):\n",
    "    for e in range(n_epoches):\n",
    "        sns.heatmap(\n",
    "            rdm_le[l][e], \n",
    "            cmap='viridis', square = True, \n",
    "            vmax=1, vmin=0, \n",
    "            xticklabels = False, yticklabels = False, cbar=False, \n",
    "            ax = axs[l,e]\n",
    "        )\n",
    "f.tight_layout()\n",
    "\n",
    "f.suptitle('RDM ~ Depth x Training, %s, %s' % (data_name, model_name), y = 1.02)\n",
    "f.text(0, 0.5, 'Network layer', va='center', rotation='vertical')\n",
    "f.text(0.5, 0, 'Training epoches', ha='center')"
   ]
  },
  {
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}