notebooks/10 class Classification Of Cells Using Neural Network.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 10 Class Classification of Cells Using Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Python version     : 3.6\n",
      "Tensorflow version : 1.4.0\n",
      "numpy version      : 1.13.3\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "import os\n",
    "from datetime import datetime\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sn\n",
    "\n",
    "print('Python version     :', '{0[0]}.{0[1]}'.format(sys.version_info))\n",
    "print('Tensorflow version :', tf.__version__)\n",
    "print('numpy version      :', np.__version__)\n",
    "np.random.seed(0)\n",
    "tf.set_random_seed(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dataset\n",
    "\n",
    "Our dataset consists of total of over 4,200 cells with a total of 10 categories;  \n",
    "Each cell in the dataset is represented as a 2D matrix of transistor devices and their parameters given below:\n",
    "\n",
    "\n",
    "| Device | param1 | param2 | param3 | param4 | param5 | ...... | ...... | param13 |\n",
    "|--------|--------|--------|--------|--------|--------|--------|--------|---------|\n",
    "|   10   |    6   |    6   |    3   |    3   |    5   |        |        |    2    |\n",
    "|   10   |    8   |    3   |    2   |    4   |    4   |    2   |    3   |    6    |\n",
    "|   20   |    7   |    3   |    2   |    4   |    4   |    2   |    3   |    2    |\n",
    "|  ...   |  ...   |  ...   |  ...   |  ...   |  ...   |  ...   |  ...   |  ...    |\n",
    "|  ...   |  ...   |  ...   |  ...   |  ...   |  ...   |  ...   |  ...   |  ...    |\n",
    "|   20   |    7   |    3   |    2   |    4   |    5   |    2   |    3   |    2    |\n",
    "\n",
    "\n",
    "Entire dataset consists of 4,200 such tables is shown in the next picture."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "![Cell Dataset As Stack Of Tables](https://github.com/srohit0/mida/blob/master/data/CellDatasetAsStackOfTables.jpg?raw=true)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load cells\n",
    "\n",
    "1. Set parameters of the cells (like max devices and feature per device etc).\n",
    "2. Read cells off the disk"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# max_devices_per_cell is also max number of rows in circuit files in the training set\n",
    "max_devices_per_cell = 369; \n",
    "# feature_per_device is different from max number of cols (14) in circuit files in training set\n",
    "feature_per_device   = 5; \n",
    "feature_indices = (0,2,5,8,11)\n",
    "\n",
    "def load_one_cell(cell_file):\n",
    "    cell_cols = 14\n",
    "    cell_data = np.zeros(shape=(max_devices_per_cell, cell_cols), dtype=np.float32)\n",
    "    cell_mat  = np.loadtxt(cell_file, dtype=np.float32, ndmin=2)\n",
    "    nrows     = cell_mat.shape[0]\n",
    "    ncols     = cell_mat.shape[1]\n",
    "    cell_data[:nrows, :ncols] = cell_mat\n",
    "    cell_data = cell_data[:, feature_indices]\n",
    "    return cell_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FDN : shape= (192, 369, 5) , Mean= 0.867669 , sigma= 3.38401\n",
      "LD : shape= (192, 369, 5) , Mean= 0.448001 , sigma= 2.27175\n",
      "ADDH : shape= (208, 369, 5) , Mean= 0.57027 , sigma= 2.51812\n",
      "MAJ : shape= (320, 369, 5) , Mean= 0.717342 , sigma= 2.72987\n",
      "AOA : shape= (384, 369, 5) , Mean= 0.699328 , sigma= 2.90194\n",
      "ADDF : shape= (464, 369, 5) , Mean= 1.71977 , sigma= 4.98359\n",
      "CKG : shape= (560, 369, 5) , Mean= 1.47629 , sigma= 4.81816\n",
      "FDP : shape= (608, 369, 5) , Mean= 0.777529 , sigma= 3.18849\n",
      "INV : shape= (614, 369, 5) , Mean= 0.348261 , sigma= 2.0792\n",
      "EN : shape= (736, 369, 5) , Mean= 0.850906 , sigma= 3.03733\n",
      "dataset: shape= (4278, 369, 5) , Mean= 0.889867 , sigma= 3.45478\n",
      "labels : shape= (4278, 10)\n"
     ]
    }
   ],
   "source": [
    "def load_cells(folder):\n",
    "  #Load the data for a single label.\n",
    "  cell_files = os.listdir(folder)\n",
    "  dataset = np.zeros(shape=(len(cell_files), max_devices_per_cell, feature_per_device),\n",
    "                         dtype=np.float32)\n",
    "  cell_index = 0\n",
    "  for cell in cell_files:\n",
    "    cell_file = os.path.join(folder, cell)\n",
    "    try:\n",
    "      cell_data = load_one_cell(cell_file)\n",
    "\n",
    "      cell_rows = cell_data.shape[0]\n",
    "      cell_cols = cell_data.shape[1]\n",
    "      dataset[cell_index, :cell_rows, :cell_cols] = cell_data\n",
    "\n",
    "      cell_index = cell_index + 1\n",
    "    except IOError as e:\n",
    "      print('Could not read:', cell_file, ':', e, '- it\\'s ok, skipping.')\n",
    "    \n",
    "  dataset = dataset[0:cell_index, :, :]\n",
    "\n",
    "  print(os.path.basename(folder), ': shape=', dataset.shape, ', Mean=', np.mean(dataset), ', sigma=', np.std(dataset))\n",
    "  return dataset\n",
    "\n",
    "label_list = ['FDN', 'LD', 'ADDH', 'MAJ', 'AOA', 'ADDF', 'CKG', 'FDP', 'INV', 'EN']\n",
    "n_labels   = len(label_list); # num_classes\n",
    "for label_index in range(n_labels):\n",
    "    cell_dataset = load_cells(os.path.join('..','data','cells',label_list[label_index]))\n",
    "    cell_labels = np.zeros((cell_dataset.shape[0], n_labels))\n",
    "    cell_labels[:, label_index] = 1\n",
    "    if label_index:\n",
    "        dataset = np.append(dataset, cell_dataset, axis=0)\n",
    "        labels  = np.append(labels,  cell_labels,  axis=0)\n",
    "    else:\n",
    "        dataset = cell_dataset\n",
    "        labels  = cell_labels\n",
    "\n",
    "print('dataset: shape=', dataset.shape, ', Mean=', np.mean(dataset), ', sigma=', np.std(dataset))\n",
    "print('labels : shape=', labels.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Randomize Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Full dataset: no. of cells= (4278, 369, 5) Min= 0.0 Max= 50.0 Mean= 0.889867 Sigma= 3.45478\n"
     ]
    }
   ],
   "source": [
    "# randomize dataset and labels together to keep'em in sync.\n",
    "def randomize(dataset, labels):\n",
    "  permutation = np.random.permutation(labels.shape[0])\n",
    "  shuffled_dataset = dataset[permutation,:,:]\n",
    "  shuffled_labels = labels[permutation]\n",
    "  return shuffled_dataset, shuffled_labels\n",
    "\n",
    "# randomize dataset\n",
    "dataset, labels = randomize(dataset, labels)\n",
    "\n",
    "print('Full dataset: no. of cells=', dataset.shape, 'Min=', np.min(dataset), 'Max=',np.max(dataset), \n",
    "      'Mean=', np.mean(dataset), 'Sigma=', np.std(dataset))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Reformat Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4278, 1845)\n"
     ]
    }
   ],
   "source": [
    "def reformat(dataset):\n",
    "    dataset = dataset.reshape((-1, max_devices_per_cell*feature_per_device))\n",
    "    return dataset\n",
    "\n",
    "dataset = reformat(dataset)\n",
    "print(dataset.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Normalize Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Min= 0.0 Max= 1.0 Mean= 0.0177973 sigma= 0.0690956\n",
      "Dataset size (4278, 1845)\n"
     ]
    }
   ],
   "source": [
    "# normalize the dataset between 0 and 1.\n",
    "data_mean = 0;\n",
    "data_std  = (np.max(dataset)-np.min(dataset));\n",
    "\n",
    "def normalize(d, mean, std):\n",
    "    return (d - mean) / std\n",
    "\n",
    "def denormalize(d, mean, std):\n",
    "    return (d * std) + mean\n",
    "\n",
    "dataset = normalize(dataset, data_mean, data_std)\n",
    "\n",
    "print('Min=', np.min(dataset), 'Max=',np.max(dataset), 'Mean=',np.mean(dataset), 'sigma=', np.std(dataset))\n",
    "print('Dataset size', dataset.shape)\n",
    "#print('dataset feature min vector = ', np.min(dataset, axis=0))\n",
    "#print('dataset feature max vector = ', np.max(dataset, axis=0))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Divide dataset\n",
    "into training dataset and validation dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3422, 1845)\n"
     ]
    }
   ],
   "source": [
    "# training dataset percentage is 80%\n",
    "def divide_dataset(dataset, labels):\n",
    "    train_pct = 0.8; \n",
    "    # divide dataset into training and validation set\n",
    "    train_index = int(dataset.shape[0]*train_pct)\n",
    "    t_X = dataset[:train_index, :]\n",
    "    t_Y = labels[:train_index,:]\n",
    "    v_X = dataset[train_index:,:]\n",
    "    v_Y = labels[train_index:,:]\n",
    "    \n",
    "    return t_X, t_Y, v_X, v_Y\n",
    "\n",
    "train_X, train_Y, valid_X, valid_Y = divide_dataset(dataset, labels)\n",
    "\n",
    "print(train_X.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Declare Graph Variables\n",
    "\n",
    "Place to declare place holders and variables for compute graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "SIGMA      = np.sqrt(2.0/train_X.shape[0])\n",
    "\n",
    "# Parameters\n",
    "learning_rate = 1e-2\n",
    "l2_loss_beta  = 1e-5\n",
    "training_epochs = 1000\n",
    "display_step = 50\n",
    "\n",
    "# Network Parameters\n",
    "n_hidden_1 = 256 # 1st layer number of neurons\n",
    "n_hidden_2 = 32   # 2nd layer number of neurons\n",
    "n_features = train_X.shape[1]; # num_inputs\n",
    "\n",
    "# tf Graph input\n",
    "t_dataset = tf.placeholder(tf.float32, shape=(None, n_features), name='t_dataset');\n",
    "t_labels  = tf.placeholder(tf.float32, shape=(None, n_labels), name='t_labels');\n",
    "\n",
    "\n",
    "weights = {\n",
    "    'h1': tf.Variable(tf.truncated_normal(shape=[n_features, n_hidden_1], mean=3.0*SIGMA, \n",
    "                                          stddev=SIGMA, dtype=tf.float32), trainable=True),\n",
    "    'h2': tf.Variable(tf.truncated_normal(shape=[n_hidden_1, n_hidden_2], mean=3.0*SIGMA, \n",
    "                                           stddev=SIGMA, dtype=tf.float32), trainable=True),\n",
    "    'out': tf.Variable(tf.truncated_normal(shape=[n_hidden_2, n_labels], mean=3.0*SIGMA, \n",
    "                                           stddev=SIGMA, dtype=tf.float32), trainable=True)\n",
    "}\n",
    "biases = {\n",
    "    'b1': tf.Variable(tf.constant(value=SIGMA, dtype=tf.float32, shape=[n_hidden_1]), trainable=True),\n",
    "    'b2': tf.Variable(tf.constant(value=SIGMA, dtype=tf.float32, shape=[n_hidden_2]), trainable=True),\n",
    "    'out': tf.Variable(tf.constant(value=SIGMA, dtype=tf.float32, shape=[n_labels]), trainable=True)\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Compute Graph\n",
    "Construct a Neural Network model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def neural_net(x):\n",
    "    # Hidden fully connected layer with 256 neurons\n",
    "    layer_1 = tf.add(tf.matmul(x, weights['h1']), biases['b1'])\n",
    "    # Hidden fully connected layer with 256 neurons\n",
    "    layer_2 = tf.add(tf.matmul(layer_1, weights['h2']), biases['b2'])\n",
    "    # Output fully connected layer with a neuron for each class\n",
    "    out_layer = tf.matmul(layer_2, weights['out']) + biases['out']\n",
    "    return out_layer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Parameters and Function\n",
    "Neural network parameters and functionss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Construct model\n",
    "predictions = neural_net(t_dataset)\n",
    "\n",
    "t_labels_cls       = tf.argmax(t_labels, axis=1)\n",
    "predictions_cls    = tf.argmax(predictions, axis=1)\n",
    "correct_prediction = tf.equal(predictions_cls, t_labels_cls)\n",
    "accuracy           = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))\n",
    "\n",
    "# \n",
    "cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(logits=predictions,\n",
    "                                                               labels=t_labels_cls)\n",
    "\n",
    "# L2 loss\n",
    "regularizers = tf.nn.l2_loss(weights['h1']) + tf.nn.l2_loss(weights['h2']) + tf.nn.l2_loss(weights['out'])\n",
    "\n",
    "# Cost\n",
    "cost = tf.reduce_mean(cross_entropy + l2_loss_beta * regularizers)\n",
    "\n",
    "# Optimization algorithm\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate).minimize(cost)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[   16 sec ] Epoch: 0050 cost= 1.94959 train acc= 0.31327 valid acc= 0.315\n",
      "[   31 sec ] Epoch: 0100 cost= 1.41288 train acc= 0.45354 valid acc= 0.450\n",
      "[   45 sec ] Epoch: 0150 cost= 1.05944 train acc= 0.59527 valid acc= 0.547\n",
      "[   61 sec ] Epoch: 0200 cost= 0.81571 train acc= 0.68440 valid acc= 0.651\n",
      "[   76 sec ] Epoch: 0250 cost= 0.64334 train acc= 0.75541 valid acc= 0.706\n",
      "[   91 sec ] Epoch: 0300 cost= 0.53457 train acc= 0.80625 valid acc= 0.744\n",
      "[  105 sec ] Epoch: 0350 cost= 0.46312 train acc= 0.83694 valid acc= 0.782\n",
      "[  120 sec ] Epoch: 0400 cost= 0.40644 train acc= 0.86265 valid acc= 0.794\n",
      "[  135 sec ] Epoch: 0450 cost= 0.34390 train acc= 0.89217 valid acc= 0.821\n",
      "[  150 sec ] Epoch: 0500 cost= 0.27899 train acc= 0.91642 valid acc= 0.842\n",
      "[  168 sec ] Epoch: 0550 cost= 0.22639 train acc= 0.93396 valid acc= 0.855\n",
      "[  185 sec ] Epoch: 0600 cost= 0.18298 train acc= 0.95149 valid acc= 0.875\n",
      "[  201 sec ] Epoch: 0650 cost= 0.14901 train acc= 0.96230 valid acc= 0.888\n",
      "[  215 sec ] Epoch: 0700 cost= 0.49565 train acc= 0.83869 valid acc= 0.827\n",
      "[  230 sec ] Epoch: 0750 cost= 0.15865 train acc= 0.96172 valid acc= 0.904\n",
      "[  245 sec ] Epoch: 0800 cost= 0.12858 train acc= 0.97195 valid acc= 0.907\n",
      "[  260 sec ] Epoch: 0850 cost= 0.11212 train acc= 0.97750 valid acc= 0.905\n",
      "[  276 sec ] Epoch: 0900 cost= 0.10019 train acc= 0.98159 valid acc= 0.908\n",
      "[  291 sec ] Epoch: 0950 cost= 0.09083 train acc= 0.98480 valid acc= 0.910\n",
      "[  306 sec ] Epoch: 1000 cost= 0.08315 train acc= 0.98714 valid acc= 0.911\n",
      "\n",
      "Optimization Finished!\n"
     ]
    }
   ],
   "source": [
    "# Start training\n",
    "sess = tf.Session()\n",
    "init = tf.global_variables_initializer()\n",
    "sess.run(init)\n",
    "\n",
    "t_start = datetime.now()\n",
    "\n",
    "for epoch in range(training_epochs):\n",
    "\n",
    "    sess.run(optimizer, feed_dict={t_dataset:train_X, t_labels:train_Y})\n",
    "\n",
    "    #Display logs per epoch step\n",
    "    if (epoch+1) % display_step == 0:\n",
    "\n",
    "        t_cost, t_acc = sess.run([cost, accuracy], feed_dict={t_dataset:train_X, t_labels:train_Y})\n",
    "        v_cost, v_acc = sess.run([cost, accuracy], feed_dict={t_dataset:valid_X, t_labels:valid_Y})\n",
    "        print (\"[\", \"{:4d}\".format((datetime.now()-t_start).seconds), \n",
    "               \"sec ] Epoch:\", '%04d' % (epoch+1), \n",
    "               \"cost=\", \"{:.5f}\".format(t_cost),\n",
    "               \"train acc=\", \"{:.5f}\".format(t_acc),\n",
    "               \"valid acc=\", \"{:.3f}\".format(v_acc))\n",
    "\n",
    "print( \"\\nOptimization Finished!\" )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Cost and Accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training   cost    = 0.0831519\n",
      "Validation cost    = 0.316682 \n",
      "\n",
      "Training   Accuracy = 0.987142\n",
      "Validation Accuracy = 0.911215\n"
     ]
    }
   ],
   "source": [
    "# Find cost on training dataset.\n",
    "t_cost, t_acc = sess.run([cost, accuracy], feed_dict={t_dataset:train_X, t_labels:train_Y})\n",
    "v_cost, v_acc = sess.run([cost, accuracy], feed_dict={t_dataset:valid_X, t_labels:valid_Y})\n",
    "\n",
    "print (\"Training   cost    =\", t_cost)\n",
    "print (\"Validation cost    =\", v_cost, '\\n')\n",
    "print(\"Training   Accuracy =\", t_acc)\n",
    "print(\"Validation Accuracy =\", v_acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0AAAANXCAYAAAAcsD9hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x212627ad978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pred = sess.run(predictions, feed_dict={t_dataset:dataset, t_labels:labels})\n",
    "cm = tf.confusion_matrix(labels=np.argmax(labels, axis=1), predictions=np.argmax(pred, axis=1), num_classes=n_labels)\n",
    "cm_ary = cm.eval(session=sess)\n",
    "\n",
    "plt.figure(figsize = (1.5*n_labels,1.5*n_labels))\n",
    "sn.heatmap(cm_ary,   xticklabels=label_list, yticklabels=label_list,\n",
    "           cmap=\"Blues\", annot=True, fmt='d', annot_kws={\"size\": 20})# font size\n",
    "sn.set(font_scale=2.0)#for label size\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion\n",
    "Neural networks gave us reasonable accuracy of over 90% in both training and validation sets.\n",
    "\n",
    "In the confusion matrix, it is easy to see that EN (ExNor function) and MAJ (majority function) have large numbers of false predicions. \n",
    "\n",
    "We also know that these two categories of cells have similar structural implementation. It is speculated that neural network **`magically`** figures this out. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "#sess.close()"
   ]
  }
 ],
 "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}