Baby Neural Turing Machines.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Baby Neural Turing Machines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import libraries and relevant dependencies\n",
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "import torch.optim as optim\n",
    "import random\n",
    "import string\n",
    "from torch.autograd import Variable\n",
    "\n",
    "# Dyck library\n",
    "from tasks.dyck_generator import DyckLanguage\n",
    "\n",
    "# RNN Models\n",
    "from models.ntm_models import BNTM_Softmax, BNTM_SoftmaxTemperature, BNTM_GumbelSoftmax\n",
    "\n",
    "# Set default tensor type \"double\"\n",
    "torch.set_default_tensor_type('torch.DoubleTensor')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Fix the random seed:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RANDOM SEED: 1729\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<torch._C.Generator at 0x11b3ffcf0>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "randomseed_num = 1729\n",
    "print ('RANDOM SEED: {}'.format(randomseed_num))\n",
    "random.seed (randomseed_num)\n",
    "np.random.seed (randomseed_num)\n",
    "torch.manual_seed(randomseed_num)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### GPU/CPU Checkpoint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cpu\n"
     ]
    }
   ],
   "source": [
    "# GPU/CPU Check\n",
    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\") ## GPU stuff\n",
    "print (device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dyck Languages"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A probabilistic context-free grammar for $\\mathcal{D}_n$ can be written as follows:\n",
    "\n",
    "\\begin{align*}\n",
    "S \\rightarrow \\begin{cases} \n",
    "(_i\\, S\\, )_i & \\text{with probability } \\frac{p}{n} \\\\\n",
    "S\\,S & \\text{with probability } q \\\\ \n",
    "\\varepsilon & \\text{with probability } 1 - (p+q) \n",
    "\\end{cases}\n",
    "\\end{align*}\n",
    "where $0 < p, q < 1$ and $p+q < 1$."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training and Test Corpora Generation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Training corpus window : [`MIN_SIZE`, `MAX_SIZE`]\n",
    "\n",
    "Test corpus window: [`MAX_SIZE+2`, `2*MAX_SIZE`]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading data...\n",
      "500 samples generated.\n",
      "1000 samples generated.\n",
      "1500 samples generated.\n",
      "2000 samples generated.\n",
      "2500 samples generated.\n",
      "3000 samples generated.\n",
      "3500 samples generated.\n",
      "4000 samples generated.\n",
      "4500 samples generated.\n",
      "5000 samples generated.\n",
      "500 samples generated.\n",
      "1000 samples generated.\n",
      "1500 samples generated.\n",
      "2000 samples generated.\n",
      "2500 samples generated.\n",
      "3000 samples generated.\n",
      "3500 samples generated.\n",
      "4000 samples generated.\n",
      "4500 samples generated.\n",
      "5000 samples generated.\n",
      "[([[[]]]())]\n",
      "212222211120\n",
      "tensor([[[0., 0., 1., 0.]],\n",
      "\n",
      "        [[1., 0., 0., 0.]],\n",
      "\n",
      "        [[0., 0., 1., 0.]],\n",
      "\n",
      "        [[0., 0., 1., 0.]],\n",
      "\n",
      "        [[0., 0., 1., 0.]],\n",
      "\n",
      "        [[0., 0., 0., 1.]],\n",
      "\n",
      "        [[0., 0., 0., 1.]],\n",
      "\n",
      "        [[0., 0., 0., 1.]],\n",
      "\n",
      "        [[1., 0., 0., 0.]],\n",
      "\n",
      "        [[0., 1., 0., 0.]],\n",
      "\n",
      "        [[0., 1., 0., 0.]],\n",
      "\n",
      "        [[0., 0., 0., 1.]]])\n",
      "tensor([[1., 0., 1., 1.],\n",
      "        [1., 1., 1., 0.],\n",
      "        [1., 0., 1., 1.],\n",
      "        [1., 0., 1., 1.],\n",
      "        [1., 0., 1., 1.],\n",
      "        [1., 0., 1., 1.],\n",
      "        [1., 0., 1., 1.],\n",
      "        [1., 1., 1., 0.],\n",
      "        [1., 1., 1., 0.],\n",
      "        [1., 1., 1., 0.],\n",
      "        [1., 0., 1., 1.],\n",
      "        [1., 0., 1., 0.]])\n",
      "((()([[[((()))(([[(())()((([()])))[][]()([])]]([(([[[]]]))]([((()))]))))([])]]][[[]](())])))\n",
      "11111222111112112211121211121211122222121212211211222221121121111121111212122212222211121110\n"
     ]
    }
   ],
   "source": [
    "## Parameters of the Probabilistic Dyck Language \n",
    "NUM_PAR = 2\n",
    "MIN_SIZE = 2\n",
    "MAX_SIZE = 50\n",
    "P_VAL = 0.5\n",
    "Q_VAL = 0.25\n",
    "\n",
    "# Number of samples in the training corpus\n",
    "TRAINING_SIZE = 5000\n",
    "# Number of samples in the test corpus\n",
    "TEST_SIZE = 5000\n",
    "\n",
    "# Create a Dyck language generator\n",
    "Dyck = DyckLanguage (NUM_PAR, P_VAL, Q_VAL)\n",
    "all_letters = word_set = Dyck.return_vocab ()\n",
    "n_letters = vocab_size = len (word_set)\n",
    "\n",
    "print('Loading data...')\n",
    "\n",
    "training_input, training_output, st = Dyck.training_set_generator (TRAINING_SIZE, MIN_SIZE, MAX_SIZE)\n",
    "test_input, test_output, st2 = Dyck.training_set_generator (TEST_SIZE, MAX_SIZE + 2, 2 * MAX_SIZE)\n",
    "\n",
    "for i in range (1):\n",
    "    print (training_input[i])\n",
    "    print (training_output[i])\n",
    "    print (Dyck.lineToTensor(training_input[i]))\n",
    "    print (Dyck.lineToTensorSigmoid(training_output[i]))\n",
    "    print (test_input[i])\n",
    "    print (test_output[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Baby-NTM  Parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Number of hidden units\n",
    "n_hidden = 12\n",
    "# Number of hidden layers\n",
    "n_layers = 1\n",
    "# Memory size\n",
    "memory_size = 104\n",
    "memory_dim = 5\n",
    "\n",
    "# Parameters for the temperature-based methods\n",
    "temp = 1.0\n",
    "temp_min = 0.5\n",
    "ANNEAL_RATE = .00001"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Stack-RNN with Softmax\n",
    "model = BNTM_Softmax (n_hidden, vocab_size, vocab_size, n_layers, memory_size, memory_dim).to(device)\n",
    "# Learning rate\n",
    "learning_rate = .01\n",
    "# Minimum Squared Error (MSE) loss\n",
    "criterion = nn.MSELoss() \n",
    "# Adam optimizer (https://arxiv.org/abs/1412.6980)\n",
    "optim = torch.optim.Adam(model.parameters(), lr = learning_rate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model details:\n",
      "BNTM_Softmax(\n",
      "  (rnn): RNN(4, 12)\n",
      "  (W_m): Linear(in_features=5, out_features=12, bias=True)\n",
      "  (W_y): Linear(in_features=12, out_features=4, bias=True)\n",
      "  (W_n): Linear(in_features=12, out_features=5, bias=True)\n",
      "  (W_a): Linear(in_features=12, out_features=5, bias=True)\n",
      "  (softmax): Softmax(dim=0)\n",
      "  (sigmoid): Sigmoid()\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "print ('Model details:')\n",
    "print (model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Number of epochs to train our model\n",
    "epoch_num = 5\n",
    "# Output threshold\n",
    "epsilon = 0.5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training and Testing the Stack-RNN Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def test_model (model, data_input, data_output):\n",
    "    # Turn on the eval mode\n",
    "    model.eval()\n",
    "    # Total number of \"correctly\" predicted samples\n",
    "    correct_num = 0\n",
    "    with torch.no_grad():\n",
    "        for i in range (len(data_output)):\n",
    "            len_input = len (data_input[i])\n",
    "            model.zero_grad ()\n",
    "            # Initialize the hidden state\n",
    "            hidden = model.init_hidden()\n",
    "            # Initialize the memory\n",
    "            memory = torch.zeros (memory_size, memory_dim).to(device)\n",
    "            # Target values\n",
    "            target = Dyck.lineToTensorSigmoid(data_output[i]).to(device) \n",
    "            # Output values\n",
    "            output_vals = torch.zeros (target.shape)\n",
    "            \n",
    "            for j in range (len_input):\n",
    "                output, hidden, memory = model (Dyck.lineToTensor(data_input[i][j]).to(device), hidden, memory)\n",
    "                output_vals [j] = output\n",
    "\n",
    "            # Binarize the entries based on the output threshold\n",
    "            out_np = np.int_(output_vals.detach().numpy() >= epsilon)\n",
    "            target_np = np.int_(target.detach().numpy())\n",
    "            \n",
    "            # (Double-)check whether the output values and the target values are the same\n",
    "            if np.all(np.equal(out_np, target_np)) and (out_np.flatten() == target_np.flatten()).all():\n",
    "                # If so, increase `correct_num` by one\n",
    "                correct_num += 1\n",
    "                \n",
    "    return float(correct_num)/len(data_output) * 100, correct_num"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train (model, optimizer, criterion, epoch_num=3):\n",
    "    # Turn on the train model for the model\n",
    "    model.train()\n",
    "    # Arrays for loss and \"moving\" accuracy per epoch\n",
    "    loss_arr = []\n",
    "    correct_arr = []\n",
    "    for epoch in range(1, epoch_num + 1):\n",
    "        print ('Epoch: {}'.format(epoch))\n",
    "        \n",
    "        # Total loss per epoch\n",
    "        total_loss = 0\n",
    "        # Total number of \"correctly\" predicted samples so far in the epoch\n",
    "        counter = 0\n",
    "\n",
    "        for i in range (TRAINING_SIZE):\n",
    "            len_input = len (training_input[i])\n",
    "            # Good-old zero grad\n",
    "            model.zero_grad ()\n",
    "            # Initialize the hidden state\n",
    "            hidden = model.init_hidden()\n",
    "            # Initialize the memory \n",
    "            memory = torch.zeros (memory_size, memory_dim).to(device)\n",
    "            # Target values\n",
    "            target = Dyck.lineToTensorSigmoid(training_output[i]).to(device) \n",
    "            # Output values\n",
    "            output_vals = torch.zeros (target.shape)\n",
    "\n",
    "            for j in range (len_input):\n",
    "                output, hidden, memory = model (Dyck.lineToTensor(training_input[i][j]).to(device), hidden, memory)\n",
    "                output_vals [j] = output\n",
    "            \n",
    "            # MSE (y, y_bar)\n",
    "            loss = criterion (output_vals, target)\n",
    "            # Add the current loss to the total loss\n",
    "            total_loss += loss.item()\n",
    "            # Backprop! \n",
    "            loss.backward ()\n",
    "            optimizer.step ()\n",
    "            \n",
    "            # Print the performance of the model every 500 steps\n",
    "            if i % 250 == 0:\n",
    "                print ('Sample Number {}: '.format(i))\n",
    "                print ('Input : {}'.format(training_input[i]))\n",
    "                print ('Output: {}'.format(training_output[i]))\n",
    "                print ('* Counter: {}'.format(counter))\n",
    "                print ('* Avg Loss: {}'.format(total_loss/(i+1))) \n",
    "\n",
    "            # Binarize the entries based on the output threshold\n",
    "            out_np = np.int_(output_vals.detach().numpy() >= epsilon)\n",
    "            target_np = np.int_(target.detach().numpy())\n",
    "                \n",
    "            # \"Moving\" training accuracy\n",
    "            if np.all(np.equal(out_np, target_np)) and (out_np.flatten() == target_np.flatten()).all():\n",
    "                counter += 1\n",
    "                \n",
    "            # At the end of the epoch, append our total loss and \"moving\" accuracy\n",
    "            if i == TRAINING_SIZE - 1:\n",
    "                print ('Counter: {}'.format(float(counter)/TRAINING_SIZE))\n",
    "                loss_arr.append (total_loss)\n",
    "                correct_arr.append(counter) \n",
    "\n",
    "        if epoch % 1 == 0:\n",
    "            print ('Training Accuracy %: ', correct_arr)\n",
    "            print ('Loss: ', loss_arr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Let there be light!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 1\n",
      "Sample Number 0: \n",
      "Input : [([[[]]]())]\n",
      "Output: 212222211120\n",
      "* Counter: 0\n",
      "* Avg Loss: 0.28577524031818285\n",
      "Sample Number 250: \n",
      "Input : [[[(()([[(())]]))]]]\n",
      "Output: 22211112211122112220\n",
      "* Counter: 1\n",
      "* Avg Loss: 0.0837031479525154\n",
      "Sample Number 500: \n",
      "Input : ([(()(()))])\n",
      "Output: 121111111210\n",
      "* Counter: 7\n",
      "* Avg Loss: 0.06505831459496109\n",
      "Sample Number 750: \n",
      "Input : [(([[]])[[][]])]\n",
      "Output: 2112221122222120\n",
      "* Counter: 31\n",
      "* Avg Loss: 0.05563432172454156\n",
      "Sample Number 1000: \n",
      "Input : [[[[]][[]]]]\n",
      "Output: 222222222220\n",
      "* Counter: 57\n",
      "* Avg Loss: 0.04954909904902164\n",
      "Sample Number 1250: \n",
      "Input : [([[((()))]])]\n",
      "Output: 21221111122120\n",
      "* Counter: 234\n",
      "* Avg Loss: 0.04096930254240002\n",
      "Sample Number 1500: \n",
      "Input : [(([([])([[[[[(())[]]]]]])]()[()]))()][([()]([]))]\n",
      "Output: 21121212122222111222222212111212112120212121121120\n",
      "* Counter: 478\n",
      "* Avg Loss: 0.03440341456924498\n",
      "Sample Number 1750: \n",
      "Input : (([([])]))(())\n",
      "Output: 11212121101110\n",
      "* Counter: 711\n",
      "* Avg Loss: 0.02972583457456173\n",
      "Sample Number 2000: \n",
      "Input : ([[[]]])((([()](()))))()[([])]\n",
      "Output: 122222101112121111111010212120\n",
      "* Counter: 961\n",
      "* Avg Loss: 0.026052531459735826\n",
      "Sample Number 2250: \n",
      "Input : [((([()[[](((([([])]))))]])))]\n",
      "Output: 211121222211112121211112211120\n",
      "* Counter: 1211\n",
      "* Avg Loss: 0.023182489191475076\n",
      "Sample Number 2500: \n",
      "Input : ()[[]]()\n",
      "Output: 10222010\n",
      "* Counter: 1461\n",
      "* Avg Loss: 0.020881048267820126\n",
      "Sample Number 2750: \n",
      "Input : ([(([]))])([[]])\n",
      "Output: 1211211210122210\n",
      "* Counter: 1711\n",
      "* Avg Loss: 0.01899422001327016\n",
      "Sample Number 3000: \n",
      "Input : [[(([]))[]([[([[([])]])]()])()()]]\n",
      "Output: 2211211222122122121221221212121220\n",
      "* Counter: 1961\n",
      "* Avg Loss: 0.01741951776911101\n",
      "Sample Number 3250: \n",
      "Input : [[[([()])]][(())]]\n",
      "Output: 222121212222111220\n",
      "* Counter: 2211\n",
      "* Avg Loss: 0.016085490225316457\n",
      "Sample Number 3500: \n",
      "Input : (((([]))))[([[]][])]\n",
      "Output: 11112111102122212120\n",
      "* Counter: 2461\n",
      "* Avg Loss: 0.014940835949126577\n",
      "Sample Number 3750: \n",
      "Input : ([[]()][[(((())))]])([()]([[()]])([[]]))\n",
      "Output: 1222121221111111221012121122122111222110\n",
      "* Counter: 2711\n",
      "* Avg Loss: 0.0139481741578528\n",
      "Sample Number 4000: \n",
      "Input : [([[([[([])]])((([])())())]]([][[]])())]\n",
      "Output: 2122122121221211121111111221121222111120\n",
      "* Counter: 2961\n",
      "* Avg Loss: 0.013079021090917823\n",
      "Sample Number 4250: \n",
      "Input : ()([])([((([([([])[]])])))])()()\n",
      "Output: 10121012111212121222121112101010\n",
      "* Counter: 3211\n",
      "* Avg Loss: 0.012311705616018943\n",
      "Sample Number 4500: \n",
      "Input : [(((([][([])]))))]\n",
      "Output: 211112121212111120\n",
      "* Counter: 3461\n",
      "* Avg Loss: 0.011629371725823788\n",
      "Sample Number 4750: \n",
      "Input : ([[][[[]]]](())[(([[(())]]))([])])\n",
      "Output: 1222222222111112112211122112121210\n",
      "* Counter: 3711\n",
      "* Avg Loss: 0.01101859871604837\n",
      "Counter: 0.7922\n",
      "Training Accuracy %:  [3961]\n",
      "Loss:  [52.354061373115464]\n",
      "Epoch: 2\n",
      "Sample Number 0: \n",
      "Input : [([[[]]]())]\n",
      "Output: 212222211120\n",
      "* Counter: 0\n",
      "* Avg Loss: 1.3280545888898526e-05\n",
      "Sample Number 250: \n",
      "Input : [[[(()([[(())]]))]]]\n",
      "Output: 22211112211122112220\n",
      "* Counter: 250\n",
      "* Avg Loss: 1.7633620865289934e-05\n",
      "Sample Number 500: \n",
      "Input : ([(()(()))])\n",
      "Output: 121111111210\n",
      "* Counter: 500\n",
      "* Avg Loss: 1.5763037365877155e-05\n",
      "Sample Number 750: \n",
      "Input : [(([[]])[[][]])]\n",
      "Output: 2112221122222120\n",
      "* Counter: 750\n",
      "* Avg Loss: 1.436593614990783e-05\n",
      "Sample Number 1000: \n",
      "Input : [[[[]][[]]]]\n",
      "Output: 222222222220\n",
      "* Counter: 1000\n",
      "* Avg Loss: 1.3293684518152303e-05\n",
      "Sample Number 1250: \n",
      "Input : [([[((()))]])]\n",
      "Output: 21221111122120\n",
      "* Counter: 1250\n",
      "* Avg Loss: 1.2296005942398417e-05\n",
      "Sample Number 1500: \n",
      "Input : [(([([])([[[[[(())[]]]]]])]()[()]))()][([()]([]))]\n",
      "Output: 21121212122222111222222212111212112120212121121120\n",
      "* Counter: 1500\n",
      "* Avg Loss: 1.1460974147427605e-05\n",
      "Sample Number 1750: \n",
      "Input : (([([])]))(())\n",
      "Output: 11212121101110\n",
      "* Counter: 1750\n",
      "* Avg Loss: 1.0715437761856678e-05\n",
      "Sample Number 2000: \n",
      "Input : ([[[]]])((([()](()))))()[([])]\n",
      "Output: 122222101112121111111010212120\n",
      "* Counter: 2000\n",
      "* Avg Loss: 1.0046009610964743e-05\n",
      "Sample Number 2250: \n",
      "Input : [((([()[[](((([([])]))))]])))]\n",
      "Output: 211121222211112121211112211120\n",
      "* Counter: 2250\n",
      "* Avg Loss: 9.432049132776778e-06\n",
      "Sample Number 2500: \n",
      "Input : ()[[]]()\n",
      "Output: 10222010\n",
      "* Counter: 2500\n",
      "* Avg Loss: 8.889472039104423e-06\n",
      "Sample Number 2750: \n",
      "Input : ([(([]))])([[]])\n",
      "Output: 1211211210122210\n",
      "* Counter: 2750\n",
      "* Avg Loss: 8.395681864127496e-06\n",
      "Sample Number 3000: \n",
      "Input : [[(([]))[]([[([[([])]])]()])()()]]\n",
      "Output: 2211211222122122121221221212121220\n",
      "* Counter: 3000\n",
      "* Avg Loss: 7.93919530786998e-06\n",
      "Sample Number 3250: \n",
      "Input : [[[([()])]][(())]]\n",
      "Output: 222121212222111220\n",
      "* Counter: 3250\n",
      "* Avg Loss: 7.523330915710623e-06\n",
      "Sample Number 3500: \n",
      "Input : (((([]))))[([[]][])]\n",
      "Output: 11112111102122212120\n",
      "* Counter: 3500\n",
      "* Avg Loss: 7.1383087488970366e-06\n",
      "Sample Number 3750: \n",
      "Input : ([[]()][[(((())))]])([()]([[()]])([[]]))\n",
      "Output: 1222121221111111221012121122122111222110\n",
      "* Counter: 3750\n",
      "* Avg Loss: 6.788549614695698e-06\n",
      "Sample Number 4000: \n",
      "Input : [([[([[([])]])((([])())())]]([][[]])())]\n",
      "Output: 2122122121221211121111111221121222111120\n",
      "* Counter: 4000\n",
      "* Avg Loss: 6.4687264127716115e-06\n",
      "Sample Number 4250: \n",
      "Input : ()([])([((([([([])[]])])))])()()\n",
      "Output: 10121012111212121222121112101010\n",
      "* Counter: 4250\n",
      "* Avg Loss: 6.173563181962773e-06\n",
      "Sample Number 4500: \n",
      "Input : [(((([][([])]))))]\n",
      "Output: 211112121212111120\n",
      "* Counter: 4500\n",
      "* Avg Loss: 5.902776937267716e-06\n",
      "Sample Number 4750: \n",
      "Input : ([[][[[]]]](())[(([[(())]]))([])])\n",
      "Output: 1222222222111112112211122112121210\n",
      "* Counter: 4750\n",
      "* Avg Loss: 5.6514415169014794e-06\n",
      "Counter: 1.0\n",
      "Training Accuracy %:  [3961, 5000]\n",
      "Loss:  [52.354061373115464, 0.027094410170337646]\n",
      "Epoch: 3\n",
      "Sample Number 0: \n",
      "Input : [([[[]]]())]\n",
      "Output: 212222211120\n",
      "* Counter: 0\n",
      "* Avg Loss: 7.275682238904773e-07\n",
      "Sample Number 250: \n",
      "Input : [[[(()([[(())]]))]]]\n",
      "Output: 22211112211122112220\n",
      "* Counter: 250\n",
      "* Avg Loss: 9.320995071368863e-07\n",
      "Sample Number 500: \n",
      "Input : ([(()(()))])\n",
      "Output: 121111111210\n",
      "* Counter: 500\n",
      "* Avg Loss: 8.521422435680202e-07\n",
      "Sample Number 750: \n",
      "Input : [(([[]])[[][]])]\n",
      "Output: 2112221122222120\n",
      "* Counter: 750\n",
      "* Avg Loss: 7.872139162930154e-07\n",
      "Sample Number 1000: \n",
      "Input : [[[[]][[]]]]\n",
      "Output: 222222222220\n",
      "* Counter: 1000\n",
      "* Avg Loss: 7.343099014566827e-07\n",
      "Sample Number 1250: \n",
      "Input : [([[((()))]])]\n",
      "Output: 21221111122120\n",
      "* Counter: 1250\n",
      "* Avg Loss: 6.847585210646409e-07\n",
      "Sample Number 1500: \n",
      "Input : [(([([])([[[[[(())[]]]]]])]()[()]))()][([()]([]))]\n",
      "Output: 21121212122222111222222212111212112120212121121120\n",
      "* Counter: 1500\n",
      "* Avg Loss: 6.438792903848332e-07\n",
      "Sample Number 1750: \n",
      "Input : (([([])]))(())\n",
      "Output: 11212121101110\n",
      "* Counter: 1750\n",
      "* Avg Loss: 6.063146532144561e-07\n",
      "Sample Number 2000: \n",
      "Input : ([[[]]])((([()](()))))()[([])]\n",
      "Output: 122222101112121111111010212120\n",
      "* Counter: 2000\n",
      "* Avg Loss: 5.713310005818719e-07\n",
      "Sample Number 2250: \n",
      "Input : [((([()[[](((([([])]))))]])))]\n",
      "Output: 211121222211112121211112211120\n",
      "* Counter: 2250\n",
      "* Avg Loss: 5.392432418322106e-07\n",
      "Sample Number 2500: \n",
      "Input : ()[[]]()\n",
      "Output: 10222010\n",
      "* Counter: 2500\n",
      "* Avg Loss: 5.105547044714326e-07\n",
      "Sample Number 2750: \n",
      "Input : ([(([]))])([[]])\n",
      "Output: 1211211210122210\n",
      "* Counter: 2750\n",
      "* Avg Loss: 4.840681145037346e-07\n",
      "Sample Number 3000: \n",
      "Input : [[(([]))[]([[([[([])]])]()])()()]]\n",
      "Output: 2211211222122122121221221212121220\n",
      "* Counter: 3000\n",
      "* Avg Loss: 4.5923094395290523e-07\n",
      "Sample Number 3250: \n",
      "Input : [[[([()])]][(())]]\n",
      "Output: 222121212222111220\n",
      "* Counter: 3250\n",
      "* Avg Loss: 4.36957399921065e-07\n",
      "Sample Number 3500: \n",
      "Input : (((([]))))[([[]][])]\n",
      "Output: 11112111102122212120\n",
      "* Counter: 3500\n",
      "* Avg Loss: 4.156604039918185e-07\n",
      "Sample Number 3750: \n",
      "Input : ([[]()][[(((())))]])([()]([[()]])([[]]))\n",
      "Output: 1222121221111111221012121122122111222110\n",
      "* Counter: 3750\n",
      "* Avg Loss: 3.9616992626005513e-07\n",
      "Sample Number 4000: \n",
      "Input : [([[([[([])]])((([])())())]]([][[]])())]\n",
      "Output: 2122122121221211121111111221121222111120\n",
      "* Counter: 4000\n",
      "* Avg Loss: 3.782715978395199e-07\n",
      "Sample Number 4250: \n",
      "Input : ()([])([((([([([])[]])])))])()()\n",
      "Output: 10121012111212121222121112101010\n",
      "* Counter: 4250\n",
      "* Avg Loss: 3.616247150471796e-07\n",
      "Sample Number 4500: \n",
      "Input : [(((([][([])]))))]\n",
      "Output: 211112121212111120\n",
      "* Counter: 4500\n",
      "* Avg Loss: 3.4626204183457815e-07\n",
      "Sample Number 4750: \n",
      "Input : ([[][[[]]]](())[(([[(())]]))([])])\n",
      "Output: 1222222222111112112211122112121210\n",
      "* Counter: 4750\n",
      "* Avg Loss: 3.324308149346935e-07\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Counter: 1.0\n",
      "Training Accuracy %:  [3961, 5000, 5000]\n",
      "Loss:  [52.354061373115464, 0.027094410170337646, 0.0015958646551512323]\n",
      "Epoch: 4\n",
      "Sample Number 0: \n",
      "Input : [([[[]]]())]\n",
      "Output: 212222211120\n",
      "* Counter: 0\n",
      "* Avg Loss: 5.3298557344247174e-08\n",
      "Sample Number 250: \n",
      "Input : [[[(()([[(())]]))]]]\n",
      "Output: 22211112211122112220\n",
      "* Counter: 250\n",
      "* Avg Loss: 6.240215271955903e-08\n",
      "Sample Number 500: \n",
      "Input : ([(()(()))])\n",
      "Output: 121111111210\n",
      "* Counter: 500\n",
      "* Avg Loss: 5.6709140709256964e-08\n",
      "Sample Number 750: \n",
      "Input : [(([[]])[[][]])]\n",
      "Output: 2112221122222120\n",
      "* Counter: 750\n",
      "* Avg Loss: 5.237454204366951e-08\n",
      "Sample Number 1000: \n",
      "Input : [[[[]][[]]]]\n",
      "Output: 222222222220\n",
      "* Counter: 1000\n",
      "* Avg Loss: 4.892151193153148e-08\n",
      "Sample Number 1250: \n",
      "Input : [([[((()))]])]\n",
      "Output: 21221111122120\n",
      "* Counter: 1250\n",
      "* Avg Loss: 4.576648886522762e-08\n",
      "Sample Number 1500: \n",
      "Input : [(([([])([[[[[(())[]]]]]])]()[()]))()][([()]([]))]\n",
      "Output: 21121212122222111222222212111212112120212121121120\n",
      "* Counter: 1500\n",
      "* Avg Loss: 4.301662582703498e-08\n",
      "Sample Number 1750: \n",
      "Input : (([([])]))(())\n",
      "Output: 11212121101110\n",
      "* Counter: 1750\n",
      "* Avg Loss: 4.057228904373591e-08\n",
      "Sample Number 2000: \n",
      "Input : ([[[]]])((([()](()))))()[([])]\n",
      "Output: 122222101112121111111010212120\n",
      "* Counter: 2000\n",
      "* Avg Loss: 3.829019938325839e-08\n",
      "Sample Number 2250: \n",
      "Input : [((([()[[](((([([])]))))]])))]\n",
      "Output: 211121222211112121211112211120\n",
      "* Counter: 2250\n",
      "* Avg Loss: 3.618537741019477e-08\n",
      "Sample Number 2500: \n",
      "Input : ()[[]]()\n",
      "Output: 10222010\n",
      "* Counter: 2500\n",
      "* Avg Loss: 3.430439239470401e-08\n",
      "Sample Number 2750: \n",
      "Input : ([(([]))])([[]])\n",
      "Output: 1211211210122210\n",
      "* Counter: 2750\n",
      "* Avg Loss: 3.257674582338281e-08\n",
      "Sample Number 3000: \n",
      "Input : [[(([]))[]([[([[([])]])]()])()()]]\n",
      "Output: 2211211222122122121221221212121220\n",
      "* Counter: 3000\n",
      "* Avg Loss: 3.095445052796526e-08\n",
      "Sample Number 3250: \n",
      "Input : [[[([()])]][(())]]\n",
      "Output: 222121212222111220\n",
      "* Counter: 3250\n",
      "* Avg Loss: 2.9531390106762183e-08\n",
      "Sample Number 3500: \n",
      "Input : (((([]))))[([[]][])]\n",
      "Output: 11112111102122212120\n",
      "* Counter: 3500\n",
      "* Avg Loss: 2.8123828660506034e-08\n",
      "Sample Number 3750: \n",
      "Input : ([[]()][[(((())))]])([()]([[()]])([[]]))\n",
      "Output: 1222121221111111221012121122122111222110\n",
      "* Counter: 3750\n",
      "* Avg Loss: 2.6837025446782535e-08\n",
      "Sample Number 4000: \n",
      "Input : [([[([[([])]])((([])())())]]([][[]])())]\n",
      "Output: 2122122121221211121111111221121222111120\n",
      "* Counter: 4000\n",
      "* Avg Loss: 2.5653821416002072e-08\n",
      "Sample Number 4250: \n",
      "Input : ()([])([((([([([])[]])])))])()()\n",
      "Output: 10121012111212121222121112101010\n",
      "* Counter: 4250\n",
      "* Avg Loss: 2.4551224334006473e-08\n",
      "Sample Number 4500: \n",
      "Input : [(((([][([])]))))]\n",
      "Output: 211112121212111120\n",
      "* Counter: 4500\n",
      "* Avg Loss: 2.3531781524923968e-08\n",
      "Sample Number 4750: \n",
      "Input : ([[][[[]]]](())[(([[(())]]))([])])\n",
      "Output: 1222222222111112112211122112121210\n",
      "* Counter: 4656\n",
      "* Avg Loss: 0.0006950378123050635\n",
      "Counter: 0.9812\n",
      "Training Accuracy %:  [3961, 5000, 5000, 4906]\n",
      "Loss:  [52.354061373115464, 0.027094410170337646, 0.0015958646551512323, 3.310932254234565]\n",
      "Epoch: 5\n",
      "Sample Number 0: \n",
      "Input : [([[[]]]())]\n",
      "Output: 212222211120\n",
      "* Counter: 0\n",
      "* Avg Loss: 1.2764805468895382e-05\n",
      "Sample Number 250: \n",
      "Input : [[[(()([[(())]]))]]]\n",
      "Output: 22211112211122112220\n",
      "* Counter: 250\n",
      "* Avg Loss: 1.2725468186846562e-05\n",
      "Sample Number 500: \n",
      "Input : ([(()(()))])\n",
      "Output: 121111111210\n",
      "* Counter: 500\n",
      "* Avg Loss: 1.0830784990482081e-05\n",
      "Sample Number 750: \n",
      "Input : [(([[]])[[][]])]\n",
      "Output: 2112221122222120\n",
      "* Counter: 750\n",
      "* Avg Loss: 9.46414131232392e-06\n",
      "Sample Number 1000: \n",
      "Input : [[[[]][[]]]]\n",
      "Output: 222222222220\n",
      "* Counter: 1000\n",
      "* Avg Loss: 8.505948481545677e-06\n",
      "Sample Number 1250: \n",
      "Input : [([[((()))]])]\n",
      "Output: 21221111122120\n",
      "* Counter: 1250\n",
      "* Avg Loss: 7.740106407458854e-06\n",
      "Sample Number 1500: \n",
      "Input : [(([([])([[[[[(())[]]]]]])]()[()]))()][([()]([]))]\n",
      "Output: 21121212122222111222222212111212112120212121121120\n",
      "* Counter: 1500\n",
      "* Avg Loss: 7.071100353060916e-06\n",
      "Sample Number 1750: \n",
      "Input : (([([])]))(())\n",
      "Output: 11212121101110\n",
      "* Counter: 1750\n",
      "* Avg Loss: 6.538264039274999e-06\n",
      "Sample Number 2000: \n",
      "Input : ([[[]]])((([()](()))))()[([])]\n",
      "Output: 122222101112121111111010212120\n",
      "* Counter: 2000\n",
      "* Avg Loss: 6.083802083151011e-06\n",
      "Sample Number 2250: \n",
      "Input : [((([()[[](((([([])]))))]])))]\n",
      "Output: 211121222211112121211112211120\n",
      "* Counter: 2250\n",
      "* Avg Loss: 5.674695117365524e-06\n",
      "Sample Number 2500: \n",
      "Input : ()[[]]()\n",
      "Output: 10222010\n",
      "* Counter: 2500\n",
      "* Avg Loss: 5.324059160800788e-06\n",
      "Sample Number 2750: \n",
      "Input : ([(([]))])([[]])\n",
      "Output: 1211211210122210\n",
      "* Counter: 2750\n",
      "* Avg Loss: 5.009910457164749e-06\n",
      "Sample Number 3000: \n",
      "Input : [[(([]))[]([[([[([])]])]()])()()]]\n",
      "Output: 2211211222122122121221221212121220\n",
      "* Counter: 3000\n",
      "* Avg Loss: 4.7218837920344346e-06\n",
      "Sample Number 3250: \n",
      "Input : [[[([()])]][(())]]\n",
      "Output: 222121212222111220\n",
      "* Counter: 3250\n",
      "* Avg Loss: 4.460221318804952e-06\n",
      "Sample Number 3500: \n",
      "Input : (((([]))))[([[]][])]\n",
      "Output: 11112111102122212120\n",
      "* Counter: 3500\n",
      "* Avg Loss: 4.215796279278141e-06\n",
      "Sample Number 3750: \n",
      "Input : ([[]()][[(((())))]])([()]([[()]])([[]]))\n",
      "Output: 1222121221111111221012121122122111222110\n",
      "* Counter: 3750\n",
      "* Avg Loss: 3.999612750939929e-06\n",
      "Sample Number 4000: \n",
      "Input : [([[([[([])]])((([])())())]]([][[]])())]\n",
      "Output: 2122122121221211121111111221121222111120\n",
      "* Counter: 4000\n",
      "* Avg Loss: 3.8014186591556874e-06\n",
      "Sample Number 4250: \n",
      "Input : ()([])([((([([([])[]])])))])()()\n",
      "Output: 10121012111212121222121112101010\n",
      "* Counter: 4250\n",
      "* Avg Loss: 3.6201571246917722e-06\n",
      "Sample Number 4500: \n",
      "Input : [(((([][([])]))))]\n",
      "Output: 211112121212111120\n",
      "* Counter: 4500\n",
      "* Avg Loss: 3.4557859938726916e-06\n",
      "Sample Number 4750: \n",
      "Input : ([[][[[]]]](())[(([[(())]]))([])])\n",
      "Output: 1222222222111112112211122112121210\n",
      "* Counter: 4750\n",
      "* Avg Loss: 3.3031619112308266e-06\n",
      "Counter: 1.0\n",
      "Training Accuracy %:  [3961, 5000, 5000, 4906, 5000]\n",
      "Loss:  [52.354061373115464, 0.027094410170337646, 0.0015958646551512323, 3.310932254234565, 0.015815981013524612]\n"
     ]
    }
   ],
   "source": [
    "train (model, optim, criterion, epoch_num)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluate the Performance of the Stack-RNN Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training accuracy: (100.0, 5000).\n"
     ]
    }
   ],
   "source": [
    "# Training set accuracy \n",
    "correct_num = test_model (model, training_input, training_output)\n",
    "print ('Training accuracy: {}.'.format(correct_num))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test accuracy: (100.0, 5000).\n"
     ]
    }
   ],
   "source": [
    "# Test set accuracy \n",
    "correct_num = test_model (model, test_input, test_output)\n",
    "print ('Test accuracy: {}.'.format(correct_num))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Save/Upload the Model Weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save the model weights\n",
    "torch.save(model.state_dict(), 'models/baby_ntm_model_weights.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Load the model weights\n",
    "# model.load_state_dict(torch.load('models/baby_ntm_model_weights.pth'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualize the Memory Entires of a Baby-NTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "from plotly.graph_objs import *\n",
    "from plotly.offline import download_plotlyjs, init_notebook_mode,  iplot, plot\n",
    "import plotly.graph_objs as go\n",
    "import math\n",
    "\n",
    "import seaborn as sns; sns.set()\n",
    "\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def enable_plotly_in_cell():\n",
    "  import IPython\n",
    "  from plotly.offline import init_notebook_mode\n",
    "  display(IPython.core.display.HTML('''\n",
    "        <script src=\"/static/components/requirejs/require.js\"></script>\n",
    "  '''))\n",
    "  init_notebook_mode(connected=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get the Hidden State and Memory Configuration for a Given Input"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_hidden_and_memory_info (model, input, output):\n",
    "    # Turn on the evaluation mode for the model\n",
    "    model.eval()\n",
    "    # Hidden state values\n",
    "    hidden_states = []\n",
    "    # Memory configuration\n",
    "    memory_config = []\n",
    "    # Memory operation weights\n",
    "    operation_weights = []\n",
    "    # Most recently inserted element to the memory\n",
    "    new_elt_inserted = []\n",
    "    \n",
    "    with torch.no_grad():\n",
    "        len_input = len (input)\n",
    "        model.zero_grad ()\n",
    "        \n",
    "        # Initialize the hidden state\n",
    "        hidden = model.init_hidden()\n",
    "        # Initalize the memory configuration\n",
    "        memory = torch.zeros (memory_size, memory_dim).to(device)\n",
    "        # Target values\n",
    "        target = Dyck.lineToTensorSigmoid(output)\n",
    "        # Output values\n",
    "        output_vals = torch.zeros (target.shape).to(device)\n",
    "\n",
    "        for j in range (len_input):\n",
    "            # Feed the input to the model\n",
    "            output, hidden, memory = model (Dyck.lineToTensor(input[j]).to(device), hidden, memory)\n",
    "            # Hidden state values\n",
    "            hidden_states.append (hidden.cpu().numpy())\n",
    "            # Stack configuration \n",
    "            memory_config.append (memory.cpu().numpy())\n",
    "            # Stack operation weights\n",
    "            operation_weights.append (model.action_weights.cpu().numpy())\n",
    "            # New element inserted to the stack\n",
    "            new_elt_inserted.append (model.new_elt.cpu().numpy())\n",
    "            # Output value\n",
    "            output_vals [j] = output.view(-1)\n",
    "        \n",
    "        # Binarize the entries based on the output threshold\n",
    "        out_np = np.int_(output_vals.cpu().detach().numpy() >= epsilon)\n",
    "        target_np = np.int_(target.cpu().detach().numpy())\n",
    "        \n",
    "        # (Double-)check whether the output values and the target values are the same\n",
    "        if np.all(np.equal(out_np, target_np)) and (out_np.flatten() == target_np.flatten()).all():\n",
    "            print ('Correct!')\n",
    "        else:\n",
    "            print ('Incorrect')\n",
    "            \n",
    "    return hidden_states, memory_config, operation_weights, new_elt_inserted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_seq = '(([[]]))[[(())]]()[]'\n",
    "output_seq =  Dyck.output_generator (input_seq)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct!\n"
     ]
    }
   ],
   "source": [
    "hidden_states, memory_config, operation_weights, new_elt_inserted = get_hidden_and_memory_info (model, input_seq, output_seq)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize the Memory Operation Weights at Each Timestep"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualize_memory_operation_weights (operation_weights, input_seq, timestep=1):\n",
    "    # Stack operation labels\n",
    "    labels = ['ROTATE LEFT','ROTATE RIGHT', 'NO-MOVE', 'POP-RIGHT', 'POP-LEFT']\n",
    "    memory_op_weights = np.squeeze(operation_weights)\n",
    "    plt.figure(figsize=(16, 5))\n",
    "    fig = sns.heatmap(memory_op_weights.T, cmap=sns.light_palette(\"#34495e\"),xticklabels=input_seq, yticklabels=labels, vmin=0, vmax=1)\n",
    "    fig.set_title('Strength of Memory Operations at Each Timestep', fontsize=17)\n",
    "    cbar = fig.collections[0].colorbar\n",
    "    cbar.set_ticks(np.linspace(0,1,6))\n",
    "    plt.xlabel('Sequence', fontsize=16)\n",
    "    plt.ylabel('Actions', fontsize=16)\n",
    "    plt.xticks(fontsize=13)\n",
    "    plt.yticks(fontsize=14)\n",
    "    # plt.savefig('weight_{}.png'.format(timestep), dpi=128, bbox_inches='tight')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6kAAAFWCAYAAACLhNtWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvFvnyVgAAIABJREFUeJzs3XncpXP9+PHXWIqsiUJla3lrVUJUslRkTVKiiBBS36RCEiOiRMiSbI2iRMiSNWvI0o/KkndN2bObGfswY35/fK5jrjnOOfe5556Z+5qZ1/PxuB/3Odf1uT7X+5xz3ec+7/PZRkyaNAlJkiRJkppgjuEOQJIkSZKkFpNUSZIkSVJjmKRKkiRJkhrDJFWSJEmS1BgmqZIkSZKkxjBJlTRbi4gRwx1DN02OTTO3WenampUey8zO10LStDLXcAcgSdNSRCwP7AWsDbweeBK4CTgyMy+olXsdcDhwPHD1MITaVUQsDfwc+Cpwd7XtbuD6zPz8dDrnx4EjgeWA0Zn5rg5lRgL7AhOBJTPzkQ5lVgZurO4um5l3T494myIiPgvsAHwAeA1wP3AhcERm/mc4Y+smIt4LHAesWts2CfhxZu45bIENUkTMC/yIcr2d2qPcKOBLA1S3RGY+NA1jGwV8MjMXH8QxywB39VF0v8wc2bTXLCLWAPYE1hvuWCTN/ExSJc0yIuJdwF+AvwHfAf4HLApsDfwxInbOzGOr4isDXwROHI5YB7AeM/6D3uHAq4CNgCcGKDsC+CxwdId9W07juBopIuaiJEabAb8GjgHGActTvlzYNiK2ysw/DF+UXW0JfLBt22rAA8MQy1AsDfwfsG0fZZ8ANuix//FpEtHQPEh5HVpeBVxFeY86obb9/up3016znYF3DHcQkmYNJqmSZiW7Ac8AH8/MF1obI+Is4E/AgRFxXGa+NFwBNtjrgEsy85I+yv4Z+DxtSWpEzAF8DrgZWHGaR9gsPwA2BTbOzD/Wtl8REScBZwCnRsSKmZnDEuEgZOb1wx3DdPZi0x9jZo4HXo4xIuapbt7fKfamPx5JGgqTVEmzklbXujnrGzNzUkTsQ2l5eFVEfB74ZbX7ioi4KjPXjIgrKa2vABsCdwAfAiYB3wB2ApYFHgJ+BfwgM1+El7vCbkPp+vkj4F3Aw8DRmXlwK5aImA/4MaUFbkHgcuA0SmvcslUd+1bF74qIkzNzm+r+XBFxAKXlaBHgFmDXzGx1r+0oIt4P7A+sQumS+hfg+5l5fVsXw60jYmtg28wc1aPK3wI/j4g3Zeb9te1rUlquj6ItSY2IAA6idMOem/JhfI/M/GutzCRKy9h7KC21cwJnU1omvwp8HXgtcC3wlcy8t3bsWsA+wPsoLb1XAN/NzDur/WtW23YE9gAWonRvHgl8OzMPrdX1asp1cGJm7t7+4CPitcCuwKi2BBUoyUZEbAfcQ+n+uG3t8X2L0tq6OfAi8Efgm5n5cuv1IJ6rkcC6wLuBkzJz14h4N+X6+SjlGnkCuBjYLTMfq3d9rero2HW0eg6+SemF0LrmT6Vc8y9UZUYBy1C6Du8NvIXSPf3AzDy5Futnge9Wj/t5Svf6vTLzjvbnrnbMPFWdm1XnmAD8HdgnMy+rvZ4Av4yIkZm5TLf6BiMitqG0Cr6L8vz/FzgmM4+slZmP8kXFZynX/H+BQzPzl211bQF8ny7PzRDjfPk1q/0db0H58mQ94AVgFPA9yrWyLaV19hJgp8wcU6vri8C3Ka/RWMqXLHtl5lPV/hGUv6+tgDcDjwLnAXtm5rjqvXONWlzbZuaoiHgV5XXcCliS8jdxTGYeXjv3qOq8R1Ke0yUor/Vemdl6jSXNZpw4SdKs5DxKonpjRHwzIlaoWvfIzGsz85DMfJ4qMaiO2YWSALV8lpLIbQL8sGp1PQI4hJIwbUhpQfw2JVGtez2la97xwPrAdcCPI2KjWpmzKEnCjygfJscxZVe+E4BWUrspJbls+TQl0d6hquPNwPnVB8GOqnFiN1AS4p0oScdrgKsiYnUmdzF8Ariguv2KxKvN2cB4Sqtp3ZaU8Zhj6hsj4i2UROstwFeAL1D+/1xdJdB1B1A+SH+G8pxvDfyV8rzvTHne16R0r23VvyUl2R9L+TC8M+VD7/UR8fa2+n9CSRp2AX5BSZjaxytuQknwTury+NcC5gW6duXNzEcprfefbtv1feDtlK7mewMbAxe3rtNBPld7V+f4LPCbiHgDpZV7Mco1sg7lg/8WlOcSyvX06+r2akx57VHFMAI4l5KUnFLFeBLluf9D2+Q47wP2oyTVGwL3AqMi4j1VXR+mfAlzHaW77Q6U1+bCqst0N7+k/F3+lJKI70hJBs+MiIUorfWt8dkH8Mrn+RUiYq4uP3PUyuxYPdZLKV3fN6Mklz+rEuNWj4ELq8dyRFXuUuCkiKhfS4tVsf2weg4fpCTU07OXwS8oieCnKM/7bpQvs95Guc4PplwvL7+vRMRulGvihuq4/SjX3QUR0frCbw/KWP8jKNfV/pRruPV3+FXKtfgQU76HnEH5YuYXlOvj98ChEXFQW9zvqOo+mMlf4FwcESsN5cmQNPOyJVXSLCMzj42IxSitNj+tNj9Zfct/YmaeW5V7NCLurPbf0daiMwfwpcwcBxARbwW+Buyfma0Wzksj4iHg5Ig4otbtbl5gl8w8rzr2WkrCszFwXpUwrgNslZmnVMdcFBGvBz5exXZ/RLRaNm9pm3joUWCDKtEmIhakJMQrUCaH6uTHwH3UukBHxB+BO4FDMvODlGTuReDRPrsQPkn5kP55que5SpQ3pSQTr2srvy+lNXrtzHy8FsNtlOTmk7Wyd9Vaji+PiG2BNwGrZebY6thVqnO1EoZDgGsy8+VEJSIuA/5N+TC9ea3+4zPzN7VyJwC/ioj3Z+Yt1eYvA9e2WmE7WK4Va5f9Lf8GNoqI19ZarcYB61ZdO4mIRykf3DemJL2Dea7+npn71B7LJ6pyn2kdS+kpsAolsSYz/xMR/6tud3ut16Vcp9tnZmvM9qUR8QhwbBXDhdX2hYDVM/PWKoakJEkbA7cCH6H8TR2Ymf+rytxH+buYn/LFwhSqa2lhSi+BX9W2PwecCayYmVdExN+rXf+pvXbdvIGS+HRyKiXhAngrZdKrvWvnvZYyZnUt4Mrq+Vkd2DIzf1sVuyzKhGdrA62W0jkor8Xfqnr+Q7kmPkZJsqeHKzPzO9X5rgG2B+YBNs/MCZT3m48DH67KLEhpvfxVZu5Ye8y3U8bDfpaS7K5Bud6Prr64uyoinqZ8MUdm3hERjwPjW9dVRKxNuQ6+XGthvjQingf2iYijaz0xFqRMNHVxdexllNbpvSnXiqTZjEmqpFlKZu4fEUdQPkiuSen2uDGwcUT8BvhiZk7qUcW9rQS18jFK99Fz2lp+zqMkE+tSG0dG6YraimV8lYTMV236eHXMGW3nPLXaN5CbWglq5b/V79d2Klx1SVwFOLg+RreK63TgOxExf2Y+3ce52/0WOD0ilsvM/1K6F85JeV62aSv7CUqL5bjaczgJOB/4WkS8qhbfdW3HPgw81kpQK49RkhiAoHQP/EH9oMx8JCIuokrOam5tu/974GeUVqZbIuJNlNdi+24PnHI9QOmC2kunpOi3rQS1cnZVzxqUJHUwz9UUjyUzL6UkAXNVXYbfQumy+k5K63S/Ws/Zb9q2/4qSpK7F5CT1qVaCWmklHa1r/krgJeCmiPh9ddwVvbqoV49vPYCIWJzSCvg2Soslg3wsLY8zZYLfvq917laCtyClxfutlJmb6+f9KOU1Oast7k3b6n26laBWev69TiP1958XImIM8NcqQW15jHJtQGn1nI9Xvr9dV5Vbl5KkXkbphXBzRJxN6XXxmwHeSz9R/T63re5zKK21H2NyQv9QK0GtYn82Ii6gjxZySbMmk1RJs5zMfJKSCJ4BLy/tcDSlO+qplA9Y3Tzcdn/R6vf/61L+jW33n227/xKTh1YsBoxrS1KgdJHrxzMd6obuQzcWpiRUnep/sNq3IDA1Ser5wFOUVsqDKM/t2Zn5fMmPprAopRtht5asRZk8FvjJDvvbH3f9g/Ei1e9uj3Hhtm1TvL6Z+VxEnApsGRHfoSSrzwCnd4kVJregLgv8q0e5t1CSuHr35/oYXjLzpaoFqvU4BvNcTfFYqm64B1Ba/hekPP6bKa/v/D3ibLcI5Tp9ri3W5yJiHFM+p8+2lXmpev1b3exviIh1KV0+d6SMOR4XEcdSxhx2nMSsau07jDLe9hngdkqPAJj8JcFgTKiP6e0mIpalLP+0DmWppX8xOfFrnXdROv8dt5vium1/bqaTwf79tN7fzuxSX+v97VDKdbQ9pbV/JHBPRHwvM7st/9Oq+7EB6oa2v4vKI8DCETFHt+tE0qzLMamSZgkR8caIeCAivt6+r+oy22oZe8X6nwNoteB9krJsTfvPgYOo635goZg8a2fL6wcZU7/GUj6Qdlqrcclq31QtvVElMOcAm1ctthvSfa3KsZRWp07P38p0/xDbj9aEQ90eYz91H0/5AmFtyvjN32Vm+wf7ukuA55iyG/EUqnGT6/LKcauLtZWbk/JhvrXm7FCeq90pYwf3AF6XmUtm5ob0TqQ7eYJync7bFut8lO69g3q9MvNPmbkeJbldhzJ+cw/KuMdXqBLFcyldY5cHFqi6pXcbIzxNVEn++ZSlbT4CzJdlveBd24qOpTw/r2o7fvko47xnJq33ty/T+Xr7OpTJ5zLz2MxciXINb0G5Tn5VfQnYre4XKL05OtU9qlZ2sfaDKV20HzNBlWZPJqmSZhUPUVqfdo6ITq1GrQl0/lH9nthnvVdWvxfPzL+2fihJyo8oH6L7dQWlNaa9W+Bmbff7ja2nKtG6EfhsRMzd2h5l5tbPATf00RrUy28p42G/RWnBubxLuSspXw78o+053ILyIbhbq2E/ktJiOMX6rNXY5E9SxtX1riDz75TJmfao4uy5dm7VUn8w8KWI+Ez7/qpr44mUmWEPbtu9SdvEQ5tSukm3lv65kql/rtYA/l0lE09UsSzA5HGhLQNdX60ZVdvXvG2N2xzwOW2JiL0i4r9VN+Xnqy7JO1S7l+5y2MqU8d0/zqLV8rd+9bv1WKbJ30nNYpSu0aMy87pat+r28/6Z8ne8cdvxB1NaYWcmf6FMgrZU2/V2L6WHxAcBIuK8iDgDIDMfz8zTKF3s56CMGYdXvh5XUrpIv6at7oUpLf5L1MouVZ9QqvpCZH0m/11Ims3Y3VfSLCEzJ1Yzc55LGVt4FCUhnYOyjMw3gXNr455aXTA3iIgxVaLSqd7bI+Jk4KiIeDNlrNabKWOq5mYQE6Bk5rURcT5l+ZbXURKszSiznsLk7rut2DaNiAt6TODTj+9SPuhdFhGHVdt2o3S122YI9UJpEXucMrnJ0ZnZLWkYSZk59JKIOJLSwrI5JVn5/gDj2nqqulDuQWnROZsyW+0CVUwjKK9TP46nzEB6R48JheoOoEygdHpE/JoytvQJypchu1S/t87M29qOey/w+4g4ntId+IfApVXyBkN7rq4HPhkR36eMa10K+A6llbnedXcMvLw8yvWZ2T4B1MWUmVqPjDJj8I2UZGVvymt+Kf37E+U1ODMijqEkMjtRWtjO7nLMzZRk/MCIOJjyOm5OmekZJo93bf2dfCwi/pmZN/SIY+6IWLXH/v9U45jvAnaqfj9KmSBpD0qvg9Z5/0jpAnxcNWb2n5S/4Y3o0breRJn5RET8CPhelRheQunu/T1Kd/ZvVEWvoMzKezBlXPEilGv1LiZP2jYGeENErAf8rSp3BXBGRPyQsqzMOyh/Ow9SJvmqOzsivkf5wmt3yvP9AyTNlmxJlTTLqBLQlSitA7tSxp6ew+RlFeqtXrdQxh1+je7dVFu2oyQTWwEXUVpQr6XMavpIrwM72JLSArkvpSvo65m8HERrbOjFlFbJgyhjwaZalnUG16ZMznMKpYvdM8BHM/PKIdb9ImXiobl55SQ79XK3U74oeJLSwngeJenZKTMPGEoMVf2/pswAuiRlbN3RlC8AVh5Egn9e9btnK2rtnBMyc2vKlwxvoEwodAllXdRrgPdlZvsEWVSxPV3FuTelC+unavUO5bn6EWXJmZ0p1+k+lAlvdgBeE5OX8ziNMsb6ZEoS2/7YJlFaCY+kjCO9gNId9GBgo8F0v6wmSPoUJak5jZKYLkaZ4fj2LseMpiR7i1H+Rn5JeY4/QpkdeY2q3GOUSa8+TZm1tteESotQ3he6/bS+KNqYMsHRCZTXaAPKUIELKRMmUX0Zsx7l72kvShfhNYHNMrPXWOZGysyRlC9W1qU8lqMpMzR/NKuZzzPzp5QeExtSkvTjKcn52rXeGMdXx51DmSH9par8LylfEl5Ceb7OANZq68XxCOU98SDKe8kzwIczM6fPo5bUdCMmTZrqL7AlSYNQLVGxGnBefcxjRPwE2CYzO43L0gwQEV+hrNP4xlZX2elwjkmULqx7To/6pZlRRIyiLD/TaVy5pNmU3X0lacZ5idKqcFFEHEfpgvlhyoyndmsbBhGxFWUM6FeB46ZXgipJkvpnd19JmkEy8z7KZD4LMnkpnM8B38jMHw5nbLOxd1K+JLiCMg5PkiQNM7v7SpIkSZKGLCI+SBnasmbb9o0ocyVMAE7KzON71WNLqiRJkiRpSCJid8rEc/O0bZ8bOIyyVvYawFeq2dG7MkmVJEmSJA3Vf3jlWvBQlp8anZljqjWor6Es8dWVEydpQGOfes4+4Zoqn9h8p+EOQTOpXb/yheEO4WUbfKzn/1E1iO85mlpHH7TXcIcwhV2+e+BwhzCFJr0nH37cQKvGzXg3XXDyiOGOYTBWXv9LU/XZfqDHmZlnRsQyHXYtSFlCrOUpYKFeddmSKkmSJEmaXp4EFqjdXwAY2+sAW1IlSZIkSdPLP4G3RcQiwNPAR4FDeh1gkipJkiRJmqYiYktg/sw8LiJ2Ay6m9OQ9KTMf6HWsSaokSZIkacgy825g1er2b2rbzwPO67cex6RKkiRJkhrDJFWSJEmS1BgmqZIkSZKkxjBJlSRJkiQ1hkmqJEmSJKkxTFIlSZIkSY1hkipJkiRJagyTVEmSJElSY5ikSpIkSZIawyRVkiRJktQYJqmSJEmSpMYwSZUkSZIkNYZJqiRJkiSpMUxSJUmSJEmNYZIqSZIkSWoMk1RJkiRJUmOYpEqSJEmSGmOu4Q6gk4i4G1i6tmkSMBb4M/C1zLyvVnYBYC9gc2BJ4EHgLODAzHy8KjMK+FKPU26bmaNqdV4GrA4smZmP1bafVp2nmy2qOC/ssv/hzFy8fWNEfBI4PzM7vh4RsRPw8y513pCZq0bE8sA/u5Q5FBjdow6AX2TmTj32S5IkSdJ018gktfIt4DfV7TmBdwLHAicDawNExPzA1cAI4OvAbcBbgZHA9RGxemY+BHwD2LOqa3XgdGCJ2rnGtW5ExOLAR4F7gS2Bn9XK7QjsWt3+OHACsExt/1hgTWAi8KYOj2niwA+7q7uB1Tpsf6Ht/oqURL3uGWAC8Ifq/muA/wAbADdX254dQmySJEmSNE00OUl9skowWx6IiH2AUyJiocwcB+wPzAesmJlPV+XuiYhrgL8AhwFbVGXHAUTEGIC2uus+T2mRvJDS+vpyktpWz9hO9UQEnbZPAxP7rPPRHuWeg5eTe4AnpkOckiRJkjTVZrYxqeOr3xMjYk7gy8ARtQQVgMwcDxwEbBYRiwzyHFsAVwLnAytGxHuGFrIkSZIkqV8zTZIaEW8DfgBcVCWlbwcWBG7scsifKS3FHxjEOd4CrAKcB1wHPErvsaySJEmSpGmoyd19j4qIw6vbc1HGXp7D5DGhrRbSMV2Ob21fdBDnbE18dGVmToyIc4EvRMQemTmY8aRzRsTTHbYfmpn7DqKeuuW61PmNzDyxdj8jYlLt/j8y80NTeU5JkiRJmqGanKTuB5wBzA/sS5nt93utGXuBJ6rfi1Nmrm23ZPX78Q77utkCuCAzX6zunwVsB6wLXDCIeiYC7+uwvVtC3Y/7gI912P5I2/11gfo40/FIkiRJ0kyiyUnqo5k5GiAiNgduAv4QER+sksh/UxLQlYBrOhy/MmVG25s77HuFiFiBMoPw8tX56rZhcEkqrdinoRf7rPPuzLx/Gp9bkiRJkmaImWJMama+AGwPrADsVm2bABwP7F6tlfqyiJgb2Bs4q77O6QC2pHT1fT+lFbT1cxKwcUS8dho8FEmSJElSD01uSZ1CZt4UEScCe0fEKZn5AGUJmrWAqyNiL+B2YFnKOqnzUdZHHVBEjAA2B07NzH+07TuUMovw5pR1WvtSrbfayaNdxreOiIhPtm17ITMvr27P2a1Ol5GRJEmSNKuYKVpSa/YCXgQOAcjMZ4G1KRMqHQEkcDJwC7DyIJK3D1HGvJ7YviMz7wCuonT57decwINdfpbtcswclLVZ6z+n1/Yv063OiJhpvmyQJEmSpF4amdxk5jJdtj/G5Fl9W9uepbScjuyz7j8BI9q2Xdu+rW3/mh22nQ/M02H7Rb3q6lJ/z2My81gGbsW9s9/zVkv4DCpGSZIkSZoRZraWVEmSJEnSLMwkVZIkSZLUGCapkiRJkqTGMEmVJEmSJDWGSaokSZIkqTFMUiVJkiRJjWGSKkmSJElqDJNUSZIkSVJjmKRKkiRJkhrDJFWSJEmS1BgmqZIkSZKkxjBJlSRJkiQ1xlzDHYAkSZIkacZ423JLDXcIA7IlVZIkSZLUGCapkiRJkqTGMEmVJEmSJDWGSaokSZIkqTGcOEmSJEmSNNUiYg7gGGAFYDywfWaOru3/NrAF8BJwYGae3as+W1IlSZIkSUOxCTBPZq4G7Akc2toREQsD/wesBqwDHD5QZSapkiRJkqSh+AhwEUBmXg+sVNv3DHAPMF/189JAldndV9J0c+nvjh3uECTNRnzP0azCa7m7DT62+nCHoM4WBMbV7k+MiLkyc0J1/z7gDmBO4KCBKrMlVZIkSZI0FE8CC9Tuz1FLUNcDlgCWBZYCNomIVXpVZpIqSZIkSRqKa4H1ASJiVeDW2r4xwHPA+Mx8HhgLLNyrMrv7SpIkSZKG4mzgExFxHTAC2DYidgNGZ+a5EfFx4PqIeAm4Bri0V2UmqZIkSZKkqZaZLwE7tW2+s7Z/X2Dffuuzu68kSZIkqTFMUiVJkiRJjWGSKkmSJElqDJNUSZIkSVJjmKRKkiRJkhrDJFWSJEmS1BgmqZIkSZKkxjBJlSRJkiQ1hkmqJEmSJKkxTFIlSZIkSY1hkipJkiRJagyTVEmSJElSY5ikSpIkSZIawyRVkiRJktQYJqmSJEmSpMYwSZUkSZIkNYZJqiRJkiSpMeaakSeLiLuBpWubJgFjgT8DX8vM+2plFwD2AjYHlgQeBM4CDszMx6syo4Av9Tjltpk5qlbnZcDqwJKZ+Vht+2nVebrZoorzwi77H87Mxds3RsTywD/bNk8AHgbOAL6TmROqsg8B387MU2rHvx/YA/go8DrgAeBMYP/MfLIqs1N13Fs7nP+vwO+Bw4Hnejy+8Zk5T4/9kiRJkjRDDEdL6reAJaqfN1OSw3cDJ7cKRMT8wNXAesDXgQC2B1YBro+IVkL4jVpdn6u2LVH7+V2tzsUpyd59wJZtMe1YO2YrYHxbPX+oyk1s2976ec8Aj3nFWtm3AfsCuwC7dTsgItYFrgNeBD5VHbcL8DHgwojo+7XLzOfb4j0f+FXt/tLdj5YkSZKkGWeGtqRWnszMh2r3H4iIfYBTImKhzBwH7A/MB6yYmU9X5e6JiGuAvwCHAVtUZccBRMQYgLa66z5PadW8kNL6+rPWjrZ6xnaqJyLotL1Pj7Ydd2JErAl8Gji4vXBEzAucCJyQmV+v7bq3ah29C1iX7i27r1A/f0SMB56bysciSZIkSdNNU8akjq9+T4yIOYEvA0fUElQAMnM8cBCwWUQsMshzbAFcSWlFXDEiBmr9nN7GU7r+drI+sDgwsn1HZj5Kabm9aLpFJkmSJEnDZNiT1Ih4G/AD4KIqKX07sCBwY5dD/kxpAf7AIM7xFkpX4fMoXWgfpfdY1ukmIkZExEaUlt3fdym2GnBba+xtu8y8KzMnTa8YJUmSJGm4DEd336Mi4vDa+V8AzgF2rba1WkjHdDm+tX3RQZyzNfHRlZk5MSLOBb4QEXtk5sRB1DNnRDzdYfuhmblvj+MyIlpJ5auBhygtwkd2Kb8YMEWCGhHHAl+sbTohM1vP2XJd4pqX7omwJEmSJDXOcCSp+1Fmtp2fMoHQ0sD3aq2GT1S/FwdGdzh+yep3x1bGLrYALsjMF6v7ZwHbUcZ1XjCIeiYC7+uwvVtC3bIuJTFdGvg5cC3w48x8qUv5MZTJpOr2AQ6pbh8J1GfjvY8yoVK7sweIS5IkSZIaZTiS1EczczRARGwO3AT8ISI+WCWR/6YkoCsB13Q4fmXKWM6b+zlZRKwAvBNYvjpf3TYMLkmlFfsg3Z2Z9wOjI2Jj4BbgMeA7XcrfAOwcEQtk5lPVeR8BHgGIiGfayr/YKa5qgiRJkiRJmmkM65jUzHyBsrTMClTLsVTrhh4P7F6tlfqyiJgb2Bs4q77O6QC2pHT1fT+lFbT1cxKwcUS8dho8lL5l5p3AAcBu1TqonZxLSdS/176jWnrmFWuySpIkSdKsYDhaUqeQmTdFxInA3hFxSmY+QFmCZi3g6ojYC7gdWJYy2+18lPVRBxQRIyjrsJ6amf9o23coZRbhzYFj+423tkZru0cHMb71EEor7tER8eH2SZAy85mI+CJwTkS8HjgBeAB4L7AnZRKo4/uNWZIkSZJmFsM+u29lL+BFqjGXmfkssDZlQqUjgAROpnSTXXkQ63t+iDIO9MT2HZl5B3AVJVns15zAg11+lu23kmopnV0ps/hu3aXM5ZQZjCcBvwP+BRwN3Am8LzNPHkTckiRJkjRTGDFpkiuZqLexTz3nRSJJkiR1sPAC844Y7hgGY8uvfX+qPtv/5qj9Z9jjbEpLqiRJkiRJJqmSJEmSpOYwSZUkSZIkNcawz+4rSZIkSZox3r7c0sMdwoBsSZUkSZIkNYZJqiRJkiSpMUxSJUmSJEmNYZIqSZIkSWoj9xF7AAAfPElEQVQMk1RJkiRJUmOYpEqSJEmSGsMkVZIkSZLUGCapkiRJkqTGMEmVJEmSJDWGSaokSZIkqTFMUiVJkiRJjWGSKkmSJElqDJNUSZIkSVJjzDXcAUiSJEmSZl4RMQdwDLACMB7YPjNH1/avB+xb3b0Z2CUzJ3Wrz5ZUSZIkSdJQbALMk5mrAXsCh7Z2RMQCwE+ADTNzVeBuYNFelZmkSpIkSZKG4iPARQCZeT2wUm3fh4BbgUMj4s/Aw5n5aK/KTFIlSZIkSUOxIDCudn9iRLSGli4KrAXsAawH7BoRb+9VmUmqJEmSJGkongQWqN2fIzMnVLcfB27KzIcy82ngauB9vSozSZUkSZIkDcW1wPoAEbEqpXtvy/8D3h0Ri1atq6sCd/SqrO/ZfSMigLky8/aImB/4MbAUcGZmjhrUQ5AkSZIkzSrOBj4REdcBI4BtI2I3YHRmnhsR3wUursqenpm39aqsryQ1Ij4NnA4cDnwH+CWwIfAX4LiImD8zj5qqhyNJkiRJmmll5kvATm2b76ztPw04rd/6+u3uu3dV6XcjYgng08APMnNtYB/ga/2eUJIkSZKkbvpNUt8BnFgNft2Q0oT7+2rfX4Clp0NskiRJkqTZTL9J6lhgker2hsBdmfnv6v7ywCPTOjBJkiRJ0uyn3yT1AuDgiPg5JUn9NUBEfAM4kDJQVpIkSZKkIek3Sf0/4FJgdeBE4EfV9u0pCep3p31okiRJkqTZTV+z+2bms8DOHXatUM3kJEmSJEnSkA1mndQ5gBWA+ai1wJblUyEzr57WwUmSJEmSZi/9rpO6GmWd1CUpM/u2mwTMOQ3jkiRJkiTNhvptST2CMsPvV4H7Abv4SpIkSZKmuX6T1PcAm2bmhdMzGEmSJEnS7K3f2X3vBRacnoFIkiRJktRvkrov8IOI+MD0DEaSJEmSNHvrt7vvt4DFgRsjYgIwvm3/pMxcaJpGJkmSJEma7fSbpJ4/XaOQJEmSJIk+k9TM3G96ByJJkiRJUr8tqUTE4pRuv2tQJlF6HLgGOCIz/zd9wpMkSZIkzU76mjgpIt4K/A34CmWd1CuAR4Cdgb9X+yVJkiRJGpJ+W1IPAR4G1s7Mx1sbI2JR4BLgx8Bnpn14kiRJkqTZSb9J6trANvUEFSAzH4uIHwLHT/PIZqCIuBt4EPhQZk6qbV+T0mo8d2ZOiIgFgL2AzYElq2POAg5sf27a6m/Vc29mLt1h/2lVnZ/IzD9V20ZQWq53ApYHxgFXAftn5m1Vmf2B7YA3ZeZLbXW+lvLFwmeAp6rzdzIuMxfuFrskSZIkzUj9rpP6LPBSl30vMYixrQ22KrBDt50RMT9wNbAe8HUggO2BVYDrqzG7A1kiIlZoq/dVwLodyh4HHAAcA7wT2AB4ujrXmlWZU4ElgA91OP7TlOT0otq2N1Xl6z9v7yNuSZIkSZoh+k0urwa+HxFXZeaY1saIWAT4PqWFb2Z3D3BQRJyVmY912L8/MB+wYmY+3TomIq4B/gIcBmwxwDmuBjYG/l7btjbwT2C11oaI+BSwNbByZv6j2nwXsF1EjAdGRURk5p0RcQuwGWUSq7rNgdMz88WIaG17ODMnDBCjJEmSJA2bfltSvwO8Gbg7Iv4QEb+IiD9QEqelgN2nV4Az0E8pLY8Hd9g3J/BlykzGT9d3ZOZ44CBgsypp7+UcYKO2bZsAZ7dt2wE4t5ag1u1Hec7Xqe6fCmxadQ8GICJeR0l+Tx0gHkmSJElqlL6S1My8B3g/cAJlLObalK6iJwDvy8x/TrcIZ5xngW8A20TEh9v2vZ2y7M6NXY79M6VV+gMDnONc4AMRsQS8PO50I+APbeVW7nauzHwY+BelezLAacAbKd2OWz5DmYX52gHikSRJkqRG6XssaWY+QFkndZaVmedExPnAMRFRTzhbLaRjOhxW377oAKd4ALgZ2JAy2dQqwJjM/HetS27rfN3O1TrfolXMD0TEVZTE9IZq/+eA39QngaqMbTsPwNGZuccAcUuSJEmaBbxtuaWGO4QBdU1SI+JnwCGZeW91u5dJmfmNaRvasPk/4I7q983Vtieq34sDozscs2T1+/GI2IsyA3DLem1lz6GMSz2ezl19W+frNRHTkkw5W+9vgO8Cu0fE64E1q/jbfQCY2LZtbI/zSJIkSdIM1asldSPgROBeSlLV3ipXN4nSVXaml5l3V8vqjAR2rjb/E3gcWIlXTlAEpXvuBEpS+1fg9Nq+B4AP1u6fA+wVEa8BPgV8sUN911fneoVqFuE3MrnVFOD3wFERsWJ1rlsz844Oh//HiZMkSZIkNVnXJDUzl63dXmaGRNMcPwG2oiwB03I8paXyxMx8qrUxIuYG9gbqswI/UTuOehfbzLw1Ih6krH86X2bezCsdA1wYEStn5k1t+0YC/6O2tExmjo2ICylfJnwYOGUQj1WSJEmSGqOvMakRcTnw1cy8s8O+9wK/ysz3TevghktmvhARuwB/qm3eH1gLuLrq0ns7sCwlaZyPwbUkn1sdd1KX818cEUcDF0TEnsDllImbdqS0vG5YzSpcdypl5t/lgG26nPcNEdHe3RfgMVtYJUmSJDVBrzGpG9f2rwlsHBHv7FD048Bbpn1owyszL4uI31KtfZqZz0bE2pTldo6gLMnzMHAW8JnMfKJrZa90DrArr5zVt37+r0fEX4FdqvM9QxmH+sHMvL3DIedTumdfV01y1cn9Xba/B7itz9glSZIkaboZMWlS56GmEXE4kyffmQSM6Fiw+FFm7tVjv2ZiY596rtd4ZEmSJGm2tfAC8/bKkxrn1D9cMlWf7b+wyToz7HH26u67O3AYJTn9L7ApcEtbmYnAuPoYTUmSJEmSplaviZNeAO4BiIhlKZP1vDUz/1ltWwxYgTJeUpIkSZKkIZujz3IvUZZXOb+27QPAJcA1EbHotA5MkiRJkjT76TdJbXX73bS1ITMvAt4FzA8cMu1DkyRJkiTNbvpNUtcE9sjMv9c3Vl1/9wHWn8ZxSZIkSZJmQ/0mqSOA1/So49XTJhxJkiRJ0uys3yT1MmC/iFiqvjEi3gyMBP40jeOSJEmSJM2Gei1BU/dt4Frg3xFxO/AIsBjwbuBhYLfpE54kSZIkaXbSV0tqZt4LvJOyduq/quP+C+wBfATYZjrFJ0mSJEmajfTbkkpmPgUcARwREXMDnwK2A35CSVr3my4RSpIkSZJmG30nqQAREcD2wNbAopSuvkcBp0770CRJkiRJs5sBk9SImBf4HCU5/RDwHDAv8DXgF5n50nSNUJIkSZI02+iapEbEByiJ6ZbAfJQZfrcGrgDuB243QZUkSZIkTUu9WlJvAm4H9gFOz8wHASJioRkRmCRJkiRp9tNrdt9/UGb03RrYJSLeMWNCkiRJkiTNrromqZn5PmAF4HJgW+C2iPgrsAswqfqRJEmSJGma6blOambelpnfAd4MbEBZI3UvYARwcETsHBFvmP5hSpIkSZJmB30tQVNNkHQRcFFEzE+Z7XcryvIzP4uIazJzrekXpiRJkiSpiSJiDuAYSk/c8cD2mTm6Q5k/Audk5rG96uvZktpJZj6dmSdVSekywH7A4oOtR5IkSZI0S9gEmCczVwP2BA7tUOYAYJF+KuurJbWbzLyvOtkBQ6lHkiRJkjTT+gil5y2ZeX1ErFTfGRGbAS8BF/ZT2ZCSVM0ePrH5TsMdgmZSl/6uZ08Oqas/Xvbn4Q7hZYcfd+pwh6A++Z6jqfXV7x443CFM4ZiD9hruEKbQpPfkDT62+nCHoM4WBMbV7k+MiLkyc0JEvBvYEtiMsrzpgExSJUmSJElD8SSwQO3+HJk5obq9NfBGyqoxywAvRMTdmXlRt8pMUiVJkiRJQ3EtsBFwekSsCtza2pGZu7duR8RI4KFeCSqYpEqSJEmShuZs4BMRcR1ludJtI2I3YHRmnjvYykxSJUmSJElTrVqytH0imzs7lBvZT32DXoJGkiRJkqTpxSRVkiRJktQYJqmSJEmSpMYwSZUkSZIkNYZJqiRJkiSpMUxSJUmSJEmNYZIqSZIkSWoMk1RJkiRJUmOYpEqSJEmSGsMkVZIkSZLUGCapkiRJkqTGMEmVJEmSJDWGSaokSZIkqTFMUiVJkiRJjTHXcAcgSZIkSZox3rbs0sMdwoBsSZUkSZIkNYZJqiRJkiSpMWbZ7r4RcTdQb8ueANwHHJeZP6rKjAC+AuwELA+MA64C9s/M22p1jQK+VKvrJeAx4Hxg18x8qksMI4F92zY/B4wGRmbmWVW5NYErgLkzc0Lt+K2BHYF3AXMDtwFHZuYptTJXAtdk5t5t534r8G9gWWBN4JedYqycnJnb9NgvSZIkSTPErN6S+i1giepnOWA/4ICI2KrafxxwAHAM8E5gA+Bp4Poqcaw7s1bXm4HPAesBhw8Qw42145YAVgH+DpxWJZIdRcTPgZ8DpwMrAysAvwZ+HhHfGOCc7X5XO/+bqm2fqW0bbH2SJEmSNF3Msi2plScz86Ha/ZMjYgvgMxHxJLA1sHJm/qPafxewXUSMB0ZFRGTm+Grf8211/S8ifgbsCWzXI4YX2457KCK2AzalJMVHtB8QEetQWlDXyMw/13YdFRHPAj+JiKPrra69ZOZzlBZcIqL1mj/RFpckSZIkDbtZvSW1kwnAeGAH4Nxaglq3H7AUsE6fdQ3WRODF6vhOdgAubktQW04FVug3QZUkSZKkmclsk6RGxNwRsSkl8TyX0oX2xk5lM/Nh4F/Aql3qGhER7wd2qeoaTByvAQ4FXg38sUux1ShjYzvFNj4z7x/MOSVJkiRpZjGrd/c9KiJaY0bnBZ4FDsvMU6vJkMb0OHYMsGjt/uYRsUl1+1WUBP98YPcBYlgtIp6ubo8A5gFuBtbLzLu7HLMY8Hh9Q0Q8Vh3bsl6tpXX3iNi1rY4RA8QlSZIkSY0zqyep+wFnVLefBx7MzInV/SeAxXscuyRlxt2WPwLfrm6/CDxSjfUEICJuZ/Jswvdk5ruq27cAn6cktesAPwAOzcwre5x7DLBw27ZVqjrmBO6sfrccDxzWVn4p4LIe55AkSZKkxpnVk9RHM3N0l33XAyt12hERiwNvBG6obX66R10A61OWiYGSxLY8XzvuXxExP/DriLgrM2+gsxuAD9U3ZOZ/q9g6vWZj2mOLiB6hSpIkSVIzzTZjUjs4Btg4IlbusG8k8D/gon4ry8x7MnN09XNPj6I/oax3ekKXhBPgF8CGEdEpiX5jvzFJkiRJ0sxmVm9J7SozL46Io4ELImJP4HJgQcrSL18ENqwtPzMtzzsxIr4GXEOZeOkVS9Bk5gURcSTwp4gYCVxMmRH4k8DewH3A3dM6NkmSJEkabrNzSyqZ+XXKONMdgVuBS4BFgA8OMGZ0qOe9FjgF2C8i3tClzG7AtsBGlIT2b9X9w4F395h0SZIkSZJmWrNsS2pmLtNnuZOBkwcos81UxjCS0nW4076tancfpsNsvJl5NnD2AOdYs8v20V3qnNBpuyRJkiQ1wWzdkipJkiRJahaTVEmSJElSY5ikSpIkSZIawyRVkiRJktQYJqmSJEmSpMYwSZUkSZIkNYZJqiRJkiSpMUxSJUmSJEmNYZIqSZIkSWoMk1RJkiRJUmOYpEqSJEmSGsMkVZIkSZLUGCapkiRJkqTGMEmVJEmSJDWGSaokSZIkqTFMUiVJkiRJjWGSKkmSJElqjLmGOwBJkiRJ0swrIuYAjgFWAMYD22fm6Nr+bwKfr+5ekJn79arPllRJkiRJ0lBsAsyTmasBewKHtnZExHLAF4APAasB60TEe3tVZpIqSZIkSRqKjwAXAWTm9cBKtX33AZ/MzImZ+RIwN/B8r8pGTJo0aXoFqlnE2Kee8yKRJEmSOlh4gXlHDHcMg3Hj33OqPtuvskJ0fZwRcQJwZmZeWN2/F1guMyfUyowAfgIskJk79jqXLamSJEmSpKF4Eligdn+OtgR1HuDUqsxXB6rMJFWSJEmSNBTXAusDRMSqwK2tHVUL6jnA3zNzx8ycOFBlzu4rSZIkSRqKs4FPRMR1wAhg24jYDRgNzAmsAbw6Itaryn83M//SrTKTVEmSJEnSVKsmRNqpbfOdtdvzDKY+u/tKkiRJkhrDJFWSJEmS1BgmqZIkSZKkxjBJlSRJkiQ1hkmqJEmSJKkxnN1XkiRJkmYTb19uqeEOYUC2pEqSJEmSGsMkVZIkSZLUGCapkiRJkqTGMEmVJEmSJDWGSaokSZIkqTFMUiVJkiRJjWGSKkmSJElqDJNUSZIkSVJjmKRKkiRJkhrDJFWSJEmS1BgmqZIkSZKkxphruAMYThFxN7B0bdME4D7guMz8UVVmBPAVYCdgeWAccBWwf2beVqtrFPClWl0vAY8B5wO7ZuZTXWIYBcyVmV/ssv9KYI0uD+GbmXl4RIwE9u1SZmXgkB51AKyVmVf22C9JkiRJM4QtqfAtYInqZzlgP+CAiNiq2n8ccABwDPBOYAPgaeD6iFizra4za3W9GfgcsB5w+BBjPLxWb/3nuFqZG7uU+Ruwae3+t4D728pcN8T4JEmSJGmamK1bUitPZuZDtfsnR8QWwGci4klga2DlzPxHtf8uYLuIGA+MiojIzPHVvufb6vpfRPwM2BPYbggxPtNWbycv9ijzROtG9Zgm9lGfJEmSJM1wtqR2NgEYD+wAnFtLUOv2A5YC1umzLkmSJEnSAExSayJi7ojYlJJ4nksZz3ljp7KZ+TDwL2DVLnWNiIj3A7tUdUmSJEmSBmB3XzgqIlpjRucFngUOy8xTq0mNxvQ4dgywaO3+5hGxSXX7VZQvAc4Hdh9ijLtHxK4dtr8xM8dVt1eLiKfb9u+TmT8d4rklSZIkaYYxSS3dds+obj8PPJiZE6v7TwCL9zh2SeCK2v0/At+ubr8IPJKZz7V2RsTtTJ5N+J7MfFefMR4PHNZhe33G4FuAz7ftf6zP+iVJkiSpEUxS4dHMHN1l3/XASp12RMTiwBuBG2qbn+5RF8D6wNzV7RcHEeOYAeqFMmnTQGUkSZIkqdFMUns7BrgwIlbOzJva9o0E/gdc1G9lmXnPNIxNkiRJkmY5Jqk9ZObFEXE0cEFE7AlcDiwI7Ah8EdiwtvzMUCwREZ9s23ZPZv6zuj1f1XLb7vnMHDsNzi9JkiRJjeDsvgPIzK9TxpnuCNwKXAIsAnwwM6+cRqdZG7iw7WeX2v5dgQc7/Jwwjc4vSZIkSY0wYtKkScMdgxpu7FPPeZFIkiRJHSy8wLwjhjuGwZjaz/Yz8nHakipJkiRJagyTVEmSJElSY5ikSpIkSZIawyRVkiRJktQYJqmSJEmSpMYwSZUkSZIkNYZJqiRJkiSpMUxSJUmSJEmNYZIqSZIkSWoMk1RJkiRJUmOYpEqSJEmSGsMkVZIkSZLUGCapkiRJkqTGMEmVJEmSJDWGSaokSZIkqTFMUiVJkiRJjWGSKkmSJElqDJNUSZIkSVJjzDXcAUiSJEmSZl4RMQdwDLACMB7YPjNH1/bvAOwITAAOyMzze9VnS6okSZIkaSg2AebJzNWAPYFDWzsiYnHg/4APA+sCB0XEq3tVZpIqSZIkSRqKjwAXAWTm9cBKtX2rANdm5vjMHAeMBt7bqzK7+2pACy8w74jhjkGSJEnS0E2nz/YLAuNq9ydGxFyZOaHDvqeAhXpVZkuqJEmSJGkongQWqN2fo0pQO+1bABjbqzKTVEmSJEnSUFwLrA8QEasCt9b23QisHhHzRMRCwDuA23pVNmLSpEnTK1BJkiRJ0iyuNrvve4ERwLaUpHV0Zp5bze77FUoj6YGZeWav+kxSJUmSJEmNYXdfSZIkSVJjmKRKkiRJkhrDJWg0Q0XEZsAqmbn7cMcCzYknIpYB7gKeAXbLzOOM5eV4LgTWAm7LzJUGKj+7xFLFszBwLrBOZj5vPK+IZxmadS03Kh5ozntgSxPiaeB13LR4GvM+2KRY6ppwHVdxNO3aWYaGvAc29drRZLakaoapZvM6APjhcMcCzYunsngTPrhWGhFLZq4H7DTccUCzYgHIzLHAmcDewx0LNC+emkZcyzWNiKdp74FNiadp13ED42nM+2CTYmlpynUMzbt2aob9PbCJ146mZJKqGWln4PLMHDdgyRmjafFIU2MUsGP1jXkTjKJZ8ai7pr0HNimeUTTrOh5Fs+JRd026jsFrRzMpk1TNSNsBvx/uIGqaFo80aNUHoRuAzw13LNC8eNRT094DGxNP067jpsWjnhpzHYPXjmZeJqmaISJiCeCtwE3DHQs0Lx5piP4KrPH/27v/mCvLOo7jbyZrabnFj1Ca5bA/vrWx0qZJmzjJWRsLhZoL/rBamm1lWGtpCakpoc3U5qbCP8EfLXPZwB8BYWKrVkG0ZbHGF2mSsWpYNqm5GcrTH9d11t15+HEeoedcD71fGzuHc9/nvj6Hfbkfvlz3ue5hh+hoLY/6tHYObC1P1Vodt5ZHfRqtY7B2NAHZpGq8nAH8MzP/MewgVWt5pGPxZ0pNt6K1PBqttXNga3mgvTpuLY9Ga7GOwdrRBGSTqvEyQlv11loe6VhMBl4ZdoiO1vJotNbOga3lgfbquLU8Gq3FOgZrRxNQi3+RdGJ6FjilrnrXgtbySMdiJrB32CE6Wsuj0Vo7B7aWB9qr49byaLQW6xisHU1ANqkaF5m5D9gBzBl2Fmgvj3SMzgd+OOwQHa3lUZ/WzoGt5alaq+PW8qhPo3UM1o4mIJtUjadvAYuGHaKjtTzSmEXENOCdlBu2D11reXRErZ0Dm8nTWh23lkdH1Ewdg7WjicsmVePpPuCShu7V1Voe6dW4Elhdb9regtby6PBaOwe2lKe1Om4tjw6vpToGa0cTlE2qxk1d7e4GYPmws0B7eaq/RMTVww5RNZElIjYCq4adA9rKAlD/EbQQWDnsLNBeno4marmjiTytnQNbydNaHTeYp5nzYEtZelqpY2ivdjqGfg5ssXb03yaNjIwMO4MkSZIkSYAzqZIkSZKkhtikSpIkSZKaYZMqSZIkSWrG5GEHkCRpooiIi4HrgHcDJwN7gO8Bt9cFUyRJ0jFy4SRJkgYQEfOBR4E1wHrgReAc4EvALmBuZr4yvISSJJ0YnEmVJGkwXwA2Z+ZVnde2RMRO4DHg/cCGoSSTJOkEYpMqSdJgZgB7D/H6ZmBZb1tEzAC+DnwAeA2wBbg2M5/pvSEiFgC3AgH8GrgJ+AFwXmZuj4i1wLmZObvznoXAOmBWZu6pr10CrADeAfwN+Cbwld6MbkTsAe4DZgEfpvzcXwdc07s8OSJOAq4HrgRmUmaFb87M9Z2xlwKfAd4C7AZuycwHx/bHJ0nSYFw4SZKkwWwE3hcRj0bE4og4HSAzD2Tmysz8TUScDDwJXEBp6q4ATgd+HBFTACJiLuVy4V3AIuBHwANjDVO/H7sReKYe5w7g88A9fbveAEwBFgPLgSX1seduSpO8BlgAbAUeiogL6jg3AXcC36nbHwceiIjLx5pZkqRBOJMqSdJglgFTgY9SZkmpl/o+BNyVmX8HPkKZHZ2dmTvrPk8Af6A0rbdQLhveDSzOzIPAptrAfnKMeVYAv8jMxfX3myLieWBtRNzRm22lzPAuycwRYHNEXATMB66PiKnApyizryvq/k9ERAAXRsQO4IvA1zLzy3X75og4Fbgd+O4YM0uSdFTOpEqSNIDMfCkzPw6cSWns1gGnUWYld0TELGAe8DSwOyImR8RkygJLPwEurod6D7ChNqg93x5Llog4hbLC8GO9cepYmyg/2+d1dt9WG9SevcDr6vPzgZMoC0J1P+tFmbkSmAO8Fvh+3zgbgbPqZ5Yk6bhyJlWSpDHIzL3A/cD9tWG7AlgN3AxMA94GHDjEW5+uj28Anuvb9qcxxphCaUZvq7/6zew8f7Fv20H+85/UU+vjvsOMM60+/uww22dSLjeWJOm4sUmVJOkoImIO8DBwaWZu7b2emS8DayLiUuDtwLPAU8BVhzjMS/Xxr5RFmLqm9v1+hNFXO72+83x/fVxRc/UbtOl9oT6+sfueiDgbmNTZvohDLxqVA44jSdLAvNxXkqSj2wWcCizt31BXxz0L2AH8lLKS7p7M3J6Z24FfAZ+jfo+VslDSgjoL2zO/77D7gdMiovtzem7vSV2Z9yngrb1x6lj/osysvnnAz7UNeLmTrWc15buzWymzwjP6xpkN3EhpZCVJOq6cSZUk6Sgy8/mIWAbcFRHTgbWUmcU3URY8OgP4IOUy3qXA4xFxG/A8cDXwIcrKuFBuPfNLYH1E3Etp+K7rG3JjPc69EfEg8F5gYd8+N9ZjvED5fux0yszqQeC3A36ufRGxClgeEQcoDfXlwNnApzPzuYi4B7izLu60rW77KvBwZu4/3LElSXq1nEmVJGkAmXk3pdEcodzmZQvwDeCPlHua/r42bRcCO4FVlEtxzwQuy8wN9Ti/ozSd0ym3ollCaTi7Y22irCZ8GbABOIeyqnB3n0fq9nOBR2qWnwPzMrP/e6hH8lnKSr3XUBZQehcwv86YQmmgbwU+QVmY6do61sfGMIYkSQObNDIycvS9JEnS/0y9LcyTwHmd5lCSpP9LzqRKkiRJkpphkypJkiRJaoaX+0qSJEmSmuFMqiRJkiSpGTapkiRJkqRm2KRKkiRJkpphkypJkiRJaoZNqiRJkiSpGTapkiRJkqRm/BvFkPi/lQDCmQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1152x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_memory_operation_weights (operation_weights, input_seq)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize the Memory Configuration at Each Timestep"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualize_memory_configuration (memory_config, input, dimension=0):\n",
    "    memory_bound = 5\n",
    "    first_half = np.round(np.array(memory_config)[:, :memory_bound+1, dimension], decimals=3)\n",
    "    memory_config = np.concatenate((first_half, np.round(np.array(memory_config)[:, -(memory_bound+1):, dimension], decimals=3)), axis=1)\n",
    "    location = np.concatenate((np.arange (1, memory_bound+2), np.arange (memory_size-(memory_bound), memory_size+1))) \n",
    "    plt.figure(figsize=(34, 20))\n",
    "    fig = sns.heatmap(memory_config.T, cmap='viridis', yticklabels = location, xticklabels=input, annot=True, cbar=False)\n",
    "    fig.invert_yaxis()\n",
    "    fig.set_title('Memory Entries at Each Timestep', fontsize=17)\n",
    "    plt.xticks(fontsize=13)\n",
    "    plt.yticks(fontsize=13)\n",
    "    plt.xlabel('Sequence', fontsize=16)\n",
    "    plt.ylabel('Memory Location', fontsize=16)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2448x1440 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_memory_configuration (memory_config, input_seq, 2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize the Hidden State Values at Each Timestep"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualize_hidden_states (hidden_states, input):\n",
    "    plt.style.use('default')\n",
    "    domain = list(range(len(input)))\n",
    "    hidden_states = np.squeeze(hidden_states).T\n",
    "    for i in range (int(n_hidden/2)):\n",
    "        plt.figure()\n",
    "        for j in range (2):\n",
    "            plt.plot (domain, hidden_states[i*2+j], label='Unit {}'.format(i*2+j+1))\n",
    "        plt.legend (loc='upper right')\n",
    "        plt.xticks(domain, input_seq) \n",
    "        plt.title ('Analysis of the Hidden State Dynamics')\n",
    "        plt.ylabel ('Activation Values')\n",
    "        plt.ylim (-1.15, 1.15, 10)\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_hidden_states (hidden_states, input_seq)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Q.E.D."
   ]
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}