supplementary_notebook-full.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "BKYqKPb26eeG"
   },
   "source": [
    "# A neural-network framework for modelling auditory sensory cells and synapses\n",
    "\n",
    "Python notebook for comparing the evaluation results of the two additional CNN models to the referebce Dierich et al. IHC model and the Zilany et al. ANF model. The Dierich2020 CNN IHC model is evaluated using the BM outputs of the reference Verhulst2018 cochlear model (generated here)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "BKYqKPb26eeG"
   },
   "source": [
    "## Prerequisites\n",
    "\n",
    "- To run the full notebook for the first time, you'll have to compile the cochlea_utils.c file that is used for solving the TL model of the cochlea. This requires some C++ compiler which should be installed beforehand (more information can be found here). In a terminal, run from the Verhulstetal2018 folder:  \n",
    "For Mac/Linux: ` gcc -shared -fpic -O3 -ffast-math -o tridiag.so cochlea_utils.c `  \n",
    "For Windows: ` gcc -shared -fpic -O3 -ffast-math -o tridiag.dll cochlea_utils.c `\n",
    "\n",
    "- If running on google colab: add the following as a code block and run it to compile cochlea_utils.c in the runtime machine:  \n",
    "` !gcc -shared -fpic -O3 -ffast-math -o tridiag.so cochlea_utils.c `\n",
    "\n",
    "- The Matlab engine for python needs to be installed (see the instructions at the bottom of the `README` file). \n",
    "\n",
    "- The Zilany2014 model needs to be compiled in Matlab by running the function mexANmodel.m, found under `Zilany2014/reference/` (see also the instructions on the model's readme file)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "vhwtnExCCXmU"
   },
   "source": [
    "## Import required python packages and functions\n",
    "Import required python packages and define the necessary parameters.\n",
    "**Notice that for all the simulations, the reference models operate at 100kHz and the CNN models operate at 20kHz.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 431
    },
    "colab_type": "code",
    "id": "WMcdw9aWCe0l",
    "outputId": "e6f8453e-4636-4f25-9a12-05e5aba07ce8",
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import scipy.signal as sp_sig\n",
    "import numpy as np\n",
    "import keras\n",
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from Verhulstetal2018.get_tl_vbm import tl_vbm\n",
    "from extra_functions import *\n",
    "from time import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define model specific variables\n",
    "ds_rate = 2 # downsampling rate of the frequency channels of the reference model\n",
    "fs_connear = 20e3\n",
    "fs_tl = 100e3\n",
    "ds_factor = int(fs_tl / fs_connear)\n",
    "p0 = 2e-5 # calibrate to 2e-5 Pascal\n",
    "# load CFs \n",
    "CF_connear = np.loadtxt('connear/cf.txt')*1e3\n",
    "# scaling values for the CoNNear models\n",
    "cochlea_scaling = 1e6\n",
    "ihc_scaling = 1e1\n",
    "an_scaling = 1e-2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# To run this notebook and the reference models, the Matlab engine needs to be succesfully installed in python\n",
    "import matlab.engine\n",
    "import os"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "NqnSf4-8SUXE"
   },
   "source": [
    "## IHC stage\n",
    "The input dimensions for the CNN models are (b x L x Ncf), where b is the batch-size (for loading multiple stimuli simultaneously), L is the input length (including the context) and Ncf are the frequency channels.\n",
    "\n",
    "The ihc.json file can be loaded for the full-channel model (201 channels) or the ihc_1cf.json file for the 1-channel model. At the first block of each individual section, the necessary model is loaded and the rest of the parameters are defined.\n",
    "\n",
    "The BM outputs extracted from the TL cochlear model of Verhulst et al. were given as inputs to the Dierich2020 model and its CNN counterpart."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "4otlWg_jS68e"
   },
   "source": [
    "### IHC excitation patterns\n",
    "Compare the simulated average IHC receptor potentials across CF for tone stimuli presented at levels between 0 and 90 dB SPL.\n",
    "**You can change the `f_tone` variable to have tone stimuli of different frequencies, say 500Hz, 1kHz, 2kHz, etc..**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\Me\\.conda\\envs\\evaluation_env\\lib\\site-packages\\tensorflow_core\\python\\ops\\resource_variable_ops.py:1630: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "If using Keras pass *_constraint arguments to layers.\n",
      "Model: \"ihc_model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "bmm_input (InputLayer)       (None, 2560, 201)         0         \n",
      "_________________________________________________________________\n",
      "model_2 (Model)              (None, 2048, 201)         1317505   \n",
      "=================================================================\n",
      "Total params: 1,317,505\n",
      "Trainable params: 1,317,505\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "modeldir = 'Dierich2020/' # Dierich et al. CNN model directory\n",
    "# Define the IHC model hyperparameters\n",
    "context_left = 256\n",
    "context_right = 256\n",
    "Nenc = 3 # number of layers in the encoder - check for the input size\n",
    "# Load the 201-channel IHC model to simulate for all CFs\n",
    "N_cf = 201\n",
    "ihc = load_connear_model(modeldir,json_name=\"/ihc.json\",weights_name=\"/ihc.h5\",name=\"ihc_model\")\n",
    "ihc.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "yBGus8H8TFL5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "running human auditory model 2018: Verhulst, Altoe, Vasilkov\n",
      "cochlear simulation: done\n",
      "WARNING:tensorflow:From C:\\Users\\Me\\.conda\\envs\\evaluation_env\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:422: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Define the pure tone stimulus\n",
    "f_tone = 1e3 # frequency of the pure tone\n",
    "L = np.arange(0., 91., 10.) # levels from 0 to 90dB SPL\n",
    "stim_dur = 102.4e-3 # duration of the stimulus - 102.4 ms correspond to 2048 samples (fs = 20 kHz)\n",
    "silence_left = 12.8e-3 # silence before the onset of the stimulus - 12.8 ms correspond to 256 samples (context)\n",
    "silence_right = 12.8e-3 # silence after the stimulus - 256 samples\n",
    "win_dur = 5.0e-3 # 5ms long hanning window for gradual onset\n",
    "\n",
    "# make stimulus\n",
    "t = np.arange(0., stim_dur, 1./fs_tl)\n",
    "stim_sin = np.sin(2 * np.pi * f_tone * t) # generate the pure tone\n",
    "# apply hanning window\n",
    "winlength = int(2*win_dur * fs_tl)\n",
    "win = sp_sig.windows.hann(winlength) # double-sided hanning window\n",
    "stim_sin[:int(winlength/2)] = stim_sin[:int(winlength/2)] * win[:int(winlength/2)]\n",
    "stim_sin[-int(winlength/2):] = stim_sin[-int(winlength/2):] * win[int(winlength/2):]\n",
    "total_length = int(silence_left * fs_tl) + len(stim_sin) + int(silence_right * fs_tl)\n",
    "stim = np.zeros((len(L), total_length))\n",
    "stimrange = range(int(silence_left * fs_tl), int(silence_left * fs_tl) + len(stim_sin))\n",
    "for i in range(len(L)):\n",
    "    stim[i, stimrange] = p0 * 10**(L[i]/20) * stim_sin / rms(stim_sin) # calibrate \n",
    "\n",
    "############ Verhulstetal ############\n",
    "# get the cochlear outputs of the reference model\n",
    "output = tl_vbm(stim, L)\n",
    "\n",
    "CF_tl = output[0]['cf'][::ds_rate] # center frequencies\n",
    "window_length = int(total_length / ds_factor) # window length of the downsampled outputs\n",
    "tl_target = np.zeros((len(L), window_length * ds_factor, N_cf)) # pre-allocate\n",
    "ihc_target = np.zeros((len(L), window_length, N_cf)) # pre-allocate\n",
    "for i in range(len(L)):\n",
    "    tl_target[i,:,:] = output[i]['v'][:,::ds_rate] # cochlear output\n",
    "del output # remove the output variable to free-up some memory\n",
    "\n",
    "############ Dierichetal ############\n",
    "parent_dir = os.getcwd() # get current directory \n",
    "os.chdir(\"./\" + modeldir + \"reference/\") # navigate to the model's folder\n",
    "eng = matlab.engine.start_matlab() # start the matlab engine\n",
    "\n",
    "eng.workspace['Fs'] = fs_tl # pass the variables to the matlab workspace\n",
    "magic_constant = 0.118\n",
    "tl_in = magic_constant*tl_target\n",
    "\n",
    "for i in range(len(L)):\n",
    "    for fi in range(N_cf):\n",
    "        v = tl_in[i,:,fi].tolist()\n",
    "        v = matlab.double(v)\n",
    "        eng.workspace['v'] = v\n",
    "        ihc_out = eng.eval('IHC_model(v,Fs);', nargout=1) # simulate the IHC output of the reference model\n",
    "        ihc_out = np.array(ihc_out[0])\n",
    "        ihc_target[i,:,fi] = sp_sig.resample_poly(ihc_out, fs_connear, fs_tl)\n",
    "        \n",
    "eng.quit()\n",
    "os.chdir(parent_dir) # go back to the original folder\n",
    "    \n",
    "# remove context from the output\n",
    "ihc_target = ihc_target[:, context_left:-context_right,:]\n",
    "# downsample to 20 kHz\n",
    "tl_target = sp_sig.resample_poly(tl_target, fs_connear, fs_tl, axis=1)\n",
    "# apply proper scaling for feeding to the CoNNear\n",
    "tl_target = tl_target * cochlea_scaling # scaling for the cochlear input\n",
    "\n",
    "# compute the mean across CF for each level\n",
    "ihc_target_mean = np.mean(ihc_target,axis=1)\n",
    "\n",
    "############ CNN model #################\n",
    "# check the time dimension size\n",
    "if tl_target.shape[1] % 2**Nenc: # input size needs to be a multiple of 8\n",
    "    Npad = int(np.ceil(tl_target.shape[1]/(2**Nenc)))*(2**Nenc)-tl_target.shape[1]\n",
    "    tl_target = np.pad(tl_target,((0,0),(0,Npad),(0,0))) # zero-pad\n",
    "# simulate\n",
    "ihc_connear = ihc.predict(tl_target)\n",
    "ihc_connear = ihc_connear / ihc_scaling # scaling for the IHC output\n",
    "\n",
    "# compute the mean across CF for each level\n",
    "ihc_connear_mean = np.mean(ihc_connear,axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZAAAAEWCAYAAABIVsEJAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAACCl0lEQVR4nO2dd3xb1fXAv1d7ee8RrySOs/eCAAkBwl4FwiwUSqC7tBQo9Fegg9HSMkppgULbEEbDHiXQTEYgCdl7x9uO99bW/f0h2bEdD8mWLMd+33xeLL13x9HT0zvvnHvuuUJKiYKCgoKCQqCowi2AgoKCgsKpiaJAFBQUFBT6hKJAFBQUFBT6hKJAFBQUFBT6hKJAFBQUFBT6hKJAFBQUFBT6hKJAFPxGCPF3IcT/9aP+HiHEfD/KSSHEqL72o+Af/p5nIcR8IUTxQMikcGqhKBAFAIQQ+UIIqxCiUQhRJ4T4SghxpxCi7RqRUt4ppfxtX/uQUo6XUq7rp5wXCSG+9MlYLoR4UQgR0e64XgjxshCiwXf8Z+2O5Qoh3hdCVAohaoQQnwohxnRq/y5fvXpfO/oeZGm7AQshHhJCLOupTDsZ3hRCVPn62CmE+JkQQt2f8zLYEUIsEkJ87ru+KoUQnwkhLvUdu8V3nn7RqU5x6wOH7/xKIcTV7Y5rfPuyBvCjKLRDUSAK7blEShkBZAKPAfcCL/W3USGEpr9ttCMK+B2QCowF0oE/tjv+EDAa72dYANwjhDjfdywa+AAYAyQBm4D328m5CLgPWAhkATnAw8ESXAgxEtgIFAETpZRRwNXADCCip7qnMkKIq4A3gaV4v68k4NfAJe2K1QD3CiEie2iqBvjNUFe2pxRSSmVTNoB84JxO+2YBHmCC7/2/gN+1O34xsB2oA74CJnVq715gJ2AHNO37ANTA/cARoBHYAozwHZPAncAhoBb4KyC6kftKYFe79yXAee3e/xZ4o5u6sb6+4nzvXwMeaXd8IVDewzmTwCjf64eAZb2UWQb8N4DvZD5QDNwDVABlwOXAhcBBvDfU+9uV1wNPAaW+7SlA3+74L3xtlAK3dpJNDzwBFALHgb8DxvZy9PG6Er42f9FDmVuAL4EPgQfb7S8G5rc7v68CO4Cbffs0vs+QFe7fz3DdFAtEoVuklJvw/ojP6HxMCDENeBm4A4gDngc+6OTyuQ64CIiWUro6NfEz3/ELgUi8N7SWdscvBmYCk4FrgEXdiHkmsMcnUwxey2RHu+M7gPE91C2XUlb73o/vom6SECKum/qBcg7wVoB1kgEDkIb3qf1F4EZgOt7v5ddCiBxf2QeAOcAUvOdtFvArAJ8VdjdwLl4L7ZxO/TwO5PrqjmrXX38ZA4zAv8/9f8BdQojYbo5LX5kHhRDaIMim0E8UBaLQG6V4n9Q7czvwvJRyo5TSLaX8N15LY067Ms9IKYuklNYu6n8X+JWU8oD0sqPdjRzgMSllnZSyEFiL98bWASHEucDNnLjRWXx/69sVq6cL95AQIh2vZfOzdrstXdSlq/rdcI1vbKZt63Q8Dq8FEAhO4PdSSifwBhAPPC2lbJRS7sGrPCf5yt4A/EZKWSGlrMTrfrupVTbgn1LK3VLKZrxP9AAIIQTe7/MuKWWNlLIReAS4NkBZu6JV+fb6uaWU24H/4bVcuyvzAVCJ9/pRCDOKAlHojTS8rpLOZAI/73SzHIHXAmilqId2R+B1X3VHebvXLZxQDgAIIebgdTldJaU86Nvd5Pvb3o8eiddF1r5uAt4b1XNSytfbHWrqoi6d6/fAcilldPut0/FqIMXPttrqSCndvtetivh4u+NWTpybVKCg3bECTnwfqXT8PtqXSwBMwJZ23+Unvv09IoS4XwjR5Nv+3pX8vr/+fu5fA98TQiT3UOZXeK0tg59tKoQIRYEodIsQYiZeBfJlF4eL8D4Zt79hmjrdkHtK9VwEjOyjXFPxDobfKqVc3daZlLV4n3Qntys+GZ+Ly1c3Bq/y+EBK+ftOTe/pou7xTpZRf1gFfCtIbXVFKV7F3kqGbx94z8uITsdaqcKriMa3+y6jpJQdlHZXSCkfkVJafNudXRQ5gPe79utzSyn3A+/gHR/rrsxK4DDwfX/aVAgdigJROAkhRKQQ4mK8LpNlUspdXRR7EbhTCDFbeDH7Qmz9dff8A/itEGK0r/4kf8YahBAT8D4d/0hK+WEXRZYCvxJCxAgh8vC6Zv7V+rmAT4H1Usr7uql7mxBinE/R/Kq1bpB4EDhNCPHH1idsIcQoIcQyIUR0ENp/He9nTxBCxON9mm8NLV4O3OL7bCafLABIKT14v88nhRCJPrnSfFFp/UJKKfG6Cf9PCPEd37WlEkLME0K80E21h4Hv4I2a644H8AYXKIQRRYEotOdDIUQj3ifGB4A/4/0hn4SUcjPem/OzeCOlDuONpvGXP+O9qf0PaMAbLmz0o97P8bpWXmrnOtnT7viDeF1jBcBnwB+llJ/4jl2Bd2D+O+3qNgkhMnyf6RPgD3jHXAp824MECSnlEWAu3hDhPUKIeuBtYDP+u8l64ne+tnYCu4Ctvn1IKVfgjcpag/e7WtOp7r2+/RuEEA14raUxBAEp5VvAYryBEqV4XXC/o10Idafyx4BXAHMPba7HG4atEEaE9wFBQUFBQUEhMBQLREFBQUGhTygKREFBQUGhTygKREFBQUGhTygKREFBQUGhTwQzyV1YiY6OlqNG9T8DeHNzM2Zzt8EfQScU/QWrzf6005e6gdTxt+xAf5+DmcF8LobC7y5Y7Q70727Lli1VUspeJ412SbiTcQVry83NlcFg7dq1QWknnP0Fq83+tNOXuoHU8bfsQH+fg5nBfC6Gwu8uWO0O9O8O2CyVZIoKCgoKCgNJ2BWIEOJu36Iw8b735wohtgghdvn+nh1uGRUUFBQUTiasYyBCiBF400sXtttdhXdho1Jf2opP8eZjUlBQUFAYRITbAnkSbz6btunwUsptUsrWBHB7AIPoYVlRBQUFBYXwELZUJr71kBdKKX8ihMgHZkgpqzqVuQq4U0rZefGb1uNLgCUACQkJ05cvX95vuZqamrBYek1CGjRC0V+w2uxPO32pG0gdf8sO9Pc5mBnM52Io/O6C1e5A/+4WLFiwRUo5o08d9nX03Z8Nb0K23V1sl+FdGzrKVy4fiO9UdzzepHgj/elLicIKfptKFNbQYjCfi6HwuwtWu6dSFFZIx0Bk95bDRCAb2OFdDI10YKsQYpaUsty3Wty7wLelN4OpgoKCgsIgIyxjIFLKXVLKRClllpQyC++629N8yiMa+C/wS+lN2aygMGA0utwsLami2eXuvfAA4/RI3iyvocbZeXl5BYXwMBhnov8QGIV3AZr/8+07T0pZEUaZ+sUnn3zCT37yE9xuN9/97ne5776u1jJSCDd1ThfX7TjKtsYWPqmqZ+nEHDQqcVI5KSX7m238t7KeKqcLs1pFnFZDmkFLpcPFyqoG1AJ+MzqNUSb/V12VUnLEamdzfTNHWuxMjjBxRoyFKK0Gh8fDnXsK+LiqniSdhmfGZnJWrL9rdykohIZBoUB8Vkjr69/hWwRnKOB2u/nBD37AypUrSU9PZ+bMmVx66aWMGzcu3KIpcOKmvafJyl8KKjjYbOOm1DheKa3m3oNFPDFmBD43Ky1uD2+V1/BySRX7m22ogGitmiaXB0e7YJTRJj2VDhfnfHOA741IZGaUmYkRRhJ02g59V9idrK9rQgBVThdvlNWwu8m77LnAG5qoAqZFmtAIwYb6Zn6UkcgnVfUs3nGEJekJ3J+TgkEd7mBKheHKoFAgQ5lNmzYxatQocnJyALj22mt5//33FQUyCGh2ufnx/kL+W1kPgEWt4t8Ts1kQF0m8VsOTBcfZVN/MZYkxHGmxsbK6gSa3h4kWI4/lpnNRQlSbUmhwuSmxOTCpVWQa9Ry3O7nnYBFPFhxv629ShJHZUWYEgmNWO2tqGnC3C4KcYDHyyOg0zoiJINOoY1tDC+tqGllb08i+ZiuPjE7j1vQEfpaVzG+PlPJCcSWf1zbyg4xEzo2LJFqr/JwVBpZhc8U9/OEe9pY29Fqurs7K3w587Veb41IjefCS8T2WKSkpYcSIEW3v09PT2bhxo1/tK4SOAqudb+86xqFmG7/ISua8+EhyzQb0Ku/T/D3ZyWQZ9bxWVs0T+eXEatVcmhjN4uRYZkWZ26ySViI1aiItJ1bkTdJr+ffEHGqcLg402/imvplV1Q0sK61BIyBaq+F7IxK5LDEag0qFViXINOg6tDs72sLsaAv35qQgpWw7ZlSreCQ3nYVxkdxzoIgf7StEBaQatIww6FiSnsAFCdEhP4cKCsNGgYQL2cU8m843H4WBpczu4FvbD9Pk8vDG5JGc2cVYghCCxSmxLE6JpcrhIlqj7nI8pDditRrmRluYG23hx5lJfZa5q2tmYVwk38wdx/bGFtZUN5JvtbOtoYXv7M7n8sRozoqNwKhSMSPKTLpBh0d6FWeiTotRcXspBIFho0B6sxRaWbduHfPnzw1av+np6RQVFbW9Ly4uJjU1NWjtKwRGjdPF4u1HqXO6eWfqKCZFmHqtE68bvD8TlRBMizQzLdKbwtvpkTxTcJynCo7zXkVdW7k8s4EiImnesA+NgLFmIxlGHck6LefFR3FGjAWV8mCjECCD95cxRJg5cyaHDh3i2LFjpKWl8cYbb/Daa6+FW6xhw2qp478HirC6PRxu8Q6WqwS8NinHL+VxqqFVCX6encztIxKoc7pocLlZV9PI57WNpDQ7OT93JCU2BzsbrRxqtrOuppGXSqrINOjIMurRqwSjzQamRJg4Jy5SsVQUekRRICFGo9Hw7LPPsmjRItxuN7feeivjx/tnDSn0j3+WVPEiJmIr6zCr1aTptdwxIoGLE6KZEjn0lEd7IjVqIjVqACZEmPhhZhLr1hUzPy2+Qzm7x8N/K+t5u7yWepeLCoeHtTWNOKUkSafhp1nJ3JQS1yf3ncLQR1EgA8CFF17IhRdeGG4xhhVf1DTyq0PFTMfJB6dPRq24Z7pEr1JxZVIMVybFtO1zeDxsqGvmifxyfnmwmP+U1fDsuIyA5rQoDA8U+1RhyLG3ycrte/IZbTLwQ5oV5REgOpWKM2MjeH/qKP4+LpN8q51zvjnAd3cf442yauweT7hFVBgkKApEYUhxqNnG1duPYFKrWDoxG6OiO/qMEILLk2JYNyuPq5Nj2Vzfwk/3F3HuNwf5pr6ZUpuDg802qh2uLqMNFYY+igtLYchQ5XBx9fYjqAS8OWUkGUY9R8Mt1BAgWa/lj2NGIHMlq6obuPdgMZdsPdShTIRaxYOj0rghJVYJUx9GKApEYcjwYnElxx1O/jcjl5GKvz7oCCE4Nz6KOdEW3jpei0aARa2m2unik8p67j5QxJe1jTyZl6FEbw0TFAWiMCRocrn5V0kVFyZEMXEIhucOJiI0ar7TKZrr1rR4ni2s4NGjZTg8khcnZA2JsacGlxuB9zMrnIyiQBSGBK+WVVPvcvODEYnhFmVYohKCH2cmYVKr+NWhEu4/WMxjuekD7s6SUrKloYVGlxshoNbppsbpQgJaIYjTahhp8m+F7H1NVhbvOIJOJfhg6mhSDTo8UvLfynpeLqlkQWwkP8xIHNYTMBUFEmJuvfVWPvroIxITE9m9ezcANTU1LF68mPz8fKKioli5ciUxMTG9tKTQHQ6Ph+eLKpkbbWZalDnc4gxrvpueQJndyV8LK5gcaeL6lLiQ9ymlpNTuZFtDC38pPM6ORmuvdW5Dx/x275vdbj6qqOet4zUct7uYHW3mo4o6dCpBvdPN4h1HuD7Fm6X5qNVOrFbN13XNbK5v5vsZieQY9STqtd11N2RRFEiIueWWW/jhD3/It7/97bZ9jz32GAsXLuS+++5jyZIlPPbYYzz++ONhlPLU5s3yWkrtTv4wZkTvhRVCzgM5KWxvaOFXh0qYE2Uhx88n/kA50mJjWWk1bx+vpcLhXWRrhEHHE2NGkGc24JGSKK2GWK0aFQKnlFQ7XTx6tIyXqyXTj9cyyqTns5pG/l5USbXTRaZBR45Jz9vHa0nRaXltcg6ldifX7TjCw0dKmRlp5t6cZC5OiOZfJVU8eLiE/1V7k7SONukZLQ2sPVSCzeNhdrSFM2IsxGs1QzawQFEgIebMM88kPz+/w77333+fdevWAbBo0SLuv/9+RYH0kWaXmz8cK2NGpImFg2CBJbe7hSNHn8RhryAv7xE0muFnEamE4JmxGSz85gA/2FvAB9NGow3iTHaPlPytqJLHjpYhkZwbF8VZsRHkmgzMiDKhU3U/gJ+s1/L8+EwWfb6N7+8taNs/PyaCH2Umclq0BSEETo9EI7yBA5lGPSum5wIwtl3G5dvSE7goIZp9TVYONNtYVd3Apy169GXVqIB/l1YDYFQJRpoM3J2VzKL4yCGlTIaPAllxH5Tv6rXYlLo6OBbtX5vJE+GCxwIW5fjx46SkpAAQFxdHRcUpu9hi2PlbUSXHHS7+MSE77D/M6pov2b//V9hsRYAKq62EKZNfQquNCqtc4SDNoOOPY0Zw+558niwo557slKC1fceeAj6srOOihCgeHZ0esOvIrFbzS5ppHjuRCLWKHJOBMeaOUXudFV57xdGeZL2WZL2WBXGR3JmRyJq16zj7zPm4pWRHYwub6popdzhZXd3ALbuPMTfazF2ZyZwRYwn79RoMho8CURhyHLc7ea6ogosTopgZxLEPu72ChoYd2GylJCdfilbbcXyq/PiH6PXJxETPBKCp+RCHDz9GdfU6jMZMpk19Haerlt27f8rWbTcwZcq/0Oviu+pqSHNJYjRXV8fwVP5xzo6NZEYQvqOtDc18WFnHTzOTuDc7uc83YbOQXJQc2295OtOqd9SdsiQ/kJPKK6VVPF1wnGt2HGF2lJk/54045cPNh48C8dNS2L5uHfPnzw+pKElJSZSVlZGSkkJ1dTWJiUrkUF/4R3Eldo+HB3KClx6/snIVu3Z/HyndAByv+JhpU19BpdIBUFj4EocOPwJAWtr1AJSW/ge12sSoUfeRnvZt1Gqvz3/y5BfZufNOtm69lqlTlmIwdJTT7qiisuITkpMvQ6MJv/stFPx+dDpf1zXxw30FrJ4xBnM/w2H/XlRJpEbFDzMST6kneK1KcGt6AtenxPGf8hoePVrGOd8c4FcjU/lOWvwpG8mlzPYJA5deein//ve/Afj000+57LLLwizRqYeU8FFlHWdER5AdpEFau72Cffvvw2LOY8b0Nxk39g/U12/mwIEHsVpLKCh8kUOHHyEx4QIyRtxGSckblJYuJy3tBubOWUNmxu1tygMgLnYeU6f8C7u9km82X0lx8at4PHYAbLYytmxZzIGDD7JhwyKOV6wYkulAIjVqnh2bSYHVwa8Pl/hVx+7xUO90nbS/wGrno4o6bkqNx3KKzsswqFXcnBbPull5zIm28MChEq7dcYQSmyPcovWJ4WOBhInrrruOdevWUVVVRXp6Og8//DD33Xcf11xzDS+99BIRERGsWrUq3GKechSi4pjVwQ8y+r7KX3uk9LB33z243VbGj38Ss3kkUVHTaGk5Rn7B3ygtWw5AXNxZjB//Z1QqHSkp30KlMmAyZXbbbnT0DKZPe4MDB3/NgYO/5uixp4iKnEJT80Gczjry8h6huHgZu3f/kKioGYzM+Tlm8yiEENTXb6ep6QAxMbOIjJyK01lNY+NeoqKmnlIWy5xoCz/MSOQvhRWcFx/Fovjux4Q21zdzx558jjucXJoYw48zE8kze8cfni+qRCXgu+mnvjswWa/ltUk5vFJazUNHSrlgy0H+M3lkt2MtgxVFgYSY119/vcv9q1evBrwrIMbGBt8XO9TZiA4VcH4PN6NAKC5eSk3NF4zJ/Q1m88i2/Tk5P8NgSEfiwWLOJSpqKkJ4n34tljF+tR0RMZbp05ZTU7ue8vJ3aWzcg0DF1KlLiYqcTErytygtW86xY0+zddt1Xbah1cbidNYAoFZbSE29muysH6PVRvbzkw8Mv8hOZl1NIz/cW8A1ybFcluhdk6V9xNRrZdXcc6CIVL2Om1LjebO8hv9V1fP21FGU2538s6SKG1LiSNHrwvhJgocQgm+nxTMr2sy1249yxbbDvDwhO9xiBYSiQBROSTahZU60JSjLzTY1HeDwkceJjzu7bVyjFSFUpKVd2+8+hBDExc4jLnbeScdUKg3padeTnHQZVVWrcTpr8UgHERETsJhHU139GdU1X2Ixj8ZiyaP8+Ic+hfclUya/fNLYymBEp1LxjwlZ/OZIKcvKqnmppAq9SjAvOoI/jklnn1Tz+wNFnB4dwQvjM4nSavhRRiKXbjvEdTuOYPNIJkUY+c3otHB/lKCTZzbywbRRXLvjKFduP8w5GJnqdBGlHfy358EvoYJCJw422yhGzfcS+m99eDx29uz9GWp1BGPHPhrWgVmNxkxy8qUn7U9J+RYpKd9qex8ffzY1qdewc+edbN58FdOmvYbJlDWAkvaNTKOelyZkU+908UVtE980NLOstJpzNx/EiZkso54XJ2S1raSYatDx5uRRXLrtENEaFUsn5mAaokkaM4x6Vs7I5Q/HynmhuIJ5m/bz8Kg0rkiMHtTBAkPz21AY0vy3sg6AC/uhQFyuZvILnufrDefR1LSfcWMfR3cKhdrGxsxlxvTluD029u77RVvU2KlAlFbDxYnRPDwqjU+m5xKn1eBE8M8J2W3Ko5Vsk551M/NYNXMMSUM8VYhZo+bh0Wk8QhNpeh3f31vANTuOcLjFFm7RukVRIAqnHP+trCcXV7984YcPP8qRI3/AYEhj0qQXiI9fEEQJBwaLZQy5o39Fff1WiotfCbc4fWK02cDKmbk8TQOjzV3PiYjTaYg9Bdw5wSJbuPnv9NE8mpvOjsYWzt50gPsPFlM6CCO1FAWicEpRYLWzu8nKbJx9bsPjcXK8YgVJSZcyfdprJMQvDKKEA0ty8hXExZ3F4SNP0Nx8JNzi9Am9SkWkGHohzP1BLQTfSYvny1ljuTo5hqWlVczZsI//lNWEW7QOhF2BCCHuFkJIIUR8p/0ZQogmIcTd4ZJNYfDxUWU9ADP7oUBq6zbictWRlHhBsMQKG0II8sb8DrXayJat19LQuDvcIikEkUS9lj/lZfDV7LHMjjbzk/2F/OlY+aCZMxRWBSKEGAGcCxR2cfhJYMXAShR8ioqKWLBgAWPHjmX8+PE8/fTTgDel+7nnnsuNN97IueeeS21tbZglPTX4qKKOSRFGEoWnz21UVHyMWm0mNvbMIEoWPgyGVKZP+w9qlYGtW2+gvn5buEVSCDIZRj2vTsrh6uQY/phfzg07j1Lp6PtDVLAItwXyJHAP0EGdCiEuB44Ce8IgU1DRaDT86U9/Yt++fWzYsIG//vWv7N27ty2l+7Jly1i4cCGPPRZ4UsbhRonNwbbGFi5OiO5zGx6Pi8rKlcTHLUCtPrXzELXHbM5h+ow30eli2bHzDmy20nCLpBBkdCoVz+Rl8MjoNL6qa+KsTfv5d0kVLk/4rBERLlNICHEpsFBK+RMhRD4wQ0pZJYQwA6vwWiZ3A01Syie6aWMJsAQgISFh+vLly/stV1NTExaLpd/tdMcDDzzAFVdcwTPPPMOTTz6JXq/Hbrdz1113sXTp0qD0EazP0J92+lK3tzofSx1LMfFnGohsbvCr/c5tSrkPj3wClfgeQswISL5TASlL8chHgHiEOB1kHYgMBBNpbnZhNhsQYvBNxAv1726g+gtGu/60USxVvISJfWhIx82NWJkiXH3qf8GCBVuklH36MYRUgQghVgHJXRx6ALgfOE9KWd9JgTwBbJJSLhdCPEQPCqQ9Y8aMkQcOHOj2+OObHmd/zf5eZa6rqyM6OrrXcgB5sXncO+tev8oC5Ofnc+aZZ7J7924yMjKoq6tjnS95Y0xMTNDcWOuClBCyP+30pW5PdaSUXLDlEHaPh7Wz8vxuv3O5vXt/wfGKFZx5xjeo1adW2gh/qapex86ddyClCyHUvhBfgdfQF2Rl3klOzs8H1fyCYF2z4e4vGO3624aUkk+q6vnNkVKOWR0sio9kcVUhFy4IrH8hRJ8VSEhj46SU53S1XwgxEcgGdvgu4nRgqxBiFjAbuEoI8QcgGvAIIWxSymdDKWuoaWpq4lvf+hZPPfUUkZGnRvqJwcT7FXVsb2zhT/1YdbCyajVl5e8wYsStQ1Z5AMTHzWfe6V8jhECjiaShYQfVNevJzz9KQoKL/IK/4fE4GDXql4NKiSgEhhCCCxKiWRgXyYvFVTx6tJS9RDDJ5iDdMDBWZliCq6WUu4C2HObtLRDgjHb7H8JrgfRbefhrKYTiycTpdPKtb32LG264gSuvvBI4kdIdoKysTEnp3gMtbg+/PVLKBIuRa1P6ljfMZi9n3757sVjGMTJn6Af26XQnzlNU1DSioqZRWLCOCePP4qAujsKil2ixFpA35rfo9cq1dyqjU6n4QUYiEy1Gbt5xmPM2H+DZsZmcHRf6B9VwD6IPeaSU3HbbbYwdO5af/exnbfvbp3T/97//raR074G/FVZQYnfy29FpqPvwxOzx2Nm9+yd4PHYmjH+6Q8r14YYQgtzRDzJ61P3U1HzBho2LKCp+BY8n/BE9Cv3jzNgIfkcjSTot1+88OiDhvoNieqeUMqub/Q8NrCTBZ/369bzyyitMnDiRKVOmAPDII4+0pXR/9tlnGTNmDG+++WZ4BR2k2Nwe/upbdXBudOCDk1JK9u67j/r6zYwf/xRmc04IpDy1EEKQkXEb8fFns3//rzh48CGKi5cyccKzfmcYVhicpAkPH0/P5Z6DRfwxvxwPkl8EcTnhzgwKBTKUmTdvXrdPAatXrx7wwcNTjW/qm2lxe7g2Ja5P9aV8n+PHP2Rkzt0kJ10SZOlObUymbKZOXUZ19Vr27X+A7TtuZeaMdxWX1imOUa3i6bwMNELwp/zjaITgrqyuYpn6j+LCUhjUfF7biEbA3D6sp+1w1CD5mOSky8jMvDME0p36CCGIjz+bKZP/gcvVwI6dt9PcfHjQzHRW6BsqIXhizAiuTo7h8WPl/KXgeGj6CUmrCgpB4vPaRqZHmvu0lnZF5SeAm4zMJUq0US9ERIxn/PinaGrax4aNi/hy/VyKil9Byr7P+FcIL2oheCovgyuTYvj90TKeCMGYiKJAFAYttU4XOxutnBHTt+Vbj5d/AKRiMSt+fX9IiF/I3Dmryct7BLN5FAcPPsTWrddTVbWmbS13hVMLtRA8k5fBNckxPJFfzo/2FWL3BO+hQFEgCoOWr+qakMCZMYEPnttspdTVf4MQcxTrIwCMxhGkpS5m6pRXGJv3OE3Nh9ix83a++HIOZWVvh1s8hT6gUQmezsvgvuxk3jpey+LtR6h2uILStqJAFAYtn9c0YlarmBoZ+PjH8eMfAiCYFWyxhgVCCFJTr+KMeV8xedI/sFjGsnffPew/8GvcbsUaOdUQQvDTrGT+Pi6TbY0tXLT1IAXW/n+PigJRGLR8UdvEadEWtKrALAgpJeXHPyAqcipCJIRIuuGBSqUnPn4BU6csJSPjdkpKXmXTNxdTW/dNuEVT6AOXJ8Xw9pRR1DndXLP9CMft/Zv/oyiQEGOz2Zg1axaTJ09m/PjxPPjgg4CSzr03imwOjlrtnBGg+8rjsbN33900Ne0nJeWqEEk3/FCpNIwedR9TJv8Lj8fB1q3XUln5v3CLpdAHZkSZeXVSDpVOF9fu6N8iZIoCCTF6vZ41a9awY8cOtm/fzieffMKGDRuUdO69sLLKu3DUwgDSMbjddrZt/w7l5e+Rk/MzUlMXh0q8YUtc3BnMnvUxEZbx7Nv/AA5HVbhFUugD06PM/GtCNsf66cZSFEiIEUK0pVd2Op04nU6EELz//vvcfPPNANx888289957YZRy8PG/qgZGGvWMNPm/Zkd19Vrq6jaSN+b3ZGf9QBk8DxEajZlx457A7W5i3/4HlDkjpyhnxkbw5pRR/Wpj2MxEL3/kEez7ek/nHlNXR8FLL/vVpn5sHsn3399rObfbzfTp0zl8+DA/+MEPmD17NsePHyclJYUDBw6QkpJCRUWFX30OBxpdbtbXNfHd9PjeC7ejqmoNGk1Uv1xXNpuNTZs24XK5MBqNTJkyBaPRm7m3paWFLVu2sGXLFiZOnMjChafuWur9xWLJJSfn5xw+/Cj7DzzAmNyHUKkG3zojCj0zsw8TdNszbBRIOFGr1Wzfvp26ujquuOIKdu9W1q3uiXU1jTilZFF8lN91pPRQVb2WuLizUKn6dll7PB7efvttDh061LZvx44d3HTTTZSWlvL2229js9mIiIhg/fr1TJ48mfj4wJTcUCJjxK04nXUUFPyNluajTJjwjJIGZZgxbBSIP5YCwLF165gcotxU0dHRzJ8/n08++URJ594D/6uuJ0ajZkYA4bsNDTtxOmuIj1vQ537XrFnDoUOHuOiii5gxYwaHDx/mP//5D88//zwNDQ0kJSVxyy23YLFYeOaZZ1i1ahXXXnstLpeL2tpa6urqiIqKGjbfpRAqRo28G4tlDPv23cfGTRcxbuzjxMefHW7RFAYIZQwkxFRWVlJXVweA1Wpl1apV5OXlKencu8HlkayubmBhXCSaAMJ3q6rXACri4s4MuE+73c6nn37Kl19+yfTp05kxYwZCCEaPHs3111+PzWZj4sSJ3HbbbSQnJ2OxWJg3bx779+/n3Xff5U9/+hN//etfefXVV3n55Zdpbm4OWIZTmeSkS5g18wP0+mR27LydkpI3wi2SwgAxbCyQcFFWVsbNN9+M2+3G4/FwzTXXcPHFFzN37lwlnXsXbGlopsbp5rwA3FcAVVVriY6ajlYbHVC9srIyXnvtNRobG5k+fToXXHBBh8H3nJwc7rnnHjSajj+VOXPmsGXLFnbt2kVeXh5jxoxBq9Xy1ltvsWrVqmH3QGA2j2TmjLfYuet77D/wK9Qas5L9eBigKJAQM2nSJLZt23bS/ri4OCWdexfsarICMDfaf/eVzVZGU9NeRo28J6C+pJSsWLGibdGvESO6Xi63s/IA0Ol0LFmyBACz+YSsc+fOZf369UydOpWMjIyA5DnVUan0TJzwV7bvuJW9e39Obc1XZGf/CIMhNdyiKYQIxYWlMKg40mInUqMiXuv/s01N7XoA4uLmB9TXsWPHKCws5Mwzz+xWefSE2WzuoDwAzjzzTCIjI/n444+HZXirWm1k8qQXSUu7gbLy9/h6wzmUl38QbrEUQoTfCkQIESOEGC+EyBFCKIpHISQcbbGTYzQENIejoWEnarUFs3m033WklKxbt46IiAimTp3aF1G7RK/Xc/bZZ1NeXt4hmms4odFYGJP7IHPnrCIyYhJ79t7FkSNPKKnhhyA9KgIhRJQQ4n4hxC5gA/A8sBwoEEK8KYToe8iLgkIXHLXaGWkKbM3yhoadREZMIJDnmvz8fAoLC5k3bx5arTZQMXtk4sSJREVF8cUXXwxLK6QVozGNqVOXkpq6mPyCv7Fz1/dwuZrCLZZCEOntF/cWUAScIaUcI6WcJ6WcIaUcATwGXCaEuC3kUioMCxwSim0Oso3+KxCPx05T034iIycF1NeGDRswm81MmzYtUDF7Ra1Wc9ppp1FUVERBQUHQ2z+VUKl05I35Pbmjf0119Vq2bLkGh6M63GIpBIkeFYiU8lwp5StSyroujm2RUv5USvlSyKRTGFYcR4WEgCyQpqYDSOkkInJiAHWaOHjwIFOmTAm69dHKtGnTMJvNrFu3DpcrOGsvnKoIIRgx4mYmT36ZFmsB27d/B5erMdxiKQSB3lxYe4UQDwghRg6UQArDlzK8y9bmBKBAGhp2ARAZ4b8FsmvXLqSUTJ48OTABA0Cr1TJ//nzy8/NZunQpjY3KDTMudh6TJj5HU/NBtm2/habm/o8RSSmxWgtpbNpPU9MB7PbjSOkOgrQK/tBbqMt1wLXA/4QQVcDrwHIpZWnIJRtiuN1uZsyYQVpaGh999BE1NTUsXryYffv2MXbsWJYvX05MTEy4xQwrZb7nmZwAXFgNjTvRamMxGNL8Ki+lZNu2baSlpYV8xvjMmTMxGAy8//77PP/881x44YWMHTt2WCd5jIs7iwnjn2bf/nvZtOkiYmPPREoXHo8Hh2MSOl1sr204nfVUVq6ksvJT6uo343I1dDguhIbkpEvJyblLCSEOMb25sHZIKX8ppRwJ/ATIBDYIIdYIIW4fEAmHCE8//TRjx45te6+kcz+ZclQk6jREaNR+12lo2Elk5ES/b8rl5eVUVFQwZcqUPkoZGBMnTuT222/HYrGwfPlyli9fjts9vJ+QExMXMXfOGtLSbsBqLcTlakCykU2bLvatv979Ikc1Nev5esNC9u2/l6bmgyQmXkDemN8xccJzTJjwLGNyf0Nq6mKOV3zE1xsWcujwoziddQP34YYZfgfbSyk34FUe7wNPAs8CL4ZKsKFEcXEx//3vf3nggQf485//DMD777/PunXrOHDgADfffDPz58/n8ccfD7Ok4aUMdUDWh9vdQnPzYRITFvldZ+vWrajVaiZMmNAXEftEUlISt99+O+vXr2fNmjV88cUXw37yqE4Xy5jcB9ver127FLVmKTt23o5GE0GEZTwqtRGtNhqTMQuVSofVWkhJ6X8wm0cyedI/iIyc3O2DQ1bmnRw9+hSFhS9RWrqcrMw7SU+/GbXa/+UBFHrHLwUihJiJ1531LSAfeAE4pXJvfLH8IFVFvYcQ1tV5qN2y1a8240dYOOOa3F7L/fSnP+UPf/hDBz+4ks79ZMpQMS2A8Y/Gxr2Ax+8IrNraWrZu3crEiRPbUrQPFGq1mjPPPJPKyko+//xzcnNzSU1V3CutCJHBrJkfUV39GVXVa2lpOYLL0UhT0z7Ky98FQKUykpR0MXljfotG0/NKlQZDKuPG/YGMjNs4fOSPHD7yB4qKlzJu7B+IjT19ID7SsKBHBSKEeARYDNQCbwCnSymLB0KwocJHH31EYmIi06dPZ926deEWZ9DS4HJTjyogC6S+YTsAERH+RWCtWbMGIQQLFoRv+tIFF1zAsWPHePfdd7njjju6TJMyXFGrDSQmLiIxsaNF6Xa3IKUbtdoS8PiRxTKGKZP/QW3tRvYf+DXbtt9MVtYPkDJ0ARTDid6uXjtwgZTy4EAIE0r8sRQAX26q4M0NWL9+PR988AEff/wxNpuNhoYGbrzxRiWdeyeOtniX1gwkhLey4hMs5jHo9Qm9li0tLWXXrl3MmzePqKjAEjUGE5PJxKWXXsprr73GunXrOOecc8Imy6mCWm3qdxsxMbOZNfNdDhx4iPz8ZxGcgZTzA5p8qnAyvQ2iPyylPCiEUAshLhVC/FgI8bPWLRgCCCHuFkJIIUR8u32ThBBfCyH2CCF2CSFOWcflo48+SnFxMfn5+bzxxhucffbZLFu2TEnn3onWtZmz/VQgUlZS37CNpKRL/Sq/du1aTCYT8+bN67OMwSI3N5epU6eyfv16ioqKwi3OsEGtNjF27ONkZ/0YyRfs3XePEvLbT/xVvx8CtwBxQES7rV8IIUYA5wKF7fZpgGXAnVLK8cB8oPuwjFOU++67j5UrV3LjjTeycuVK7rvvvnCLFFaOtNgRSLIMfioQNgGQlHRxr2VtNhtHjhxh6tSpGAyD41lk0aJFREZG8t577+FwOMItzrBBCEFOzk8Q4nLKy99lz9678XiG90TP/uCvAzZdShlYrgj/eBK4B3i/3b7zgJ1Syh0AUsohk/dg/vz5bdE3Sjr3juRb7cQhMaj9e6aRciNRUdMxGtN7LXvs2DE8Hg+jR/ufbDHUGAwGLrvsMpYuXcrq1au54IILwi3SsEIlLiErezRHjv4RKV2MH/cEKlVgOdgUQPiT7E0I8TiwWkr5v6B1LMSlwEIp5U+EEPnADClllRDip8B0IBFIAN6QUv6hmzaWAEsAEhISpi9fvrzfcjU1NWGx9BzhEUxC0V+w2uxPO4HWfVBawO3iYY2t17JSFuORDyLEDahEz8unNjU1UVJSQkVFBaeffjoq1eDyeR86dIiSkhImT54c8omkA31tB0K4fnce+SlSLgdGoRI/QIjIoLQbrjb6UnfBggVbpJQz+tShlLLXDbgCaAasQAPQCDT4UW8VsLuL7TJgIxDlK5cPxPte3w0cA+IBE/A1XkXTY1+5ubkyGKxduzYo7YSzv2C12Z92Aq076ctdcvGaDX6VPXz4j3LV6lHSbq/qteyaNWvkn/70J/nGG28EJM9AYbfb5dNPPy2ffPJJabVaQ9rXQF/bgRDO3115+Udyzdqx8sv1Z8iGht1BazccbfSlLrBZ+qEHutr8fRz7EzAXMEkpI6WUEVLKXlW1lPIcKeWEzhtwFMgGdvisj3RgqxAiGSgGPpNSVkkpW4CPgeCnTFUYNLS4PRx3uEjCv/Ui6uo2A9nodHG9lm1ubqahoYFRo0b1U8rQoNPpuPzyy6mvr+e9994b1unfw0VS0kVMn/YGUrrZvOVqysreDbdIpwz+KpBDwG4ZpKtbSrlLSpkopcySUmbhVRrTpJTlwKfAJCGEyTegfhawNxj9KgxOimzeQeREPxSIlJKm5v0I/FtBsKamBmDQKhCAjIwMzjvvPPbv38+XX34ZbnGGJZGRk5g1830iI6ewd9/dHDj4cI8pVRS8+DuIXgasE0KswDs3BAAp5Z+DLZCUslYI8WfgG0ACH0sp/xvsfhQGDwW+EF5/FIjNVoLL1Yg3gK93ampqSExMDOvcD3+YM2cOJSUlrFmzhqysrD4tsavQP3S6eKZOWcrhI49TVPQyTY37mDDhL37NMxqu+GuBHANWAzqCGMbbis8SqWr3fpmUcrzP5XVPsPpRGJwU+iwQf1xYTU37APyyQKxWK/X19YMq+qo7hBBccsklWCwWPv74YzweZfnXcKBSacgd/QDjxz1JQ+MuvvnmMurrt4dbrEGLXxaIlPLhUAsylMnKyiIiIgK1Wo1Go2Hz5s1KOvd2FFjtmNUqIty9e0gbm/YDAug9ffuhQ4eQUpKXl9d/IQcAvV7PokWLeOutt9iyZQszZ84Mt0jDluTkSzFbctm183ts3XYjkye9QGzsaeEWa9DR24JSLwghukw0JIQwCyFuFULcEBrRhhZr165l+/btbN68GVDSubenwOog06DDnzRHTU37MBoz8Cc5wf79+9HpdKSl+bdWyGBg/PjxZGdns3r1apqbm8MtzrAmwpLH9BlvYjSOYMfO71Jd/UW4RRp09ObCeg74PyHEPiHEm0KI54QQLwshvgC+wuvGeivkUg5B3n//fW6++WYAbr75Zt57773wChRGCmwOMv1MotjUtA+LZWyv5VwuF4cPHyYuLm7Qzf3oCSEEF1xwAXa7nfXr14dbnGGPXhfPtKmvYjLlsGv392ls3BdukQYVPbqwpJTbgWuEEBZgBpCCdy7IPinlgdCLFzzW/usFKgqO9lqurq6O45994lebiZk5LLhlSa/lhBCcd955CCG44447WLJkiZLO3YeUkkKrnfmxvQ+puVxNWK2FpCRfSU0v+QmOHTuGw+EgPj6+54Lt8Djc2PbX4ChsJPLsEahMoVkvvTcSExOZOHEimzZtYu7cuUREBG24UaEP6HSxTJn8Et9svpIdO29n5ox3lYF1H/6OgTQB60IrytBl/fr1pKamUlFRwbnnnnvK+OQHggqHC6tHkmnQ9Vq2udmbFNofC6TVfeXPuJLH4abxs2KavihBOk4k14u+OKfXuqHirLPOYteuXXz55ZdKmpNBgF6fxKRJz7Nly7Xs3v0jpk5dhkqlpOIfNmfAH0sBCEluqtaFgxITE7niiivYtGmTks7dR2sIrz8uLO8AeqsCOdRtObfbzYEDBxg9enSv7iv7sXpq/nMAd50d48R4zLNTaNl6nKavS7GcnoomJjzJF+Pi4pgyZQqbN29mzpw5wzbAYjARGTGBvLzfsXfvzzmW/wwjc4KSkPyU5tRxDp+iNDc3t61E2NzczP/+9z8mTJigpHP3UeAL4c009m6BNDXtQ6OJxGDoeSW/TZs20dTUxKRJPef/tO6tpvKl3QiNioQlE4m7YSyGUdFEnpcFAhpWFvj9OULBWWedhVqt5q233sLlUjLGDgZSki8nJeUq8vOfo6p6XbjFCTuKAgkxx48fZ968eUyePJlZs2Zx0UUXcf755yvp3H0UWB0IIF3vjwLZj8U8psdV6RoaGli7di2jRo0iN7f7RcSse6upXrYXbbKJhO9NRp8T3XZME63HMjeVlm0VOMvDFwkVHR3N5ZdfTklJCZ9++mnY5FDoyJjcB7FYxrJr1/epqRnegQ7+rok+A3gAyPTVEYCUoUnxPqTIyclhx44dJ+1X0rl7KbDZSdFre03jLqWkufkISUkX9Vjuk08+wePxcOGFF/aoaBo/K0YTayTh9omo9Cf/DCLmj6Bly3Fq3z1Mwh2TEKrAllINFuPGjWPu3Ll8/fXX5ObmnhKTIoc6arWJqVP+xdZtN7Jj5xKmTvk30dF9S2Z7quOvBfIq8E/gW8AlwMW+vwoK/aLI6iDDjwF0p7MGl6sek6n7ge3jx4+zd+9e5s2bR2xsbLflXHV2HAUNmKYldqk8ANRmLVEX5eAoaKD5m/LeP0gIOeecc4iOjmbdunVKssVBgk4Xx7Spr2AwpLJz1/ewWofnypL+KpBKKeUHUspjUsqC1i2kkikMC4psDtL9UCAtLccAMJmyuy1TXFwMwMSJXc59bcO6s9Lb1qSeQzFN0xLRj4yifsUxXPX2HsuGErVazbx58ygpKeHo0d5D0RUGBp0unsmTXkBKNzt2LsHlagq3SAOOvwrkQSHEP4QQ1wkhrmzdQiqZwpDH6ZGU2Z2M8EuBeG+cJmP3CqSsrAy9Xt9rxFLLzkq0aRY08cYeywkhiL5iNLglFX/ZhnVv+BbHnDJlCpGRkXz22WeKFTKIMJmymTjhL7S0HGHP3p8h5fDKYeavAvkOMAU4H6/rqtWNpaDQZ0rtDjzglwJpbjmKELoel7AtKysjOTm5x9BdV7UVZ3ETpkn+TTDUxhtJ/MEU1BE6qpfupWlDmV/1go1Go+H000+nsLCQvXuV1Q0GE7GxpzN69P9RVbUaKd8JtzgDir8KZLKUcoaU8mYp5Xd8260hlUxhyFPsC+H1zwI5hsmUiRDqLo+73e622f09trPLm/TZONH/mcTaZDOJP5iCLjuShlUFSGd4njKnTZtGWloa7777LoWFhWGRQaFr0tNuJC3teiQrhtWCVP4qkA1CiHEhlURh2NG6kJR/YyBHexz/qKqqwuVy9apArHuq0aZb0MQGNkFQaFREnpOJp8lJ8+bwDKprtVquv/56IiMjee2116iqquq9ksKAIIQgd/SvgTz27b+f+vqt4RZpQPBXgcwDtgshDgghdgohdgkhdoZSsKFEXV0dV111FXl5eYwdO5avv/6ampoazj33XG688UbOPfdcamtrwy3mgFNscyKAVEPPOac8HhdWa2GPEVjl5d6bek8KxN3kwFnciHFs70vhdoU+JwpdRgSNnxUj3eGxQsxmMzfddBPgTciprBsyeFCptKjE9zAYktmx805aWvLDLVLI8VeBnA+MBs6jizBeIYSSZ6EHfvKTn3D++eezf/9+duzYwdixY5V07nhdWEk6Lfpe0o3YbEVI6erRAikrK0Oj0fSYPNF2oBYkGPK6D/HtCSEEEWdn4K6z07K9sk9tBIOYmBguuOACioqK2LRpU9jkUDgZISxMnvQiINm67Ya26MGhil8KpH3objdhvKtDJN8pT0NDA59//jm33XYbADqdjujoaCWdO60hvL1nvPUnhNefAXTb/hpUETq0qebAhfVhGBODJt5Iy/bwZk+eNGkSo0ePZtWqVYora5BhNo9i2tRleDwOtm69Abs9fA8boSZYyRTDM003AOo+PIKjtPe0FGl1KioO+Oed06Waib5kZI9ljh49SkJCAt/5znfYsWMH06dP5+mnn1bSueO1QKZFmnot1+wL4TV348LyeDyUlZUxefLkbtuQbg+2g7UYJ8b3OEO9N4QQGMbE0LSxHOn0ILThyQYkhODiiy/m+eefZ9myZdx2221K2vdBhMUyhqlTX2Hz5ivZv/9+Jk16oV/X3WAlWFe/EpjeDS6Xi61bt/K9732Pbdu2YTabh6W7qjNuKSmx+zuJ8ChabQxabdee0traWhwOR4/jH46CBqTdjbGP7qv26HNjwOXBnl/f77b6Q1RUFDfccAMtLS288sorWK3WsMqj0JEISx4jR/6Cquo1lJUNzXX3hk06994shVb2rlvH6PnBS/GVnp5Oeno6s2fPBuCqq67iscceG/bp3I/bnbhkICG83buvWgfQk5OTuy1j3V8LaoF+dHTAsnZGnx0FaoHtUB2G0eEd/ktLS2Px4sW8+uqrvPvuu1x77bWn1AqMQ50R6TdTWbmSg4d+g1pjJinxwnCLFFSCdaUNPdssSCQnJzNixAgOHPAu4Lh69WrGjRs37NO5Bx7C230EVnW1d4Z4zwPoNeizIrvNfRUIKp0afWYk9kODI3Ju5MiRLFq0iIMHDyrL4A4yhFAxYfxTmM1j2L37Rxw89PshlUmgx1+TEOK/wGvAe1LKngYQFgZVqiHGX/7yF2644QYcDgc5OTn885//xOPxcM011/Dss88yZswY3nzzzXCLOaD4O4nQ7bbjcFRhNIzotkxNTQ0WiwWdruu23M1OXMdbME3J7LvAndDnxtDwST7uRgfqiN6VYKiZNWsWRUVFrFmzhujo6F7zgSkMHHp9ItOnvcahQ49QVPQyOl08WZl3hFusoNDb49gLwLXAk0KItcDrwMdSSkf7QlLKmhDJNyRoXVmuM8M5nXuxzQlAWi8KxOHwRrDo9d27+GpqanrMvus45h2raL/mR38xjIqmAbAdrsM8NfzuRyEEl1xyCfX19bz99tsUFxdz3nnnoVZ3PXNfYWBRqXTk5j6Iw1nDkSN/xGLOJT5+QbjF6jc9urCklO9LKa/Duw7IO8DNQKEQ4mUhxLkDIaDC0KTI5iBOq8HUyzogrQpEp+s+9UhvCsR+tB6hVaFLs/RN2C7QplpQmTXYDw4ONxaAXq/nlltuYfbs2WzcuJG1a9eGWySFdgghGDf2cSIs49i95yc0NOwKt0j9xt95IFYp5X+klFfgnUw4FfgkpJIpDGmKbQ6/BtDtDm94s07ftQJxOBw0NTX1okDq0GVFIjTBG1wWKoEhNxbbgRqke/D4tNVqNRdccAFTpkzhq6++agswUBgcqNVGJk1+Aa02mu07bj3lJxr69YsSQiQJIX4khFgPvAf8D5geSsEUhjb+TiJ02L2T5PTdWCCtKWC6UyAqBzjLW7yRU0HGMC4WT4sLR0F4w3m74rzzzsNgMPDhhx8q6U4GGQZ9MlOneANotmy9lprar8MsUd/pUYEIIW4XQqwBtgK5wD1Syhwp5b1Syu3BEEAIcbcQQgoh4n3vtUKIf/vybe0TQvwyGP0oDB6kbw6I/xaIQKvtOn9VTY13+K07BWL0eZj0OSFQILkxoBZY9w6+IUCTycT5559PSUkJK1euHFKRP0MBkymbadNeQ6OJYtu2b1NU9K9wi9QnehtEPw14DFglQ7BSihBiBHAu0D439dWAXko5UQhhAvYKIV6XUuYHu3+F8FDpcGHzSL9CeB2OSrTaWFSqri/VVgXS3SJSxhrhHf9ID/4sbZVeg2FUNNa91URdlD3oZhpPnDiR4uJivv76a6SU6PX6cIuk0A6LeTQzZ7zL3r0/5+Ch32IwpJGQcGoNLfc2iP4dKeX/QqE8fDwJ3EPHmewSMAshNIARcAANIepfIQwEsg6Iw17ZawSWyWTCaOx6dUFjjUCXGdzxj/YYxsXhrrHhOt4Skvb7gxCCCy64gNmzZ7NhwwYOHz6sWCKDDI3GzPjxTxEZMYk9e39GY+O+cIsUECJcF5QQ4lJgoZTyJ0KIfGCGlLJKCKEFXsE7t8QE3CWlfKGbNpYASwASEhKmL1++vN9yNTU1YbEEL1qnsLCQ3/zmN23vy8rK+M53vsN5553Hb37zG0pLS0lNTeXBBx8MWi6jYH2G/rTTU92vpZanMfMHGsgQnh7ruD2/BSyoVXd12f727dtxu91Mn37ykJzaDtlr1VSP9lA7MjTXudoG2etC20d/kVJy5MgRiouLSU1NZfTo0YPOWgr27y5c/fW1XSlr8cjfAXbs9hswGecOWP8LFizYIqWc0acOpZTdbkB2T8d724BVwO4utsuAjUCUr1w+EO97fTrwKqAFEoEDQE5vfeXm5spgsHbt2qC00xUul0smJSXJ/Px8+Ytf/EI++uijcu3atfLRRx+V99xzT9D6CdZn6E87PdX9S365TFqzTTY4Xb3W+eLL0+SePb/otv0///nP8q233uqyn+Ztx2XRvZ9Le2GD33L3hfK/bJXlz24LaR/9xePxyH/84x/ywQcflP/6179kVVVVuEXqQCh/dwPZX3/atVpL5MZNl8lVq3NkQeHLA9Y/sFn28R7fm13/FoAQok/p2qWU50gpJ3TegKNANrDDZ32kA1uFEMnA9cAnUkqnlLICWA/0TTsOMlavXs3IkSPJzMwc1unci+1OojVqIjQ9T3KT0oPDUdVtCK/L5aK+vr7bAXTboTrcWok2iPM/usI4Lg5nUSPuBntI++kPQghycnK4+OKLKS0t5bnnnmPXrlN/HsJQwmBIZfq0/wDTOHTodxyvWBFukXqlt0F0lRDiQSBXCPGzzgellH/uS6dSyl14rQsAOrmwCoGzhRDL8Lqw5gBP9aWf9qxYscKvmPi6ujqOHfMvNjs5OZkLLrjAbxneeOMNrrvuOoBhnc69yOpfFl6nsxYpXX0K4ZVSYj9cizXWO2cjlBjHxdHwvwKs+2qwzO55Sd1wIoRgxowZbalz3nnnHdRqNePGKatVDxbUaj0qcTsRES+wd+/PMRhSiYrsfpmCcNObBXItYMOraCK62ELBXwELXlfXN8A/pZSn/PK5DoeDDz74gKuvvjrcooSdYj9DeB0O7xyQ7mah9xTC66qy4q530BIX+nEJTZIJdawB297qkPcVDCIiIrjhhhtIS0vjrbfe4uDBg+EWSaEdQuiYNOl5dNo49uy5C5er93WMwkWPFoiU8gDwuBBip5QyZPaUlDKr3esmvKG8QcVfSyFUualWrFjBtGnTSEpKAhi26dyllBTZHJwZ07tbyW5vnYXe9bnpyQKxH6oDoCU+9ApECIFxXBxNX5fisbuCkvE31Oj1em688UaWLl3Kf/7zH66//npGjvRvyQOF0KPTxTFu3BNs3XYDhw8/Sl7e78ItUpf4G9v4lRDiz0KIzb7tT0KI4M/MGsK8/vrrbe4rYNimc691uWlxe/y0QHyJFHVdp2mvra1Fp9NhMp28qqHtUC3qWAOubhY8bKiqYN+X62hpCM4scuO4WHBLbAfrgtLeQGAwGLjxxhuJj4/n9ddfZ8+ePeEWSaEdMTGzycj4LiWlr1Ne/kG4xekSfxXIy0AjcI1vawD+GSqhhhotLS2sXLmSK6+8sm3ffffdx8qVK7nxxhtZuXIl9913XxglHDiKA1gHxN6WSLFrC6Suro6oqKiTQlKl24P9aD2GUdHdtv3Z0pf4+C9P8PclN/H2ow9ibWr08xN0jS4zCpVJc8q4sVoxmUx8+9vfJjExkTfffJO33nqLlpbBN6dluDIy5y6io2ayd98vqK7+LNzinIS/CmSklPJBKeVR3/Yw0P0KPwodMJlMVFdXExV1wmiLi4tj9erVLFu2jNWrV/eYDHAoEchCUg5HJWq1GY3G3OXx+vp6oqOjT9pvz/cuX2vI7Xp2utNu4+j2zYyedRqzr7yGot07eOeRX2Pvx41TqAWGMYMvuaI/mM1mbrvtNhYsWMDevXt57rnnOHToULjFUgBUKj2TJ7+IxTyGnbu+T33DjnCL1AF/FYhVCDGv9Y0Q4nRAWYBZIWACmYVut1eg68Z9BV4F0l4pt2LbWw0a4V27vAvyd27DZbcz+dwLOf2aG7nkZ7+kIv8o7z7+MG6Xy89PcjKGcXF4WlxhXyu9L6jVas466yy++93vYjQaefXVV/nggw+w2wdvaPJwQaOJYMqUl9HpEti5805s9sGTYdlfBXIn8FchRL4v5PZZYGgsqaUwoBTZHJjVKqJ7mQMC3igsfTfuK5fLhdVqPckCkVJi3VeDYVQMKl3XfRza+BUGSwTp4yYAMHL6bM65/QeU7N/DsW0nL/zlL4bcGNCIU86N1Z7U1FSWLFnCaaedxtatW/nb3/5GSUlJuMUa9uh08Uye9DxudzM7d96J220Lt0iA/+uB7JBSTgYmAZOklFOHQmitwsDTug6IP6k0HI6KbicRtj4Zd7ZAXBUtuGtsGMZ27RJ0u5wc3bKJkdNno9aciJYad8bZGCMi2be+735mlV6NYWQ01n01p3TOKa1Wy3nnncett96KlJJ//vOfygD7IMBiGcP4cX+isXE3+/bfNyiusYAyzEkpG6SUSmJDhT7jXQfEvzXE7fbKbueA2GzeJ7DOFkhranVjNwqkcPdO7C3NjJ59Wof9ao2G3DnzOLplEw5b372zgzm5YqBkZGRw++23k5yczJtvvslnn302KG5aw5mEhHMZmfNzjh//kPyC58ItTmAKREGhvxTbnH6Nf7jdVtzupm5nobcqkM4WiG1fNdp0C+rIrlOXH9q4Hq3BSObEKScdy5t3Fi6HnSPfbOhVvu5oVVzWU9iN1R6LxcLNN9/MpEmTWLt2LW+//TZOpzPcYg1rMjPvJCnpUo4e/TMVlZ+GVRZFgSgMGA0uN/Uut98RWND9UrY2mw21Wt0h86i70YGjqBHj2K4Xn3I5nRzcuJ5RM2aj0Z0sQ1ruWCLiEvrlxlJH6tGOiBgyCgS8Lq0rrriChQsXsnv3bt58801c/Qg2UOgfQgjG5j1KZORk9uz5OY2Ne8Mmi98KRAhxmhDieiHEt1u3UAo2lHjyyScZP348EyZM4LrrrsNms1FTU8O5557LjTfeyLnnnts2q3ooc2IOSO9L2bbOQu/OArHb7URGRqJSnbiEneXNIEGXFdllnfztW7A3NzN23vwujwuVirzTz6Rg57Z+TTA0jovFWdw0qJMrBooQgjPOOIOLLrqIgwcP8vbbb+N2u8Mt1rBFrTYwaeLf0Wqj2LHzdpqbD4dFDn/XRH8FeAKYB8z0bUMiQ26oKSkp4ZlnnmHz5s3s3r0bt9vNG2+8wWOPPcbChQtZtmwZCxcu5LHHHgu3qCGn0OpVIBmG3lfGO5EHq+soLJvNdtL4h7vWe8PWxBq6rLPvy3UYI6PI6MJ91Uru7NPxuN0U7treq4zdYRzntYAG41K3/WXmzJmcf/757Nu3jxUrVihjImFEr09k8qR/IKWLzVuuoa6u7xGEfcVfC2QGcLqU8vtSyh/5th+HUrChRGvIqcvloqWlhdTU1GGZzr0okDkgjtY8WN27sE6KwKq1gYouxz/sLS0c3bKJMXPndYi+6kxi9ki0BiMlB/ruFtAkmlDHGYaUG6s9c+bM4fTTT2fz5s1s3Lgx3OIMayIixjJj+lvodLFs33EbUg7sNedv1rfdQDJQFkJZQsrBg7+lsan35SLdnjq2bO1yAcSTiLCMJTf3/3osk5aWxt13301GRgZGo5HzzjuP8847b1imcy+yOTCpVcRp/ZgDYq9ECDU67cmTAV0uFw6H4yQF4q6zo47UI9Qnhwgf2vQVLqejW/dVKyq1mtTcPEr2912BCCEwjj21kisGysKFC6murubTTz/FYrEwYcKEcIs0bDEaRzBl8sts3HQRHvkvpLxywFac9NcCiQf2CiE+FUJ80LqFUrChQm1tLe+//z7Hjh2jtLSU5uZmli1bFm6xwkKhze73HBC7oxKtNg4hTlY2DQ3eSPLOLixXrQ11TNfusQNffU5UUjIpo/N67TttzDgqC/Oxt/Q9jfaJ5IpDc2xLpVJx5ZVXMmLECN5++21lnkiYMRozGDXyPmAvxSUDd3/x99HooVAKMRD0Zim0sm7dOqZPmx+0fletWkV2djYJCV5XzJVXXslXX301LNO5F9kcZPg5B8ThqETfjfuqrq4OODmE111rRz/y5NQmHo+bkv17GXfWQr+UV1reOJCS0oP7yZ5y8lrr/nAiuWINpoldf45THZ1Oxw033MCrr77Km2++yYoVK0hMTOTCCy8kPr77FDQKoSEt7XoOHPwPBw8+hN1eTk72XahUobV+/Z2J/llXW0glGyJkZGSwYcMGWlpakFKyevVqxo4dO+zSuUspKbT6t5AUeF1Y3U0irK/3Rki1t0Ck24O7wY46+mQLpKakGKfdRsqoXL/6Thk1BqFS9c+NpRYY8mKx7j/1kisGgl6v54YbbuCcc85h1KhRlJeX89JLL1FQUBBu0YYdQghU4oekplxDQcHf2bXre3g8oQ239jcKa44Q4hshRJMQwiGEcAshlBnpfjB79myuuuoqpk2bxsSJE/F4PCxZsmTYpXOvd7lp9HMdEPC6sLrLg9VqgURGngjXddc7QIIm5uQIrPLD3hX3kv1UIFqDgaTskZQc6J9bxpAXi7S6cBT3L1X8YEev1zNv3jwuv/xyvvvd72IymVi6dClVVVXhFm3YIYSOsWMfJTf3Iaqq13Dg4IMhjZTzdwzkWeA64BBgBL7r26fgBw8//DD79+9n9+7dvPLKK+j1+mGXzr3QF4GVYexdgUjpxuGo6jYTb319PTqdDk27aCpXrXdmeldjIOVHDqIzmohNSfNb3rS8cZQfOojb1fdZ14ZR0SDAdmDohfN2R2xsLLfccgtCCL788stwizNsGZF+E1mZ36O09A0KCv4esn78nkgopTwMqKWUbinlP4H5IZNKYcgRSAivw1kLeLpdyra5uRldp5nkbXNAok+2QMoOHyR55GiEyv/EC2ljxuNyOjh+tO8TtFQmLboREUN2IL07IiIimDFjBjt27BgWE2QHKzk5Pyc56TKOHH2C8vL3Q9KHv7+oFiGEDtguhPiDEOIuoOtVfhQUuqCobRKhHwqkl1noLS0taLUdZ7O762wgOGkMxONyUVWY77f7qpWU0WOAE+6vvmIYE4uzpAl3k6Nf7ZxqnHbaaahUKsUKCSNCCMaOfYyY6Dns3XcvNbVfB70PfxXITb6yPwSagRHAt4IujcKQpdDmIEKtIsqvdUB6zoPV3Nx8kgJx1dpRRegQmo6XdEtVBR63O2AFYo6JxRQVTUX+sYDqdcaQGwMS7Ifq+tXOqUZkZCRTp05l27ZtbdGGCgOPSqVj4sS/YTJlsWvX92hq6t8D0Unt+1NISlkACCBFSvmwlPJnPpeWgoJfFNkcZBj9nANi9yqQniyQk11Yti4H0FsqvKu3pYwMTIEIIUjMyqGi4GhA9TqjTbOgMmuGnRsL4KyzzsJisfDqq6+2BT4oDDxabSRTJr+MSmVk+45bsdmCp9D9jcK6BNgOfOJ7P0WZSKgQCIW2AEJ4Wy2QLqKwWmehn2SB1HUdwttcUYYlNg5LbNcZensiISuH6qLCfg2kC5XAMDoG28FapGfohvN2RUREBDfeeCNOp5NXX321LQW/wsBjMKQyZfJLuFyNbNl6LS0t/bOsW/HXhfUQMAuoA5BSbgeygiKBwpBHSklRAArE7qhAo4lAre7ComjxLtTUXoFIj8RdZ+/SAmmuKCd55Og+yZ2YmY3H7aK6uKhP9VsxjIvD0+zEfuzUWyu9vyQmJrJ48WKqqqr48MMPleSLYSQiYhzTpr6C293C5i2LaWo60O82/VUgLinl8Lv6g8TTTz/NhAkTGD9+PE899RTAsErnXu100+L2+JWFF/CF8HbvvoKOCsTd4ACPPCmE12FtwV5fS1L2qD7JnZg9EoCK/P65sQx5sQidCuuOyn61c6qSk5PD2WefzZ49e/jmm2/63I7NZuP48eMUFRUp1kwfiYycxIzpy1EJDdu234LV2r+HI7+TKQohrgfUQojRwI+Br/rV8zBh9+7dvPjii2zatAmdTsf555/PRRddxIsvvsjChQt54IEH2LBhA4899hiPP/54uMUNCYGE8IJ3LZCAFEid92bS2QKpKPCa6a2KIFCik1PQ6PVU5B8BzulTGwAqnRrD2Disu6uIvmwkQj381nE7/fTTKSws5NNPPyUtLY20NP/n5FRWVvL111+zc+fOtoWszGYzl156KWPGjAmVyEMWkymbKVP+xZat17Jte/+WdfL3Sv4RMB6wA68DDcBP+9XzMGHfvn3MmTMHk8mERqPhrLPO4t133x1W6dwLbd45GiP8mEQIvjxYASkQb/vqqI7tV/osh4Ss7MAE9qFSqUnIzKayn5FYAKZJCXhaXNgP1/W7rVMRlUrFFVdcgcVi4c0338Rq9W/d+a1bt/L888+zc+dOJk2axNVXX80111yDxWLh9ddf57PPlIxKfcFiyWXK5JdwOvvn+fDLApFStgAP+LZTkv87VMzupt4v2jpp4elth/xqc4LFyG9Hp/dcZsIEHnjgAaqrqzEajXz88cfMmDFjWKVzb11IKjOAQfTuJhF2qUCavYPcKkvH9ivyj6ExGLHEBD6A3kpiZg77vlyH9HgCmojYGcOYGIRBTcvOKgxjhnbWge4wmUxcffXVvPzyy7z33ntce+21PUblrV27ls8++4zs7Gy+9a1vdVi+ODc3lw8++IC1a9eSkJDAuHHjBuIjDCmioqYyY/rbQN9cvNCLBdI+dXtXW5979bb9kBCiRAix3bdd2O7YL4UQh4UQB4QQi/rTT7gZO3Ys9957L+eeey7nn38+kydP7pCCYzhwzGonQafB4sccEJerCbe7BX03aUxaFUj7c+hpdoIAlbHjea3IP4IxPrFfayMkZuXgsLZQX9k/BS80KozjvG4s6Ry+S8Gmp6dz3nnnceDAAb766oQXvPPgemNjI19++SXjx4/npptu6qA8wPv9X3rppaSlpfHee+8N6QewUGI2982920pvd7K5QBFet9VGvHNBgsmTUson2u8QQowDrsXrMksFVgkhcqWU/frV9WYptLJu3TrmTw3uar233XYbt912GwD3338/6enpwyqd+zGrnSy/B9C7D+EFrwIxGAwd1kL3NDtRmbQI1YnL0+1yUV1UQPyEqf2Q3KtAwOsOi05K7ldbpulJtGytoGV7JeaZ/WvrVGb27NkUFBSwatUqNBoNO3bsoKamhrlz52IweMexNmzYgMfj4eyzz+7wXbdHo9GwePFinn/+ed544w1uv/12jEbjQH6UYU9vNnkycD8wAXgaOBeoCnE698uAN6SUdinlMeAw3hDiU5bWp6PCwkLeeecdrrvuumGVzr3A6iDL5O8Aes+z0FtaWjCZTB32eZqdqMwdn4VqSotxu1yY4vunmOMyMhFC5RtI7x/6nCi0ySaaviod1uGsQgguu+wyYmJiWLFiBY2NjdhsNlatWgV4o602b97MuHHjiIvr2f0YGRnJNddcQ11dHe+88w4ej2cgPoKCD+HvhSyE0OPNyPtH4DdSyr/0q2MhHgJuwTsgvxn4uZSyVgjxLLBBSrnMV+4lYIWU8q0u2lgCLAFISEiYvnz58v6IBEBTU9NJ5nJ/+fGPf0xDQwNqtZrvf//7TJ8+nfr6eh5++GHKy8tJTk7moYce6pCevD8E6zP0p53Wug4J3yaaq7HyLWHvtY7JvA8p/45KPIwQJ1uN27dvx+PxkJub2yZb2kbvc1DJ7BM3j+oDe8hfs4KsSxcTlzaiT5+hlT1v/BN9ZBSjLryyX+0ARBYJEveoKJ7lxjbAQyGhuLb7Q0tLC3V1dSQlJXHw4EGOHz9OXl4eNTU1VFRUMH36dCIiIvxqq6SkhEOHDpGZmUl2du9BE6E6F8FoNxi/u0BYsGDBFilln9wuvTrjfYrjIrzKIwt4BnjHn8aFEKvwWjGdeQD4G/BbQPr+/gm4la7dZF1qOSnlC8ALAGPGjJHz58/3R6weWbduHcFopz07d+7scv9ll10Wkv6C1WZ/2mmtu7/ZCpsOsGBcHvOTTl7fvHOdnJHxHDoEp59+ITrdyXfYffv2ERUVhcViaZOtfOsWtAlGRs8/MZC6rvAQxTo9sSlp/T4XLbs2U7Rvd1DOqcfhpuzRTYy2xhM3f2y/2wuEUFxrwUJKic1mY//+/QghmDlzJhdddFFA9d9//322b9/OnDlzGDu253MbqnMRjHaD8bsbKHpUIEKIf+N1X60AHpZS7g6kcSmlX8HzQogXgY98b4vxJmtsJR0oDaRfhcFDgS8CK8vvEN4qhNCi1UZ3ebylpYWUlJQO+zwtTlTmjtZbRf4x4jMy+xU51UpCljcSq6WhHlPkyUvmBoJKp8Y8K5mmL4px19tRR/k3NjTU0Wg0XHfddRQXF5OXlxfwU7QQgosuuoiKigreffdd4uPj25aRVggdvf26bgJygZ8AXwkhGnxbY39XJBRCtL8LXAG0KqcPgGuFEHohRDYwGtjUn74UwsexFq/bKtvo5yC6vQKdLg4hTr40pZS0tLRgNp9YSUB6pE+BaDuUq8w/2jYA3l8SM1sH0oOTP8gyMxk80LxViRxqT3JyMjNmzOiz+0ar1bJ48WK0Wi1vvPGGMlt9AOhRgUgpVVLKCN8W2W6LkFL212H/ByHELiHETmABcJevzz3AcmAv3uSNP+hvBJZC+DhmtROtUROj9S90uaelbB0OB263u8MgurS5wEMHBdJUU42tuYmEjL5NIOxM60TE/mbmbUUTb0SXHUXL5vJhPZgeCqKiorj66qupra1VBtUHgLDlVJBS3iSlnCilnCSlvFRKWdbu2O+llCOllGOklCvCJaNC/ymwOsj0030FrZMIe56F3l6BtE4iVLdTINXFhQDEjcgIWN6uMEVGYYmLp+JY/yOxWjHPSMJVbcOR3y9DXqELsrKyWLRoEQcPHlRmqoeY4ZeUR2FAOWa1++2+Am8Yb295sNorEE/rLPQOCsSbIC4uPTgKBLyZeSsLguPCAjBOjEfo1TRvPh60NhVOMGvWLKZMmcJnn33G/v37wy3OkEVRIAohw+mRFNsdfisQKd04nTW95sHqqEC8yfVUphMusuqSQowRkf0e8G5PYvZIakqKcdqD41dX6dSYJidg3VmJx+YKSpsKJ2gdVE9NTeWdd95RZqqHCEWBKISMYpsDt4Qsvy2QBkD2mgerSwvEcsICqSkpIraHuR9ul4fPXjvAR8/u8FMu70C6lB6qigr8rtMb5tkpSKeHpg3Kkq+hoHVQXafTsWzZMurrlRUpgo2iQBRCxjGrNwLL3xBe8P7Ae8uD1WEMpMWnQExeBSKlpLqokLj0rhWItdHB+09tY/fnJRTsrqaxxj+LIiEruJFYALo0C/rcGJq+KMHjUOJEQkFUVBQ33HADNpuNV199te0aUggOigJRCBmtCsT/MRCvAukpD5YQoi1fEoCnyYnQqlDpvIkaW+rrsDU3dTn+4XZ6+PAvO6goaGTaokwAyo/691QalZiE3mym/MhBPz+Lf0QuGIGn2UnzN+VBbVfhBCkpKVx77bVUV1fzyiuv+J1KXqF3FAWiEDLyrXZMahUJOv9CeGWrBdKNC6u5uRmTydQhu27nOSBtA+hpJyuQr949TGVhI+fdNp5Zl2aj0ar8ViBCCFJGjaHscHAViD47Cl1WJE2fFyNdSshpqMjJyWHx4sUcP36cV155Bbu957Q6Cv6hKBCFkHGo2c4okz6AdOqtFkjXCfS6T6TYToGU+EJ4O7mwju2oZOeaYiYtSCdnSgJqtYrErEjKj/jvF08ZPYaqogIc1uC6QSLPzsBd71CskBCTm5vL4sWLqaysZPPmzRw8GNyHgeGIokAUQsb+Zht5ZkPvBVuR9Wg00ahUXbu8ulIg7k4KpKakCJ3RhDmmYx6tjR8cJS7NzGlXnlg8J3lkFFVFTTj9HH9IHZ0HUlJ+xL8Fx/xFPzoaXXYkDasLlbGQEDNmzBiWLFmCTqfjtdde4+uvvw63SKc0igJRCAlNUlDucDLG7P/6DJJ69N1MIgROSmMC4GlxdZpEWERc+ogOVo/T7qamtNlreWhPXPLJOVF4PJLKAv8m8yWP8q6/XXowuPMKhBBEnZ+Np8lJ03ol7VuoSUhIYNq0aYwdO5ZPP/2UlStXKhkB+oiiQBRCQpHv0grIAqG+20mE4E1VfZICaXJ2nANSXHjSAHp1SRNSQkJGx9TgyTnebDzlR/1TIAaLhdjUdMoOBX9imj4zEkNeLI2fFbXNrlcIHWq1mquvvprp06ezfv163n//fdxuxfoLFEWBKISEIrxRUYEqkO4mEbpcLmw2W4dEe9LpQTrcbS4sa2MDLfV1J63/UVnYCJysQIwWHdFJJsoCGgfJo+zQgZA8sUadn4V0uGn4JD/obSucjEql4uKLL+ass85i+/btSgLGPqAoEIWQUIyaCLWKVL2298K0rold320erObmZoAOCqRtDohPgdQd907Ii0lN61C3srARY4QWc/TJYyvJOZGUH633WyGk5uZhbWyg/njwB7y1yWYs89Jo/qYce74y6W0gEEKwYMECLrroIo4cOcILL7xAebkSzOAvigJRCAlFqMkzG/2OwHK7mwBnty6spqYmgA4uLE+nRIrNdXXeMtEdB9ArixpJGBHRpSxJ2VHYmpw0VPn35Jky2jsOEgo3FkDkOZmoo/XUvnNYCesdQGbOnMnNN9+Mw+HgH//4Bzt2+J+lYDijKBCFoCOlpBgVeRb/3Vd2uzdXUXep3FsVSHsLpHMixZb6OgBMUSdyYHnckpqSZuIzul4aNTHTu7/VzdUbcSMy0BqMlBzY51f5QFHp1ERfPgpXRQsNq4KXNkWhdzIzM7njjjtIS0vj3XffZcWKFUo6+F5QFIhC0Kl0uGhExZgAxj8cjkoAdN2kMenKhdWtAomMbitjrwePR5IwomsFEpdqQaUWVBb6N5CuUqlJzxtH4e7QPaEa82Ixz0ym8bNibEfqQtaPwslERETw7W9/m9mzZ7Nx40bee+89ZXC9BxQFohB09jd73UGBDKC3WSDdzELvyoXl7qxAGurQm8xodCdyb1lrvH9bLY3OqLUq4tIsVBT4Z4EAZE6aRm1ZCQ1VocvwGnVJDpo4I7XLDyhRWQOMWq3m/PPP5+yzz2bnzp0sW7ZMyebbDYoCUQg6+5u9uYYCs0CqAHocA9Hr9Wi1JwblPc1OEKAyesN4W+rqOrivAGy1Er1JQ0Rc97IkZERQWdjo90B65qQpABTs3O5X+b6g0qmJvXYM7iYnNW/sR7qVeQoDiRCCM888k0svvZTS0lL+9re/sXLlSsUa6YSiQBSCzv5mG5F4SND5F4EFYLeXA1o0mq5XSm5ubj5prWx3gwOVRYtQeQfHWxrqMUVFdyhjq4X4bgbQW0nIiMDe4vJ7ID0uPQNLTCz5O7f5Vb6v6NIjiLl8FPZDddR/ErwswAr+M23aNH784x8zZcoU1q9fz9KlS2ls9N9aHeooCkQh6OxrsjGCwJ7UWqwFQEK3N/ruJhGqLSfcVS31dR3GP9xuD7a6k+d/dCbQgXQhBJmTplK4azseT2ifSM0zkzHPTaHpixKaNinrhoQDs9nMZZddxhVXXEFJSQl//etf2bFjhzJ7HUWBKAQZu8fD3mYrWYEqkJZ8IKnb411aII0O1JGdFEg7C6S2rAXpgYSMjvU6E+hAOkDmxCnYmhqpOHbU7zp9JfriHAxjYqh79zAtu6pC3p9C10yePJk777yThIQE3n33XV577bVhv0iVokAUgsrORit2j2RMAApESjdWayGiBwXS1NTUpQJRRXgViMftxtrU2GEMpFUhdBeB1UpfBtIzJk4BoCDEbiwAoVYRe8NYdBmR1LyxH9vh2pD3qdA18fHxfOc73+H888/n2LFjPPfcc2zZsmXYWiOKAlEIKt/Ue8Ntc/F/nW+brRQpHSC6ViCtaUw6uLAkeJocqH0KxNrYAFJiioppK1JZ2IRKA9GJps5NnkSgA+nm6BgSMrPJ37nVr/L9RaVTE3/zOLQJRqqX7sNRpPjhw4VKpWLOnDl8//vfJyUlhQ8//JBXXnmF2trhp9gVBaIQVL6pbybLqCNa+P9E5nVf0a0F0tUcELUD8NCmQJrrvD/ejhZII4Zo2gbZeyLQgXSA7KkzKD2wD3tLs991+oPKpCX+1gmoLFqq/rkbZ/nA9KvQNbGxsXz729/moosuori4mOeee46SkpJhNflQUSAKQUNKyab6ZmZGmXsv3I4Wa2uEUdcKpKs5IGrfgnKqiM6z0KMB7+TBquJGDB2zmnRLoAPpANlTpuNxuynYtd3vOv1FHakn4dYJoFFR8fxO7AGM2ygEH5VKxcyZM/n+979PRkYGhw4d4uWXXx42+bQUBaIQNI5ZHVQ7XcwKVIG05KNWm4GoLo93lcZE41MgrRZIS4N3MLM1CqvueAsuhwdjjH+5uPoykJ6aOxa9ycyxbZv9rhMMNPFGEu+cjNqkoerFXdgODj/XyWAjOjqaG2+8kby8PGpqanj++ef55JNPhvzSuYoCUQgareMfMyIDUyDWlmOYjFndhvB26cKye8u2KRCfC8scHQ2csCQMJ4ZEeqQvA+kqtZrMydM4tm3zgA+iamINJNw5GU28kap/76FlZ+WA9q9wMkIIkpOT+dGPfsS0adPYsGEDzz77LHv37h2yg+yKAlEIGt/UNxOlUQc0Ax2gxZqP0ZTZ7fEuXVgO719VOwtErdGgM3oHzCuLGlFrVei7npfYJYEOpAPkTJ1Bc10tFfmhD+ftjDpCR8KSSehGRFDz+n4aPysasjeqUwmj0cgll1zCbbfdhslkYvny5bz66qvU1NSEW7SgEzYFIoR4SAhRIoTY7tsu9O0/VwixRQixy/f37HDJqBAYm+qbmR5pQuVnCncAj8eB1VqMyZTdbZmu0pho7CD0alQ678JV3jkgMW1WTGVBI/HpFr8G0Fvpy0B61uRpAORv3+J3nWCiMmqIv3UCxgnx1K/Ip3rZPjw2/yPgFELHiBEjWLJkCYsWLaKwsJDnnnuOzz//HJdr6Hw/4bZAnpRSTvFtH/v2VQGXSCknAjcDr4RPPAV/qXa4ONhiC3gA3WotBjyYjFndlulqEqHafsJ9Ba0KxDuGIj2SKt8aIIHQl4F0c3QMSTmjOfzN1wH1FUxUOjWx1+cRdVE2tn3VVDy7XYnQGiSo1Wrmzp3LD37wA0aPHs2aNWv4+9//ztGjA2+xhoJwK5CTkFJuk1KW+t7uAQxCiJOXklMYVPyv2juIfXZcAD4jTkRg9WaBdE5jorGLNvcVQHO7WehNdXYcNjdx6T3PQO9MXwbSAfJOP5PyI4eoLikKqF4wEUIQcUY6CbdPwmN3UfHX7TR+WaIkYRwkREVFsXjxYq6//npcLhdLly5l2bJllJWd2ulpRLh8pkKIh4BbgAZgM/BzKWVtpzJXAXdKKc/ppo0lwBKAhISE6cuXL++3XF3NeA4loegvWG0G0s4fpZkC1PyFBoTwv65HfoqUy1GJp2lupss6mzZtwmw2M378+LZ96evAGS04PsV7/e5c+jyR6RlknX0Bzccl+WslmfMFWE62XnriyKce1DrIWuD/s5WzpZmdS/9O8pSZpM050+96oUJth8RdKsxVAnuEpHKcB1vMwF/bgTAUfnf+tut2uykpKaGwsBCXy0ViYiJZWVmYTKZ+y9aXugsWLNgipZzRpw6llCHbgFXA7i62y/AG/avxWkG/B17uVHc8cAQY6U9fubm5MhisXbs2KO2Es79gtelvO41Ol8xYt13+38HigOvu2/eAXPfZtB7rPProo/K///1vh30F938maz84LKWU0uPxyCevv0x+9uo/pZRS7vmyRD57x2pZV9ES8LlY88o++eJdn0mPxxNQvXcee0j+/Y6bpNvtCqheqPB4PLJ5Z6UsfWSDLLr3c1n9n/3y8xVrwy1WtwyF312g7ba0tMiVK1fK3/72t/Khhx6S77zzjqyqquqXbH2pC2yWfbzHa/qkdfxXTl1aDp0RQrwIfNTufTrwLvBtKeWREImnECRW1zRg90guSOh6HkdPNDUf6tF91dzcjM1mI9oXngvgsbtRuUVbIkV7SzNulwtTpLf/+korKpUgIjZwz2dCRgR7vyylocpGVILR73rjz1rI0a3fULhrR9vAejgRQmCaGI9hTAyNawpp/KKELI+KWushIs4agSY2sEg5heBjNBo555xzmD17NuvXr2fz5s3s3LmTxMREJk6cSFxcXLhF7JVwRmGltHt7BV7LBCFENPBf4JdSyvVhEE0hQD6urCdOqwl4AqHbbaWhYSfR0dO7LdPqI05JOXG5eBq9MbwqXyr31lnoZt8YSEOlFUucAZU68Ms7OcerhEoCnJyXM302BrOF3etWBdxnKFHp1ESdn03yz2fQkC5p3nyc8ic2U/PmQRzFgYUsK4SGiIgIzj//fH7yk58we/ZsKisrefbZZ3nnnXcG/UqIIbVAeuEPQogpgATygTt8+38IjAL+Twjxf75950kpB/eZHKbY3B5WVTdwRWIM6gDCdwHq67cipYOYmLndlmlVIMnJyW373D4F0mqBtCoQY6sCqbIGZD20Jy7NjCVGT8Guasadnup3PY1Wy/j5C9m64kNqF99ITLL/dQcCTayByvGSMTfOpOnzYpo2ltOy5TiaJJPXUhkbhzbV3OPCWwqhpVWRqFQqpJR888037Ny5k1GjRjF37lxycnIG3fcTNgUipbypm/2/A343wOIo9JF1NY00uz1c2Af3VU3t1wihITqq+/G7srIyYmJiMBpPKIQ2BdIpkWKrBVJfaWV0ZmDRYK0IIcicGM+BjeW4nR7UWv+tmJmXXsWOlZ/w9Vuvc+EPf96n/kONJkpP9CUjiVyYQcvOKlq2VdCwupCGVYWoLFr0I6MxjIxGPyoadYx+0N2whgM6nY758+czb948Nm/ezKZNm3jllVdISkpi5syZTJgwAYNhcLggw2mBKAwBni+uIFWv5YyYwOZcANTWfk1kxEQ0mu6jRsrKyjq4r+CEAmkN460qKkQIFdHJKdiandhbXET20QIByJoYx57PSyg9VMeIcX5mY8Q7J2Tq+RfzzYfvMOuyq4gf0f3s+nCjMmmxzEnBMicFd5MD2/5a7IdrsR2pw7qj0ldGgzbZjDbFfOJvkgmhVYdZ+uGB2WzmrLPO4rTTTmP37t1s2LCBjz76iE8//ZSJEycybdo00tLSwqrkFQWi0Ge21jfzdV0zD49KRRvAjG8Al6uRxsZdZGbc0W0Zq9VKbW0t06Z1HJT2NDqRQqIyei/fivwjxKSmodUbqC33zuHo7MKqfull3I0NxN9+Oypzz2M16WNi0GhV5O+qCkiBAMy89FvsWPkx6//zCpfd/auA6oYLtUWHeUYS5hlJSClxVbRgP1qPs6wZZ1kzzZvKkU5finLhTeaoTWmvWCyoo3SKtRIitFotU6dOZcqUKZSUlLBlyxZ27drF1q1biYuLY9KkSUyaNImYGD8TvwURRYEo9Jm/FlUQpVFzQ0rg0SJ1dZuR0t3j+EdrSuyTLJAGOy79iXU+KvOPkZY3DvC6r6CjArEfO0bFn/4EHg/1739Aym9+g+WMed32q9GpSc+LIX9XFfOuGR3QjdEYEcmsy67myzeWcuDrLxgz9wy/6w4GhBBok8xok04oWemRuGpsXoVS7lUqjuImrDtPLK8rDBq0KSZ0KZY2a0WTZGpLNaPQf4QQpKenk56ezqJFi9izZw87d+5k7dq1rF27lvT0dHQ6HTU1NcTGBvbg01cUBaLQJ4622Pm4sp4fZSRi0QR+k6it/RqVSkdUVPchr10NoAM4Chpw+Dxm1sYGGqsrSczKAU4okIi4Ez7i6udfQOh0pP3pCSr+/CQld93FqLVrUEd073bLnBhP/q5qastaiE0NLLpsxiVXcnjzBla9+FdSx4wlIjY+oPqDDaESaOONaOONMPHEZ/HYXF6F4lMqzrJmmjcfRzp8yxkL0MQZO7rAks3K2EoQMBgMTJ8+nenTp1NXV8euXbvYs2cPxcXFPPPMMyQlJTF27Fjy8vJISkoK2flWFIhCn/hTfjlaIfhuekKf6tfUfk1U5DTU6u4HA8vKyoiMjOwws9ZVY8NVbaMlzxt+2poFNzFrJOCNwDJG6tAZvJe2urKS+g8/JPbGG4lYuBBNcjL537qK2jfeIP7227vtO3tSPJ+/cZA9X5ZwxjW5AX02tUbDhT/8OUvv/TErnv0zV/7yYTTtEkEOFVQGDfqsKPRZJwIopEfiru1krZQ2Yd3VzlrRqzsqlRQzYujkFxxwoqOjOeOMMzjjjDNYsWIF0dHR7Nu3j3Xr1rFu3ToiIyPJyckhJyeH7OxsInp4cAoURYEoBMyKyjrePl7LXZlJJOoDvzHW12+jqWkvo0c90GO5rgbQbYe9EVct8V4FUulTIAlZ3smIDZVWouJPuK/MKz5BaDTE3nYrAMbx4zGffjo1/15K7E03oeommsUcrWfMnGT2fFHKtEWZmKMCm5QYk5LGObd9n0+ee5KPnnqMS+66D7Vm6CmRzgiVQBNnRBNnxDihnbVid+M8fsJScZY307KtArnBa63koKJ82zftlIoFbYpirQSK0Whk7ty5zJ07l6amJg4cOMCRI0c4cOAA27dvByAxMZHs7GwyMzNJT0/vV3+KAlEIiEqHk7sPFDPRYuSurK6XoO2NY/nPotXGkJq6uNsyDoeDqqoqJkyY0GG//VAd6kgdTrPXVVWRfxRLXHyHWehpud7BRFdtLYaNG4levBhtYmJbG3FLllB4883Uv/suMddd160M08/P5MDXZWxfWcjpV40O+HOOP2shTpuN1S//jQ+ffIwLf3Q3OkPfo8NOZVR6NfqMSPQZJ8KrpZS4a+04y5s5+NVu0g1mnGXNWPdWe2eH4bNW2lkq2mQz2mQTKr1y6+oNi8XS5ubyeDyUl5dz9OhRjh49ypYtW9i4cWO/+1C+BQW/cXkkP9tfRKPLzTNTRqJTBT7Tu6FhJ9XV6xiZczcaTfdjC7t27QIgIyOjbZ/0SGyH6zCOiwNxQoEkZnqtD7fTQ1OdvS2Et/GTTxBuN9FXfatD26ZZMzFOnkzViy8SddllqHxJ7DoTnWhi9Kwkdn9ewrRFmRjbZf/1lymLLgJgzT+f57UHfs4lP/slcWkjAm5nKCKEQBNrQBNroLZCMnm+NxCiN2sFQB2jR5toQpNkRptoQptkQpNoQqVXBu27QqVSkZqaSmpqKvPmzcPlclFeXk5RUREPP/xwn9tVFIiCXzg9ku/vLWBldQO/H53GWEvfnqSP5f8VjSaK9PQbuy3jcrn4/PPPSUtLIzv7RJ4sZ0kT0urCMDoa6kpxOuzUlBYzepY3kquuogXkiQis+vc/wJWagn7s2A7tCyFIvPvnFNz0bSqeeork++/vVpYZF2RxaNNxPnvtAItunxDQAlWtTFl0ETEpafz3mT/w6i/vYu7V1zPtgktRa5SfX1d0a63U2U8olYoWXMdbsB2ug3Yp69XRejQ+heJVMN6/KoNyrtuj0WjaIrr61U6Q5FEYwjS53PxoXyErqup5aGQqt/Vx4Lys/D2qqlaRnf1TNJruB/K2b99OfX09F198cQf/d+v4h35UNGyG6sICpMfTNoC+c10xKo0gLTcGR0EB1u3bsV5xeZc+dNPMmcTceCO1rywj8rzzMM3oejZ8TLKZuVeO4qu3D/PNx/nMurj7xI89kTlpCjc9/gyr/vFXPl/2MnvWrWL6xZeTd/pZaHXKcje9IYRAE2NAE2PwWqA+pFviqrHiOt6Cs6KlTbE0Ha0D1wnFoorQesdm4r2buRwcpU1oYg2KcukHyplT6JHV1Q3cc6CIUruT341O63PUVXn5++zd+wtiYuaSmdF99FN762PUqFFt+z0ON9YdVd6BVV8SxdJD+wFIyMqhscbG/q/KGDcvFUuMnspXPwIhsM2c2W1fiT+7i6bPPqP0vl+S8a9/oUtP67LclHNGUFPaxDcfHUNv0jBpfnqfLJGIuHguv+fXHN68gfVvvML//v4Mn7/yMqPnnM6YOWeQljcOjS5wN9lwRqgF2gQT2gQT7W1i6ZG4a2w4fYrFVWXFVW3Ftr8GT5OTFNRUbN/mbcOgRhNtQB2j91owMYYOf1UWrTKQ3w2KAlE4CZvbw8dV9fxFWti38yijTXo+mDY64OVqAez24xQUvkhR0b+Jjp7J5Ekvdhu6a7VaWbZsGQ0NDVx66aVtP1qPw031v/bgPN5M7HV5ADQU5bPj0w9IzB5JVGISn79+EIBpizJxNzVT//77mGbNwtPDhCqVyUTaH/9A4ZI7yL/qKtKe/DPmuSdPbBRCMP/6PFoanHy5/BBHt1Vy2pWjSMyKCPjGIoRg9My5jJoxh+K9u9i5+lP2f/kZu1Z/ilqrJXlkLvEZWcSlpRObOoKY1DTM0TGKuytAhEq0WRvG8R0nunpsLjZ++iWTs8fjrrXjqrPhrrXjrrVhP1qPtLs7NqYWqMxa1GYtKkvrX92J10YNwqBGZdCgbQZ3kwOVXoMIII/aqUrYViQMNqmZafJ793+/3+243W7U6o4Dca23CCGl77Xsdp9KSkAgkCBBLT0IKVFLDyo8qKUHtQSVx/taOp3o1CrUUnpz65/0dXS6QcnW3roo49vpdLnRdr7hSIFbCpxCg12lw67S4lBpcag0tGj01OrNVBssFETGUBARg0utJra5iTNKijiz8BhaKRFSdOxRgEAihQehdoHKhUpvRaVvwWOowBBXiza2BIQHR/komvfPAY8WD+AREjfgBhxC0iIkdSo3AkGGW0eSW4caDSYZQRxJWIhkH1spdefjslupryhFazASnzsZjyea+pIxmBPKSI3aSvLb69HUNlL53Ys5HGckMS2ZZo8Vp+x6soGhsoGRL63BWF5HS3osdeNGYEuKxhFjxm3U4dGqvR9Wgv14Co3HRiLdWlR6K7rIOjSmFtR6GyqNE6Fyg/AghAQhvX9VsuN564TH5cJe04CtrgFHfSPOZivS3fEmptKoUem0qLRaVGo1Qq1CqNQIlcq7qVUnlJkQvn4ECOH7ngQtNismkwnh3dH6Bfrq+Wq0b+Mk5XjifesnEn25fciTz0JTc/NJyxaLLs9WV+35d7B9a03NzVi6SWmjkVoM0ogBEwZpRCf16NCjlTp06NFJHVr0aHp5/vbgxoULF07cuHELNx7cePDgpovXwoMHD3aHHa1Oi8SDbPvX/nX7vb79wvvOarViMBqQnU9KN/d5ieSqR3/d5xUJh4wC0Y4ZJ+P+/lq4xegXQnpQ4UaDGxVu1HgQeBDIDpuq3b72rwE8bTVVeFC3teRGg130nMFTLZ1kcYwx7GMy2xnHLlS+dqUUyFblJQVSqhBColK5vTfJLmhpiaSuLpmS4nHYbD1PXjJKHameGGY5R2HmhJwe6abBWc2e2q8objlw4lypU9FZLkOovI4Lva2GaduexGivoSQW/naRmoPp/lsHBrvk3G2SmYc85Bb3vFCOU2OkKm4SlQlTaIjIwKGP9rsff5BSgmzG465BemrB04KULUiPFaQVKZ0gnYALKV1tr8ETVDkUekYttOjVRnQqA1qVHq3Qef92uelQCw1qofX99W2qE69VIjwRZCMeP1NRIGkZafLH9/+w3+04nS602hNPFhJAeG/P3jPlu1UL2t22RZu+l+LEa4Tvhi4EHiFwC1Xb5n0vcHhAaDS+/aLdMVWbyvC09u97UvS09S2Qra99ZdTSA243GhWopNfKEUjU0o1eOtF7nOg9Du9r6UAvHRg8NmJdDUR5GlHhAeG9Ebk9TtQaNUL4rAzwHmunMDxuDdKjRno0SLcGj9OA22HGWm9CLSJaT0OHp0QVAjUCFaCWAh0qdKhosVkxGPV45ImnMZvKhhSytRVUGg0anYGGxgaiIyJR4UanaUItrUjpxiOgdmwaLp3AJd2UlJWQmZqJRWVEpzp5Il+XV7/NjramAU1tEyqbA+F0IaRsuxZo91cCbo8Gl0uP26NDejTe70sKkL7vUKp88vfwWztJ17V7cu5WD544L217pO9KFdL30Cnbnj4bG5uIsJi8LfvKtZVB+tSPt7xsV88/Q6CLz9ZlPdnl7g4WwUmnyvcQ46e78KRSXVRramrGYjnZApG9feZeZOi8Jrn/d1cJUqBC0NJsxWK2IIQK0dU/KfCW7PxPhc1qx2Q88QAm2v3f/kO1vpISFj3yk8G5JvpAbsqa6MFvc6DXZg6kjr9lB/r7HMwM5nMxFH53wWr3VFoTfeiP8igoKCgohARFgSgoKCgo9AlFgSgoKCgo9AlFgSgoKCgo9AlFgSgoKCgo9AlFgSgoKCgo9AlFgSgoKCgo9AlFgSgoKCgo9IkhMxNdCNEIHOi1YO9EAfVBaCec/cUDVb2W6p3+yNaXuoHU8bdssM7FUGAwn4uB/t2F6lwE43MM9O9ujJSybwul93UG4mDb6Mdsyk7tvDDAcge9v8FwLvpSN5A6/pYN1rkYCttgPhdh+N2F5FwE43OE4XenzEQPIh8O8f4CoT+y9aVuIHUG83lTCJyh8n0G43MM9O+uzwwlF9Zm2deEYEMM5VycQDkXJ1DOxQmUc3GC/pyLoWSBvBBuAQYRyrk4gXIuTqCcixMo5+IEfT4XQ8YCUVBQUFAYWIaSBaKgoKCgMIAoCkRBQUFBoU+cEgpECHG+EOKAEOKwEOK+Lo7fIITY6du+EkJMbncsXwixSwixXQixeWAlDy1+nJf5Qoh632ffLoT4dTjkDDVCiJeFEBVCiN3dHBdCiGd852mnEGLaQMs4UAghRggh1goh9gkh9gghftJFmWFzPgCEEGohxDYhxEddHBs250IIcZfvmtgthHhdiI5rXPfpXAxk7HUf45rVwBEgB9ABO4BxncqcBsT4Xl8AbGx3LB+ID/fnCNN5mQ98FG5ZB+BcnAlMA3Z3c/xCYAXelTzntL8+htoGpADTfK8jgINdXBfD5nz4Pu/PgNe6+i0Ml3MBpAHHAKPv/XLglv6ei1PBApkFHJZSHpVSOoA3gMvaF5BSfiWlrPW93QCkD7CM4aDX8zJckFJ+DtT0UOQyYKn0sgGIFkKkDIx0A4uUskxKudX3uhHYh/fm0Z5hcz6EEOnARcA/uikybM4FoAGMQggNYAJKOx0P+FycCgokDShq976Yk38Q7bkNrxZtRQL/E0JsEUIsCYF84cLf8zJXCLFDCLFCCDF+YEQbdAR6DQ0JhBBZwFRgY6dDw+l8PAXcA3i6OT4szoWUsgR4AigEyoB6KeX/OhUL+FycCgpEdLGvy9hjIcQCvArk3na7T5dSTsPr2vqBEOLM4IsYFvw5L1uBTCnlZOAvwHuhFmqQ4vc1NFQQQliAt4GfSikbOh/uosqQOx9CiIuBCinllp6KdbFvKJ6LGLwWRjaQCpiFEDd2LtZF1R7PxamgQIqBEe3ep3Oy6YUQYhJeM/UyKWV1634pZanvbwXwLl7Xz1Cg1/MipWyQUjb5Xn8MaIUQ8QMn4qDBr2toqCCE0OJVHq9KKd/poshwOR+nA5cKIfLxunjPFkIs61RmuJyLc4BjUspKKaUTeAfv2HF7Aj4Xp4IC+QYYLYTIFkLogGuBD9oXEEJk4D0hN0kpD7bbbxZCRLS+Bs4DuozUOQXx57wkCyGE7/UsvN939UktDX0+AL7tizKZg9d8Lwu3UKHA932/BOyTUv65m2LD4nxIKX8ppUyXUmbh/X2skVJ2fuoeFucCr+tqjhDC5LtGFuIdH2tPwOdCExpZg4eU0iWE+CHwKd7Io5ellHuEEHf6jv8d+DUQBzznu1+6pDe3SxLwrm+fBnhNSvlJGD5G0PHzvFwFfE8I4QKswLXSF24xlBBCvI434ixeCFEMPAhooe08fIw3wuQw0AJ8JzySDginAzcBu4QQ23377gcyYFiej5Po9BsZFudCSrlRCPEWXre2C9gGvNDfc6GkMlFQUFBQ6BOnggtLQUFBQWEQoigQBQUFBYU+oSgQBQUFBYU+oSgQBQUFBYU+oSgQBQUFBYU+oSgQBYUu8M2heUMIcUQIsVcI8bEQIlcIkSWEsIoTGY63++bhdK4/VQjxD9/rh4QQd3dR5l9CiKs67WvqQSadEOJzXy4jBYWwo1yICgqd8E20ehf4t5TyWt++KXjnFRUBR6SUU3pp5n7gd8GUS0rpEEKsBhYDrwazbQWFvqBYIAoKJ7MAcPomVwEgpdwupfzCn8q+7AeTpJQ7ujh2uy+xpbGXNn7TzsIpEUL803foPeAGvz+JgkIIURSIgsLJTAB6SsA3st3N/a9dHJ9BFylzfJkDLgEul1Jafbv/2N4d1lpWSvlrn5VzFt70M8/6Du0GZgb6gRQUQoHiwlJQCJzeXFgpQGWnfTfhTVZ3uS+ZXSu/kFK+1fqm/RiIz5X2KvBka0ZZKaVbCOEQQkT41vtQUAgbigWioHAye4Dp/ahvBQyd9u0GsghssbOHgGIp5T877dcDtr4Kp6AQLBQFoqBwMmsAvRDi9tYdQoiZQoiz/Ky/DxjVad824A7gAyFEam8N+NayOBf4caf9cUBlJytGQSEsKApEQaETvozFVwDn+sJ49+C1BvxaJ0JKuR+Ial1KoN3+L4G7gf/6sS7Lz/Eu/LPJNz7yG9/+BXizpioohB0lG6+CQggQQtwFNEopu1uLu6/tvgP8Ukp5IJjtKij0BcUCUVAIDX8D7MFs0Ddh8T1FeSgMFhQLREFBQUGhTygWiIKCgoJCn1AUiIKCgoJCn1AUiIKCgoJCn1AUiIKCgoJCn1AUiIKCgoJCn/h/5SRkj+Z3S/kAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "############ Plots ###################\n",
    "# Plot the mean Vihc patterns for the reference model\n",
    "CF_rep=np.tile(CF_tl, (len(L),1))\n",
    "plt.semilogx(CF_rep.T/1e3, 1e3*ihc_target_mean.T)\n",
    "plt.xlim(0.25,8.), plt.grid(which='both'), \n",
    "plt.xticks(ticks=(0.25, 0.5, 1., 2., 4., 8.) , labels=(0.25, 0.5, 1., 2., 4., 8.))\n",
    "#plt.ylim(-59, -49.5)\n",
    "plt.xlabel('CF (kHz)')\n",
    "plt.ylabel('Mean of V_m (mV)')\n",
    "plt.title('Dierich2020 IHC model')\n",
    "plt.legend(L.astype(int), frameon=False)\n",
    "plt.show()\n",
    "\n",
    "# Plot the mean Vihc patterns for the CoNNear model\n",
    "CF_rep=np.tile(CF_connear, (len(L),1))\n",
    "plt.semilogx(CF_rep.T/1e3, 1e3*ihc_connear_mean.T)\n",
    "plt.xlim(0.25,8.), plt.grid(which='both'), \n",
    "plt.xticks(ticks=(0.25, 0.5, 1., 2., 4., 8.) , labels=(0.25, 0.5, 1., 2., 4., 8.))\n",
    "#plt.ylim(-59, -49.5)\n",
    "plt.xlabel('CF (kHz)')\n",
    "plt.ylabel('Mean of V_m (mV)')\n",
    "plt.title('Dierich2020 IHC model - CNN')\n",
    "plt.legend(L.astype(int), frameon=False)\n",
    "plt.show()\n",
    "\n",
    "del ihc # remove the connear model variable to free-up some memory"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### IHC AC-DC ratio\n",
    "Compare the ratio of the AC and DC components of the IHC responses across CF. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "modeldir = 'Dierich2020/' # Dierich et al. CNN model directory\n",
    "# Define the IHC model hyperparameters\n",
    "context_left = 256\n",
    "context_right = 256\n",
    "Nenc = 3 # number of layers in the encoder - check for the input size\n",
    "# Load the 1-channel IHC model for this section - the results are computed for individual CFs\n",
    "N_cf = 1\n",
    "ihc = load_connear_model(modeldir,json_name=\"/ihc_1cf.json\",weights_name=\"/ihc.h5\",name=\"ihc_model\")\n",
    "#ihc.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "running human auditory model 2018: Verhulst, Altoe, Vasilkov\n",
      "cochlear simulation: done\n",
      "10/10 [==============================] - 0s 22ms/step\n"
     ]
    }
   ],
   "source": [
    "# Define the pure tone stimuli\n",
    "N = 10 # number of frequencies to simulate between 0.15 and 8 kHz\n",
    "f_tones = np.logspace(np.log10(150),np.log10(8000),num=N) # pick N frequencies in logarithmic spacing\n",
    "# match the tone frequencies to the corresponding CFs\n",
    "for j, f_tone in enumerate(f_tones):\n",
    "    fno, _ = min(enumerate(CF_connear), key=lambda x: abs( x [1]- f_tone))\n",
    "    f_tones[j] = CF_connear[int(fno)]\n",
    "    \n",
    "L = [80.] # 80 dB SPL\n",
    "stim_dur = 80e-3 # duration of the stimulus\n",
    "silence_left = 35.2e-3 # silence before the onset of the stimulus - 12.8 ms correspond to 256 samples (context)\n",
    "silence_right = 12.8e-3 # silence after the stimulus - 256 samples\n",
    "win_dur = 5.0e-3 # 5ms long ramp window for gradual onset\n",
    "\n",
    "# indicate the time points of the response for computing the AC and DC components\n",
    "t_ac_start = silence_left + 50e-3 # 50 ms after the stimulus onset\n",
    "t_ac_dur = 20e-3 # 20 ms after t_ac_start\n",
    "t_dc_start = 7.4e-3 # 7.4 ms after the left context (15 ms before the stimulus onset)\n",
    "t_dc_dur = 10e-3 # 10 ms after t_dc_start\n",
    "\n",
    "# region of the response for computing the AC component (50 - 70 ms after the stimulus onset)\n",
    "ac_start = int(t_ac_start * fs_connear) - context_left\n",
    "ac_end = ac_start + int(t_ac_dur * fs_connear)\n",
    "ac_reg = np.arange(ac_start,ac_end,1)\n",
    "# region of the response for computing the DC component (5 - 15 ms before the stimulus onset)\n",
    "dc_start = int(t_dc_start * fs_connear)\n",
    "dc_end = dc_start + int(t_dc_dur * fs_connear)\n",
    "dc_reg = np.arange(dc_start,dc_end,1)\n",
    "\n",
    "# make stimuli\n",
    "t = np.arange(0., stim_dur, 1./fs_tl)\n",
    "winlength = int(2*win_dur * fs_tl)\n",
    "win = sp_sig.windows.bartlett(winlength) # double-sided ramp\n",
    "total_length = int(silence_left * fs_tl) + len(t) + int(silence_right * fs_tl) \n",
    "stim = np.zeros((len(f_tones), total_length))\n",
    "for j, f_tone in enumerate(f_tones):\n",
    "    stim_sin = np.sin(2 * np.pi * f_tone * t) # generate the pure tone\n",
    "    # apply ramp window\n",
    "    stim_sin[:int(winlength/2)] = stim_sin[:int(winlength/2)] * win[:int(winlength/2)]\n",
    "    stim_sin[-int(winlength/2):] = stim_sin[-int(winlength/2):] * win[int(winlength/2):]\n",
    "    stimrange = range(int(silence_left * fs_tl), int(silence_left * fs_tl) + len(stim_sin))\n",
    "    stim[j, stimrange] = p0 * 10**(L[0]/20) * stim_sin / rms(stim_sin) # calibrate \n",
    "\n",
    "############ Verhulstetal ############\n",
    "# get the outputs of the reference model \n",
    "output = tl_vbm(stim, L)\n",
    "\n",
    "CF_tl = output[0]['cf'][::ds_rate] # center frequencies\n",
    "window_length = int(total_length / ds_factor) # window length of the downsampled outputs\n",
    "tl_target = np.zeros((len(f_tones), window_length * ds_factor, N_cf)) # pre-allocate\n",
    "ihc_target = np.zeros((len(f_tones), window_length, N_cf))\n",
    "for j, f_tone in enumerate(f_tones):\n",
    "    # find the CF closest to the stimulus frequency\n",
    "    No, _ = min(enumerate(CF_tl), key=lambda x: abs(x[1] - f_tone))\n",
    "    No = int(No)\n",
    "    # extract the outputs only for the specific CF\n",
    "    tl_target[j,:,:] = output[j]['v'][:,::ds_rate][:,[No]] # cochlear output\n",
    "del output # remove the output variable to free-up some memory\n",
    "\n",
    "############ Dierichetal ############\n",
    "parent_dir = os.getcwd() # get current directory \n",
    "os.chdir(\"./\" + modeldir + \"reference/\") # navigate to the model's folder\n",
    "eng = matlab.engine.start_matlab() # start the matlab engine\n",
    "\n",
    "eng.workspace['Fs'] = fs_tl # pass the variables to the matlab workspace\n",
    "magic_constant = 0.118\n",
    "tl_in = magic_constant*tl_target\n",
    "\n",
    "for i in range(len(L)):\n",
    "    for fi in range(N_cf):\n",
    "        v = tl_in[i,:,fi].tolist()\n",
    "        v = matlab.double(v)\n",
    "        eng.workspace['v'] = v\n",
    "        ihc_out = eng.eval('IHC_model(v,Fs);', nargout=1) # simulate the IHC output of the reference model\n",
    "        ihc_out = np.array(ihc_out[0])\n",
    "        ihc_target[i,:,fi] = sp_sig.resample_poly(ihc_out, fs_connear, fs_tl)\n",
    "        \n",
    "eng.quit()\n",
    "os.chdir(parent_dir) # go back to the original folder\n",
    "    \n",
    "# remove context from the output\n",
    "ihc_target = ihc_target[:, context_left:-context_right,:]\n",
    "# downsample to 20 kHz\n",
    "tl_target = sp_sig.resample_poly(tl_target, fs_connear, fs_tl, axis=1)\n",
    "# apply proper scaling for feeding to the CoNNear\n",
    "tl_target = tl_target * cochlea_scaling # scaling for the cochlear input\n",
    "\n",
    "# compute the AC component for each frequency\n",
    "acm_target = (np.max(ihc_target[:,ac_reg],axis=1)-np.min(ihc_target[:,ac_reg],axis=1))/2\n",
    "ac_target = acm_target/np.sqrt(2.)\n",
    "# compute the DC component for each frequency\n",
    "dcm_target = (np.max(ihc_target[:,dc_reg],axis=1)-np.min(ihc_target[:,dc_reg],axis=1))/2\n",
    "dc_target = (np.min(ihc_target[:,ac_reg],axis=1)+acm_target)-(np.min(ihc_target[:,dc_reg],axis=1)+dcm_target)\n",
    "\n",
    "############ CoNNear #################\n",
    "# check the time dimension size\n",
    "if tl_target.shape[1] % 2**Nenc: # input size needs to be a multiple of 8\n",
    "    Npad = int(np.ceil(tl_target.shape[1]/(2**Nenc)))*(2**Nenc)-tl_target.shape[1]\n",
    "    tl_target = np.pad(tl_target,((0,0),(0,Npad),(0,0))) # zero-pad\n",
    "# simulate\n",
    "ihc_connear = ihc.predict(tl_target, verbose=1)\n",
    "ihc_connear = ihc_connear / ihc_scaling # scaling for the IHC output\n",
    "\n",
    "# compute the AC component for each frequency\n",
    "acm_connear = (np.max(ihc_connear[:,ac_reg],axis=1)-np.min(ihc_connear[:,ac_reg],axis=1))/2\n",
    "ac_connear = acm_connear/np.sqrt(2.)\n",
    "# compute the DC component for each frequency\n",
    "dcm_connear = (np.max(ihc_connear[:,dc_reg],axis=1)-np.min(ihc_connear[:,dc_reg],axis=1))/2\n",
    "dc_connear = (np.min(ihc_connear[:,ac_reg],axis=1)+acm_connear)-(np.min(ihc_connear[:,dc_reg],axis=1)+dcm_connear)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "############ Plots ###################\n",
    "# Plot the logarithmic decrease of the AC/DC ratio across frequency\n",
    "plt.loglog(f_tones/1e3, ac_target/dc_target,'o-')\n",
    "plt.loglog(f_tones/1e3, ac_connear/dc_connear,'o-')\n",
    "plt.xlim(0.1,10.), plt.grid(which='both'), \n",
    "plt.xticks(ticks=(0.1,1,10) , labels=(0.1,1,10))\n",
    "plt.xlabel('CF (kHz)')\n",
    "plt.ylabel('AC component / DC component')\n",
    "plt.title('AC-DC ratio')\n",
    "plt.legend(['Dierich2020 IHC model','Dierich2020 IHC model - CNN'], frameon=False)\n",
    "plt.show()\n",
    "\n",
    "del ihc # remove the connear model variable to free-up some memory"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### IHC level growth\n",
    "Compare the growth of the half-wave rectified IHC receptor potential as a function of sound level."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "modeldir = 'Dierich2020/' # Dierich et al. CNN model directory\n",
    "# Define the IHC model hyperparameters\n",
    "context_left = 256\n",
    "context_right = 256\n",
    "Nenc = 3 # number of layers in the encoder - check for the input size\n",
    "# Load the 1-channel IHC model\n",
    "N_cf = 1\n",
    "ihc = load_connear_model(modeldir,json_name=\"/ihc_1cf.json\",weights_name=\"/ihc.h5\",name=\"ihc_model\")\n",
    "#ihc.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "running human auditory model 2018: Verhulst, Altoe, Vasilkov\n",
      "cochlear simulation: done\n",
      "11/11 [==============================] - 0s 18ms/step\n"
     ]
    }
   ],
   "source": [
    "# Define the pure tone stimuli\n",
    "f_tone = 4e3 # frequency of the pure tone\n",
    "# match the tone frequency to the corresponding CF\n",
    "fno, _ = min(enumerate(CF_connear), key=lambda x: abs( x [1]- f_tone))\n",
    "f_tone = CF_connear[int(fno)]\n",
    "\n",
    "L = np.arange(0.,101.,10.) # levels from 0 to 100dB SPL\n",
    "stim_dur = 80e-3 # duration of the stimulus\n",
    "silence_left = 35.2e-3 # silence before the onset of the stimulus - 12.8 ms correspond to 256 samples (context)\n",
    "silence_right = 12.8e-3 # silence after the stimulus - 256 samples\n",
    "win_dur = 5.0e-3 # 5ms long ramp window for gradual onset\n",
    "\n",
    "# indicate the time points of the response for computing the AC and DC components\n",
    "t_ac_start = silence_left + 50e-3 # 50 ms after the stimulus onset\n",
    "t_ac_dur = 20e-3 # 20 ms after t_ac_start\n",
    "t_dc_start = 7.4e-3 # 7.4 ms after the left context (15 ms before the stimulus onset)\n",
    "t_dc_dur = 10e-3 # 10 ms after t_dc_start\n",
    "\n",
    "# region of the response for computing the AC component (50 - 70 ms after the stimulus onset)\n",
    "ac_start = int(t_ac_start * fs_connear) - context_left\n",
    "ac_end = ac_start + int(t_ac_dur * fs_connear)\n",
    "ac_reg = np.arange(ac_start,ac_end,1)\n",
    "# region of the response for computing the DC component (5 - 15 ms before the stimulus onset)\n",
    "dc_start = int(t_dc_start * fs_connear)\n",
    "dc_end = dc_start + int(t_dc_dur * fs_connear)\n",
    "dc_reg = np.arange(dc_start,dc_end,1)\n",
    "\n",
    "# make stimulus\n",
    "t = np.arange(0., stim_dur, 1./fs_tl)\n",
    "stim_sin = np.sin(2 * np.pi * f_tone * t) # generate the pure tone\n",
    "# apply hanning window\n",
    "winlength = int(2*win_dur * fs_tl)\n",
    "win = sp_sig.windows.bartlett(winlength) # double-sided ramp\n",
    "stim_sin[:int(winlength/2)] = stim_sin[:int(winlength/2)] * win[:int(winlength/2)]\n",
    "stim_sin[-int(winlength/2):] = stim_sin[-int(winlength/2):] * win[int(winlength/2):]\n",
    "total_length = int(silence_left * fs_tl) + len(stim_sin) + int(silence_right * fs_tl)\n",
    "stim = np.zeros((len(L), total_length))\n",
    "stimrange = range(int(silence_left * fs_tl), int(silence_left * fs_tl) + len(stim_sin))\n",
    "for i in range(len(L)):\n",
    "    stim[i, stimrange] = p0 * 10**(L[i]/20) * stim_sin / rms(stim_sin) # calibrate \n",
    "\n",
    "############ Verhulstetal ############\n",
    "# get the outputs of the reference model \n",
    "output = tl_vbm(stim, L)\n",
    "\n",
    "CF_tl = output[0]['cf'][::ds_rate] # center frequencies\n",
    "# find the CF closest to the stimulus frequency\n",
    "No, _ = min(enumerate(CF_tl), key=lambda x: abs(x[1] - f_tone))\n",
    "No = int(No)\n",
    "\n",
    "window_length = int(total_length / ds_factor) # window length of the downsampled outputs\n",
    "tl_target = np.zeros((len(L), window_length * ds_factor, N_cf)) # pre-allocate\n",
    "ihc_target = np.zeros((len(L), window_length, N_cf))\n",
    "for i in range(len(L)):\n",
    "    # extract the outputs only for the specific CF\n",
    "    tl_target[i,:,:] = output[i]['v'][:,::ds_rate][:,[No]] # cochlear output\n",
    "del output # remove the output variable to free-up some memory\n",
    "    \n",
    "############ Dierichetal ############\n",
    "parent_dir = os.getcwd() # get current directory \n",
    "os.chdir(\"./\" + modeldir + \"reference/\") # navigate to the model's folder\n",
    "eng = matlab.engine.start_matlab() # start the matlab engine\n",
    "\n",
    "eng.workspace['Fs'] = fs_tl # pass the variables to the matlab workspace\n",
    "magic_constant = 0.118\n",
    "tl_in = magic_constant*tl_target\n",
    "\n",
    "for i in range(len(L)):\n",
    "    for fi in range(N_cf):\n",
    "        v = tl_in[i,:,fi].tolist()\n",
    "        v = matlab.double(v)\n",
    "        eng.workspace['v'] = v\n",
    "        ihc_out = eng.eval('IHC_model(v,Fs);', nargout=1) # simulate the IHC output of the reference model\n",
    "        ihc_out = np.array(ihc_out[0])\n",
    "        ihc_target[i,:,fi] = sp_sig.resample_poly(ihc_out, fs_connear, fs_tl)\n",
    "        \n",
    "eng.quit()\n",
    "os.chdir(parent_dir) # go back to the original folder\n",
    "    \n",
    "# remove context from the output\n",
    "ihc_target = ihc_target[:, context_left:-context_right,:]\n",
    "# downsample to 20 kHz\n",
    "tl_target = sp_sig.resample_poly(tl_target, fs_connear, fs_tl, axis=1)\n",
    "# apply proper scaling for feeding to the CoNNear\n",
    "tl_target = tl_target * cochlea_scaling # scaling for the cochlear input\n",
    "\n",
    "dcm_target = np.mean(ihc_target[:,dc_reg],axis=1) # compute the DC component \n",
    "vihc_target = np.zeros((ihc_target.shape))\n",
    "for i in range(len(L)):\n",
    "    vihc_target[i,:] = ihc_target[i,:] - dcm_target[i] #half-wave rectify the response by substracting the DC component\n",
    "vihc_target_rms = rms(vihc_target,axis=1)\n",
    "\n",
    "############ CoNNear #################\n",
    "# check the time dimension size\n",
    "if tl_target.shape[1] % 2**Nenc: # input size needs to be a multiple of 8\n",
    "    Npad = int(np.ceil(tl_target.shape[1]/(2**Nenc)))*(2**Nenc)-tl_target.shape[1]\n",
    "    tl_target = np.pad(tl_target,((0,0),(0,Npad),(0,0))) # zero-pad\n",
    "# simulate\n",
    "ihc_connear = ihc.predict(tl_target, verbose=1)\n",
    "ihc_connear = ihc_connear / ihc_scaling # scaling for the IHC output\n",
    "\n",
    "dcm_connear = np.mean(ihc_connear[:,dc_reg],axis=1) # compute the DC component \n",
    "vihc_connear = np.zeros((ihc_connear.shape))\n",
    "for i in range(len(L)):\n",
    "    vihc_connear[i,:] = ihc_connear[i,:] - dcm_connear[i] #half-wave rectify the response by substracting the DC component\n",
    "vihc_connear_rms = rms(vihc_connear,axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "############ Plots ###################\n",
    "# Plot the RMS of the half-wave rectified response across level\n",
    "plt.plot(L, 1e3*vihc_target_rms,'o-')\n",
    "plt.plot(L, 1e3*vihc_connear_rms,'o-')\n",
    "plt.grid(which='both'), \n",
    "#plt.yticks(ticks=np.arange(0.,10.,1.))\n",
    "plt.xlabel('Stimulus level (dB-SPL)')\n",
    "plt.ylabel('Half-wave rectified $\\mathregular{V_{IHC}}$ (mV)')\n",
    "plt.title('RMS of half-wave rectified $\\mathregular{V_{IHC}}$')\n",
    "plt.legend(['Dierich2020 IHC model','Dierich2020 IHC model - CNN'], frameon=False)\n",
    "plt.show()\n",
    "\n",
    "del ihc # remove the connear model variable to free-up some memory"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ANF stage\n",
    "The input dimensions for the CNN models are (b x L x Ncf), where b is the batch-size (for loading multiple stimuli simultaneously), L is the input length (including the context) and Ncf are the frequency channels.\n",
    "\n",
    "The anfX.json file can be loaded for the full-channel model (201 channels) or the anfX_1cf.json file for the 1-channel model, where X corresponds to the AN fiber type (h for HSR, m for MSR or l for LSR ANF model). At the first block of each individual section, the necessary model is loaded and the rest of the parameters are defined.\n",
    "\n",
    "Auditory stimuli are generated here and given as inputs to the reference Zilany2014 cochlea-IHC model, to simulate the IHC outputs. These responses are then given as inputs to the Zilany2014 ANF model and its CNN counterpart, to simulate the mean firing rates or the PSTH responses."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ANF firing rates\n",
    "Compare the simulated ANF firing rates across time for tone stimuli presented at 70 dB SPL.\n",
    "**You can change the `f_tone` variable to have tone stimuli of different frequencies, the `f_m` variable for generating amplitude-modulated tones or the `L` variable for having different levels.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"anfh_model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "bmm_input (InputLayer)       (None, 16384, 1)          0         \n",
      "_________________________________________________________________\n",
      "model_3 (Model)              (None, 8192, 1)           1250177   \n",
      "=================================================================\n",
      "Total params: 1,250,177\n",
      "Trainable params: 1,250,177\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Model: \"anfm_model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "bmm_input (InputLayer)       (None, 16384, 1)          0         \n",
      "_________________________________________________________________\n",
      "model_3 (Model)              (None, 8192, 1)           1250177   \n",
      "=================================================================\n",
      "Total params: 1,250,177\n",
      "Trainable params: 1,250,177\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Model: \"anfl_model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "bmm_input (InputLayer)       (None, 16384, 1)          0         \n",
      "_________________________________________________________________\n",
      "model_3 (Model)              (None, 8192, 1)           1250177   \n",
      "=================================================================\n",
      "Total params: 1,250,177\n",
      "Trainable params: 1,250,177\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "modeldir = 'Zilany2014/'\n",
    "# Define the ANF model hyperparameters\n",
    "context_left = 7936 # longer left-sided context for the ANF models\n",
    "context_right = 256\n",
    "Nenc = 14 # number of layers in the encoder - check for the input size\n",
    "# Load the 1-channel ANF models\n",
    "N_cf = 1\n",
    "anfh = load_connear_model(modeldir,json_name=\"/anfh_1cf.json\",weights_name=\"/anfh.h5\",name=\"anfh_model\")\n",
    "anfh.summary()\n",
    "anfm = load_connear_model(modeldir,json_name=\"/anfm_1cf.json\",weights_name=\"/anfm.h5\",name=\"anfm_model\")\n",
    "anfm.summary()\n",
    "anfl = load_connear_model(modeldir,json_name=\"/anfl_1cf.json\",weights_name=\"/anfl.h5\",name=\"anfl_model\")\n",
    "anfl.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 387ms/step\n",
      "1/1 [==============================] - 0s 407ms/step\n",
      "1/1 [==============================] - 0s 436ms/step\n"
     ]
    }
   ],
   "source": [
    "# Define the pure tone stimulus\n",
    "f_tone = 1e3 # frequency of the pure tone\n",
    "L = [70.] # stimulus level\n",
    "m = 1 # modulation percentage \n",
    "f_m = 100 # modulation frequency - leave empty for pure-tone stimulus (no modulation applied)\n",
    "#f_m = [] # uncomment for pure-tone stimulus\n",
    "\n",
    "stim_dur = 400e-3 # duration of the stimulus - 409.6 ms correspond to 8192 samples (fs = 20 kHz)\n",
    "silence_left = 396.8e-3 + 5e-3 # silence before the onset of the stimulus - 396.8 ms correspond to 7936 samples (left context)\n",
    "silence_right = 12.8e-3 # silence after the stimulus - 256 samples (right context)\n",
    "win_dur = 7.8e-3 # 5ms long ramp window for gradual onset\n",
    "\n",
    "# match the tone frequency to the corresponding CF\n",
    "fno, _ = min(enumerate(CF_connear), key=lambda x: abs( x [1]- f_tone))\n",
    "f_tone = CF_connear[int(fno)]\n",
    "\n",
    "# make stimulus\n",
    "t = np.arange(0., stim_dur, 1./fs_tl)\n",
    "if f_m: # if f_m is defined make a SAM tone\n",
    "    stim_sin = (1 + m * np.cos(2 * np.pi * f_m * t + np.pi)) * np.sin(2 * np.pi * f_tone * t) # generate the SAM tone\n",
    "else:\n",
    "    stim_sin = np.sin(2 * np.pi * f_tone * t) # generate the pure tone\n",
    "# apply ramp\n",
    "winlength = int(2*win_dur * fs_tl)\n",
    "win = sp_sig.windows.bartlett(winlength) # double-sided ramp window\n",
    "stim_sin[:int(winlength/2)] = stim_sin[:int(winlength/2)] * win[:int(winlength/2)]\n",
    "stim_sin[-int(winlength/2):] = stim_sin[-int(winlength/2):] * win[int(winlength/2):]\n",
    "total_length = int(silence_left * fs_tl) + len(stim_sin) + int(silence_right * fs_tl)\n",
    "stim = np.zeros((len(L), total_length))\n",
    "stimrange = range(int(silence_left * fs_tl), int(silence_left * fs_tl) + len(stim_sin))\n",
    "for i in range(len(L)):\n",
    "    stim[i, stimrange] = p0 * 10**(L[i]/20) * stim_sin / rms(stim_sin) # calibrate \n",
    "\n",
    "############ Zilanyetal ############\n",
    "parent_dir = os.getcwd() # get current directory \n",
    "os.chdir(\"./\" + modeldir + \"reference/\") # navigate to the model's folder\n",
    "eng = matlab.engine.start_matlab() # start the matlab engine\n",
    "\n",
    "# pass the variables to the matlab workspace\n",
    "eng.workspace['nrep'] = 1.\n",
    "eng.workspace['Fs'] = fs_tl\n",
    "eng.workspace['noiseType'] = 0.0\n",
    "eng.workspace['implnt'] = 0.0\n",
    "eng.workspace['cohc'] = 1.0\n",
    "eng.workspace['cihc'] = 1.0\n",
    "eng.workspace['species'] = 2.0 # human\n",
    "\n",
    "sponts = [1.,2.,3.] # simulate LSR, MSR and HSR ANF fibers\n",
    "T = (total_length+1) / fs_tl # make simulation duration 1 sample longer to avoid errors\n",
    "eng.workspace['T'] = T\n",
    "\n",
    "window_length = int(total_length / ds_factor) # window length of the downsampled outputs\n",
    "ihc_target = np.zeros((len(L),window_length,N_cf))\n",
    "anfh_target = np.zeros((len(L), window_length, N_cf))\n",
    "anfm_target = np.zeros((len(L), window_length, N_cf))\n",
    "anfl_target = np.zeros((len(L), window_length, N_cf))\n",
    "for i in range(len(L)):\n",
    "    stim_in = matlab.double(stim[i,:].tolist())\n",
    "    eng.workspace['stim'] = stim_in\n",
    "    eng.workspace['f'] = float(f_tone)\n",
    "    ihc_out = eng.eval('model_IHC(stim,f,nrep,1/Fs,T,cohc,cihc,species);', nargout = 1) # simulate the IHC output\n",
    "    eng.workspace['ihc'] = ihc_out\n",
    "    ihc_out = np.array(ihc_out[0])\n",
    "    ihc_target[i,:,0] = sp_sig.resample_poly(ihc_out, fs_connear, fs_tl)[:window_length] # downsample\n",
    "\n",
    "    for sponti, spont in enumerate(sponts):\n",
    "        eng.workspace['spont'] = float(spont)\n",
    "        anf_out = eng.eval('model_Synapse(ihc,f,nrep,1/Fs,spont,noiseType,implnt);', nargout = 1) # simulate the mean ANF firing rate\n",
    "        anf_out = np.array(anf_out[0])[:total_length]\n",
    "\n",
    "        if sponti == 0:\n",
    "            anfl_target[i,:,0] = sp_sig.resample_poly(anf_out, fs_connear, fs_tl)\n",
    "        elif sponti == 1:\n",
    "            anfm_target[i,:,0] = sp_sig.resample_poly(anf_out, fs_connear, fs_tl)\n",
    "        else:\n",
    "            anfh_target[i,:,0] = sp_sig.resample_poly(anf_out, fs_connear, fs_tl)\n",
    "                    \n",
    "eng.quit()\n",
    "os.chdir(parent_dir) # go back to the original folder\n",
    "    \n",
    "# remove context from the outputs (and the last dimension)\n",
    "anfh_target = anfh_target[:, context_left:-context_right, 0]\n",
    "anfm_target = anfm_target[:, context_left:-context_right, 0]\n",
    "anfl_target = anfl_target[:, context_left:-context_right, 0]\n",
    "# apply proper scaling for feeding to the CNN model\n",
    "ihc_target = ihc_target * ihc_scaling # scaling for the ihc input\n",
    "\n",
    "############ CNN model #################\n",
    "# check the time dimension size\n",
    "if ihc_target.shape[1] % 2**Nenc: # input size needs to be a multiple of 16384\n",
    "    Npad = int(np.ceil(ihc_target.shape[1]/(2**Nenc)))*(2**Nenc)-ihc_target.shape[1]\n",
    "    ihc_target = np.pad(ihc_target,((0,0),(0,Npad),(0,0))) # zero-pad\n",
    "# simulate\n",
    "anfh_connear = anfh.predict(ihc_target, verbose=1)\n",
    "anfm_connear = anfm.predict(ihc_target, verbose=1)\n",
    "anfl_connear = anfl.predict(ihc_target, verbose=1)\n",
    "# remove last dimension \n",
    "anfh_connear = anfh_connear[:, :, 0]\n",
    "anfm_connear = anfm_connear[:, :, 0]\n",
    "anfl_connear = anfl_connear[:, :, 0]\n",
    "# scale back to the original ANF values\n",
    "anfh_connear = anfh_connear / an_scaling\n",
    "anfm_connear = anfm_connear / an_scaling\n",
    "anfl_connear = anfl_connear / an_scaling\n",
    "\n",
    "# crop the connear time dimensions to match the size of the reference responses\n",
    "anfh_connear = anfh_connear[:,:anfh_target.shape[1]]\n",
    "anfm_connear = anfm_connear[:,:anfm_target.shape[1]]\n",
    "anfl_connear = anfl_connear[:,:anfl_target.shape[1]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "############ Plots ###################\n",
    "t_ds = np.arange(0., anfh_target.shape[1]/fs_connear, 1./fs_connear) # time vector of the (downsampled) responses\n",
    "\n",
    "# Plot the firing rate patterns for the reference model\n",
    "plt.plot(1e3*t_ds,anfh_target.T,'r')\n",
    "plt.plot(1e3*t_ds,anfm_target.T,'b')\n",
    "plt.plot(1e3*t_ds,anfl_target.T,'c')\n",
    "plt.xlim(0,80.), plt.grid(which='both'), \n",
    "plt.xlabel('Time (ms)')\n",
    "plt.ylabel('Firing rate (spikes/s)')\n",
    "plt.title('Zilany2014 ANF model')\n",
    "plt.legend(['HSR','MSR','LSR'], frameon=False)\n",
    "plt.show()\n",
    "\n",
    "# Plot the firing rate patterns for the CoNNear model\n",
    "plt.plot(1e3*t_ds,anfh_connear.T,'r')\n",
    "plt.plot(1e3*t_ds,anfm_connear.T,'b')\n",
    "plt.plot(1e3*t_ds,anfl_connear.T,'c')\n",
    "plt.xlim(0,80.), plt.grid(which='both'), \n",
    "plt.xlabel('Time (ms)')\n",
    "plt.ylabel('Firing rate (spikes/s)')\n",
    "plt.title('Zilany2014 ANF model - CNN')\n",
    "plt.legend(['HSR','MSR','LSR'], frameon=False)\n",
    "plt.show()\n",
    "\n",
    "del anfh, anfm, anfl # remove the connear models variables to free-up some memory"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ANF rate-level curves\n",
    "Compare the simulated ANF rate-level curves for the three fiber models.\n",
    "**You can change the `f_tone` variable to get the curves of 1 kHz or 4 kHz.** \n",
    "\n",
    "To simulate the ANF rate-level curves reported in the paper, the mean firing rate outputs need to be fed to the Zilany2014 ANF spike generator to generate the PSTH outputs. This can be done by setting the `simulate_psth` variable to 1, but requires many stimulus repetitions and a lot of time to compute (or a powerful computer). Alternatively, the spike generator can be omited to speed up the execution, by setting the `simulate_psth` variable to 1, and plotting the same results for the mean firing rate outputs instead."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "simulate_psth = 1 # set to 0 to use the mean firing rates or to 1 to use the PSTH outputs (rather slow)\n",
    "\n",
    "modeldir = 'Zilany2014/'\n",
    "# Define the ANF model hyperparameters\n",
    "context_left = 7936\n",
    "context_right = 256\n",
    "Nenc = 14 # number of layers in the encoder - check for the input size\n",
    "# Load the 1-channel ANF models\n",
    "N_cf = 1\n",
    "if simulate_psth:\n",
    "    crop = 0 # get uncropped outputs to feed to the spike generator model\n",
    "else:\n",
    "    crop = 1\n",
    "anfh = load_connear_model(modeldir,json_name=\"/anfh_1cf.json\",weights_name=\"/anfh.h5\",name=\"anfh_model\",crop=crop)\n",
    "#anfh.summary()\n",
    "anfm = load_connear_model(modeldir,json_name=\"/anfm_1cf.json\",weights_name=\"/anfm.h5\",name=\"anfm_model\",crop=crop)\n",
    "#anfm.summary()\n",
    "anfl = load_connear_model(modeldir,json_name=\"/anfl_1cf.json\",weights_name=\"/anfl.h5\",name=\"anfl_model\",crop=crop)\n",
    "#anfl.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\Me\\.conda\\envs\\evaluation_env\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:422: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n",
      "\n",
      "11/11 [==============================] - 1s 90ms/step\n",
      "11/11 [==============================] - 1s 85ms/step\n",
      "11/11 [==============================] - 1s 84ms/step\n"
     ]
    }
   ],
   "source": [
    "# Define the pure tone stimulus\n",
    "f_tone = 4e3 # frequency of the pure tone\n",
    "L = np.arange(0.,101.,10.) # levels from 0 to 100 dB SPL\n",
    "\n",
    "stim_dur = 50e-3 # duration of the stimulus - 409.6 ms correspond to 8192 samples (fs = 20 kHz)\n",
    "silence_left = 396.8e-3 + 5e-3 # silence before the onset of the stimulus - 396.8 ms correspond to 7936 samples (left context)\n",
    "silence_right = 12.8e-3 # silence after the stimulus - 256 samples (right context)\n",
    "win_dur = 2.5e-3 # 2.5ms long ramp window for gradual onset\n",
    "\n",
    "# Zilany2014 model parameters\n",
    "nrep_spike = 100 # number of repetitions for the spike generator part of the ANF model - decrease for a faster but noisier result\n",
    "psthbinwidth = 1/fs_connear\n",
    "psthbins = int(psthbinwidth*fs_connear) # number of psth bins per psth bin\n",
    "\n",
    "# match the tone frequency to the corresponding CF\n",
    "fno, _ = min(enumerate(CF_connear), key=lambda x: abs( x [1]- f_tone))\n",
    "f_tone = CF_connear[int(fno)]\n",
    "\n",
    "# indicate the region of the response for computing the mean - 10-40 ms after the stimulus onset\n",
    "index_start = int((silence_left + 10e-3) * fs_connear) - context_left # omit silence + 10 ms (15ms = 300 samples)\n",
    "index_end = index_start + int(30e-3 * fs_connear) # keep 30 ms of response after (30ms = 600 samples)\n",
    "stim_reg = np.arange(index_start,index_end,1) # stimulus region\n",
    "    \n",
    "# make stimulus\n",
    "t = np.arange(0., stim_dur, 1./fs_tl)\n",
    "stim_sin = np.sin(2 * np.pi * f_tone * t) # generate the pure tone\n",
    "# apply ramp\n",
    "winlength = int(2*win_dur * fs_tl)\n",
    "win = sp_sig.windows.bartlett(winlength) # double-sided ramp window\n",
    "stim_sin[:int(winlength/2)] = stim_sin[:int(winlength/2)] * win[:int(winlength/2)]\n",
    "stim_sin[-int(winlength/2):] = stim_sin[-int(winlength/2):] * win[int(winlength/2):]\n",
    "total_length = int(silence_left * fs_tl) + len(stim_sin) + int(silence_right * fs_tl)\n",
    "stim = np.zeros((len(L), total_length))\n",
    "stimrange = range(int(silence_left * fs_tl), int(silence_left * fs_tl) + len(stim_sin))\n",
    "for i in range(len(L)):\n",
    "    stim[i, stimrange] = p0 * 10**(L[i]/20) * stim_sin / rms(stim_sin) # calibrate \n",
    "\n",
    "############ Zilanyetal ############\n",
    "parent_dir = os.getcwd() # get current directory \n",
    "os.chdir(\"./\" + modeldir + \"reference/\") # navigate to the model's folder\n",
    "eng = matlab.engine.start_matlab() # start the matlab engine\n",
    "\n",
    "# pass the variables to the matlab workspace\n",
    "eng.workspace['nrep'] = 1.\n",
    "eng.workspace['nrep_spike'] = float(nrep_spike)\n",
    "eng.workspace['Fs'] = fs_tl\n",
    "eng.workspace['Fs_psth'] = fs_connear\n",
    "eng.workspace['noiseType'] = 0.0\n",
    "eng.workspace['implnt'] = 0.0\n",
    "eng.workspace['cohc'] = 1.0\n",
    "eng.workspace['cihc'] = 1.0\n",
    "eng.workspace['species'] = 2.0 # human\n",
    "\n",
    "sponts = [1.,2.,3.] # simulate LSR, MSR and HSR ANF fibers\n",
    "T = (total_length+1) / fs_tl # make simulation duration 1 sample longer to avoid errors\n",
    "eng.workspace['T'] = T\n",
    "\n",
    "window_length = int(total_length / ds_factor) # window length of the downsampled outputs\n",
    "ihc_target = np.zeros((len(L),window_length,N_cf))\n",
    "anfh_target = np.zeros((len(L), window_length, N_cf))\n",
    "anfm_target = np.zeros((len(L), window_length, N_cf))\n",
    "anfl_target = np.zeros((len(L), window_length, N_cf))\n",
    "for i in range(len(L)):\n",
    "    stim_in = matlab.double(stim[i,:].tolist())\n",
    "    eng.workspace['stim'] = stim_in\n",
    "    eng.workspace['f'] = float(f_tone)\n",
    "    ihc_out = eng.eval('model_IHC(stim,f,nrep,1/Fs,T,cohc,cihc,species);', nargout = 1) # simulate the IHC output\n",
    "    eng.workspace['ihc'] = ihc_out\n",
    "    ihc_out = np.array(ihc_out[0])\n",
    "    ihc_target[i,:,0] = sp_sig.resample_poly(ihc_out, fs_connear, fs_tl)[:window_length] # downsample\n",
    "\n",
    "    for sponti, spont in enumerate(sponts):\n",
    "        eng.workspace['spont'] = float(spont)\n",
    "        anf_out = eng.eval('model_Synapse(ihc,f,nrep,1/Fs,spont,noiseType,implnt);', nargout = 1) # simulate the mean ANF firing rate\n",
    "        anf_out = np.array(anf_out[0])[:total_length]\n",
    "        \n",
    "        if sponti == 0:\n",
    "            anfl_target[i,:,0] = sp_sig.resample_poly(anf_out, fs_connear, fs_tl)\n",
    "        elif sponti == 1:\n",
    "            anfm_target[i,:,0] = sp_sig.resample_poly(anf_out, fs_connear, fs_tl)\n",
    "        else:\n",
    "            anfh_target[i,:,0] = sp_sig.resample_poly(anf_out, fs_connear, fs_tl)\n",
    "        \n",
    "        if simulate_psth: # simulate psth\n",
    "            if sponti == 0:\n",
    "                anf_out = anfl_target[i,:,0]\n",
    "            elif sponti == 1:\n",
    "                anf_out = anfm_target[i,:,0]\n",
    "            else:\n",
    "                anf_out = anfh_target[i,:,0]\n",
    "            anf_out = np.tile(anf_out,nrep_spike)\n",
    "            anf_out = matlab.double(anf_out.tolist())\n",
    "            eng.workspace['rate'] = anf_out\n",
    "            psTH = eng.eval('model_SpikeGen(rate,f,nrep_spike,1/Fs_psth,spont,noiseType,implnt);', nargout = 1)\n",
    "            psTH = np.array(psTH[0])\n",
    "\n",
    "            pr = np.sum(np.reshape(psTH,(psthbins,int(psTH.size/psthbins)),order='F'),axis=0)/nrep_spike # pr of spike in each bin\n",
    "            psTH = pr/psthbinwidth # psth in units of /s\n",
    "\n",
    "            if sponti == 0:\n",
    "                anfl_target[i,:,0] = psTH\n",
    "            elif sponti == 1:\n",
    "                anfm_target[i,:,0] = psTH\n",
    "            else:\n",
    "                anfh_target[i,:,0] = psTH\n",
    "\n",
    "# remove context from the outputs (and last dimension)\n",
    "anfh_target = anfh_target[:, context_left:-context_right, 0]\n",
    "anfm_target = anfm_target[:, context_left:-context_right, 0]\n",
    "anfl_target = anfl_target[:, context_left:-context_right, 0]\n",
    "# apply proper scaling for feeding to the CoNNear\n",
    "ihc_target = ihc_target * ihc_scaling # scaling for the cochlear input\n",
    "\n",
    "# compute the mean firing rate over the stimulus region\n",
    "anfh_target_mean = np.mean(anfh_target[:,stim_reg],axis=1)\n",
    "anfm_target_mean = np.mean(anfm_target[:,stim_reg],axis=1)\n",
    "anfl_target_mean = np.mean(anfl_target[:,stim_reg],axis=1)\n",
    "\n",
    "############ CNN model #################\n",
    "# check the time dimension size\n",
    "if ihc_target.shape[1] % 2**Nenc: # input size needs to be a multiple of 16384\n",
    "    Npad = int(np.ceil(ihc_target.shape[1]/(2**Nenc)))*(2**Nenc)-ihc_target.shape[1]\n",
    "    ihc_target = np.pad(ihc_target,((0,0),(0,Npad),(0,0))) # zero-pad\n",
    "# simulate\n",
    "anfh_connear = anfh.predict(ihc_target, verbose=1)\n",
    "anfm_connear = anfm.predict(ihc_target, verbose=1)\n",
    "anfl_connear = anfl.predict(ihc_target, verbose=1)\n",
    "\n",
    "# scale back to the original ANF values\n",
    "anfh_connear = anfh_connear / an_scaling\n",
    "anfm_connear = anfm_connear / an_scaling\n",
    "anfl_connear = anfl_connear / an_scaling\n",
    "\n",
    "if simulate_psth: # simulate psth\n",
    "    # omit the negative values\n",
    "    anfh_connear = anfh_connear.clip(0)\n",
    "    anfm_connear = anfm_connear.clip(0)\n",
    "    anfl_connear = anfl_connear.clip(0)\n",
    "    \n",
    "    for i in range(len(L)):\n",
    "        for sponti, spont in enumerate(sponts):\n",
    "            eng.workspace['spont'] = float(spont)\n",
    "            if sponti == 0:\n",
    "                anf_out = anfl_connear[i,:,0]\n",
    "            elif sponti == 1:\n",
    "                anf_out = anfm_connear[i,:,0]\n",
    "            else:\n",
    "                anf_out = anfh_connear[i,:,0]\n",
    "            anf_out = np.tile(anf_out,nrep_spike)\n",
    "            anf_out = matlab.double(anf_out.tolist())\n",
    "            eng.workspace['rate'] = anf_out\n",
    "            psTH = eng.eval('model_SpikeGen(rate,f,nrep_spike,1/Fs_psth,spont,noiseType,implnt);', nargout = 1)\n",
    "            psTH = np.array(psTH[0])\n",
    "\n",
    "            pr = np.sum(np.reshape(psTH,(psthbins,int(psTH.size/psthbins)),order='F'),axis=0)/nrep_spike # pr of spike in each bin\n",
    "            psTH = pr/psthbinwidth # psth in units of /s\n",
    "\n",
    "            if sponti == 0:\n",
    "                anfl_connear[i,:,0] = psTH\n",
    "            elif sponti == 1:\n",
    "                anfm_connear[i,:,0] = psTH\n",
    "            else:\n",
    "                anfh_connear[i,:,0] = psTH\n",
    "                \n",
    "    # remove context from the outputs\n",
    "    anfh_connear = anfh_connear[:, context_left:-context_right, :]\n",
    "    anfm_connear = anfm_connear[:, context_left:-context_right, :]\n",
    "    anfl_connear = anfl_connear[:, context_left:-context_right, :]\n",
    "\n",
    "# remove last dimension \n",
    "anfh_connear = anfh_connear[:, :, 0]\n",
    "anfm_connear = anfm_connear[:, :, 0]\n",
    "anfl_connear = anfl_connear[:, :, 0]\n",
    "\n",
    "# compute the mean firing rate over the stimulus region\n",
    "anfh_connear_mean = np.mean(anfh_connear[:,stim_reg],axis=1)\n",
    "anfm_connear_mean = np.mean(anfm_connear[:,stim_reg],axis=1)\n",
    "anfl_connear_mean = np.mean(anfl_connear[:,stim_reg],axis=1)\n",
    "\n",
    "eng.quit()\n",
    "os.chdir(parent_dir) # go back to the original folder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "############ Plots ###################\n",
    "# Plot the rate-level curves for the reference model\n",
    "plt.plot(anfh_target_mean.T,'ro-')\n",
    "plt.plot(anfm_target_mean.T,'bo-')\n",
    "plt.plot(anfl_target_mean.T,'co-')\n",
    "plt.xlim(0,10), plt.grid(which='both'), \n",
    "plt.xticks(ticks=L/10 , labels=L.astype(int))\n",
    "plt.xlabel('Stimulus level (dB SPL)')\n",
    "plt.ylabel('Firing rate (spikes/s)')\n",
    "plt.title('Zilany2014 ANF model')\n",
    "plt.legend(['HSR','MSR','LSR'], frameon=False)\n",
    "plt.show()\n",
    "\n",
    "# Plot the rate-level curves for the CoNNear model\n",
    "plt.plot(anfh_connear_mean.T,'ro-')\n",
    "plt.plot(anfm_connear_mean.T,'bo-')\n",
    "plt.plot(anfl_connear_mean.T,'co-')\n",
    "plt.xlim(0,10), plt.grid(which='both'), \n",
    "plt.xticks(ticks=L/10 , labels=L.astype(int))\n",
    "plt.xlabel('Stimulus level (dB SPL)')\n",
    "plt.ylabel('Firing rate (spikes/s)')\n",
    "plt.title('Zilany2014 ANF model - CNN')\n",
    "plt.legend(['HSR','MSR','LSR'], frameon=False)\n",
    "plt.show()\n",
    "\n",
    "del anfh, anfm, anfl # remove the connear models variables to free-up some memory"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ANF synchrony-level functions\n",
    "Compare the simulated ANF synchrony-level functions for the three fiber models.\n",
    "**You can change the `f_tone` variable to get the curves of 1 kHz or 4 kHz.**\n",
    "\n",
    "To simulate the ANF synchrony-level curves reported in the paper, the mean firing rate outputs need to be fed to the Zilany2014 ANF spike generator to generate the PSTH outputs. This can be done by setting the `simulate_psth` variable to 1, but requires many stimulus repetitions and a lot of time to compute (or a powerful computer). Alternatively, the spike generator can be omited to speed up the execution, by setting the `simulate_psth` variable to 1, and plotting the same results for the mean firing rate outputs instead."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "simulate_psth = 1 # set to 0 to use the mean firing rates or to 1 to use the PSTH outputs (rather slow)\n",
    "\n",
    "modeldir = 'Zilany2014/'\n",
    "# Define the ANF model hyperparameters\n",
    "context_left = 7936\n",
    "context_right = 256\n",
    "Nenc = 14 # number of layers in the encoder - check for the input size\n",
    "# Load the 1-channel ANF models\n",
    "N_cf = 1\n",
    "if simulate_psth:\n",
    "    crop = 0 # get uncropped outputs to feed to the spike generator model\n",
    "else:\n",
    "    crop = 1\n",
    "anfh = load_connear_model(modeldir,json_name=\"/anfh_1cf.json\",weights_name=\"/anfh.h5\",name=\"anfh_model\",crop=crop)\n",
    "#anfh.summary()\n",
    "anfm = load_connear_model(modeldir,json_name=\"/anfm_1cf.json\",weights_name=\"/anfm.h5\",name=\"anfm_model\",crop=crop)\n",
    "#anfm.summary()\n",
    "anfl = load_connear_model(modeldir,json_name=\"/anfl_1cf.json\",weights_name=\"/anfl.h5\",name=\"anfl_model\",crop=crop)\n",
    "#anfl.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "11/11 [==============================] - 1s 97ms/step\n",
      "11/11 [==============================] - 1s 98ms/step\n",
      "11/11 [==============================] - 1s 98ms/step\n"
     ]
    }
   ],
   "source": [
    "# Define the pure tone stimulus\n",
    "f_tone = 4e3 # frequency of the pure tone\n",
    "L = np.arange(0.,101.,10.) # levels from 0 to 100 dB SPL with a step of 10 - in the paper the step is 5\n",
    "f_m = 100 # modulation frequency\n",
    "m = 1 # modulation percentage \n",
    "\n",
    "stim_dur = 400e-3 # duration of the stimulus - 409.6 ms correspond to 8192 samples (fs = 20 kHz)\n",
    "silence_left = 396.8e-3 + 5e-3 # silence before the onset of the stimulus - 396.8 ms correspond to 7936 samples (left context)\n",
    "silence_right = 12.8e-3 # silence after the stimulus - 256 samples (right context)\n",
    "win_dur = 7.8e-3 # 7.8ms long ramp window for gradual onset\n",
    "\n",
    "# Zilany2014 model parameters\n",
    "nrep_spike = 100 # number of repetitions for the spike generator part of the ANF model - decrease for a faster but noisier result\n",
    "psthbinwidth = 1/fs_connear\n",
    "psthbins = int(psthbinwidth*fs_connear) # number of psth bins per psth bin\n",
    "\n",
    "# match the tone frequency to the corresponding CF\n",
    "fno, _ = min(enumerate(CF_connear), key=lambda x: abs( x [1]- f_tone))\n",
    "f_tone = CF_connear[int(fno)]\n",
    "\n",
    "# indicate the region of interest for computing the synchrony\n",
    "index_start = int(silence_left * fs_connear) - context_left # omit silence (5ms = 100 samples)\n",
    "index_end = index_start + int(stim_dur * fs_connear) # keep for the stimulus duration (400ms = 8000 samples)\n",
    "stim_reg = np.arange(index_start,index_end,1) # stimulus region\n",
    "\n",
    "# make stimulus\n",
    "t = np.arange(0., stim_dur, 1./fs_tl)\n",
    "stim_sin = (1 + m * np.cos(2 * np.pi * f_m * t + np.pi)) * np.sin(2 * np.pi * f_tone * t) # generate the SAM tone\n",
    "# apply ramp\n",
    "winlength = int(2*win_dur * fs_tl)\n",
    "win = sp_sig.windows.bartlett(winlength) # double-sided ramp window\n",
    "stim_sin[:int(winlength/2)] = stim_sin[:int(winlength/2)] * win[:int(winlength/2)]\n",
    "stim_sin[-int(winlength/2):] = stim_sin[-int(winlength/2):] * win[int(winlength/2):]\n",
    "total_length = int(silence_left * fs_tl) + len(stim_sin) + int(silence_right * fs_tl)\n",
    "stim = np.zeros((len(L), total_length))\n",
    "stimrange = range(int(silence_left * fs_tl), int(silence_left * fs_tl) + len(stim_sin))\n",
    "for i in range(len(L)):\n",
    "    stim[i, stimrange] = p0 * 10**(L[i]/20) * stim_sin / rms(stim_sin) # calibrate \n",
    "\n",
    "############ Zilanyetal ############\n",
    "parent_dir = os.getcwd() # get current directory \n",
    "os.chdir(\"./\" + modeldir + \"reference/\") # navigate to the model's folder\n",
    "eng = matlab.engine.start_matlab() # start the matlab engine\n",
    "\n",
    "# pass the variables to the matlab workspace\n",
    "eng.workspace['nrep'] = 1.\n",
    "eng.workspace['nrep_spike'] = float(nrep_spike)\n",
    "eng.workspace['Fs'] = fs_tl\n",
    "eng.workspace['Fs_psth'] = fs_connear\n",
    "eng.workspace['noiseType'] = 0.0\n",
    "eng.workspace['implnt'] = 0.0\n",
    "eng.workspace['cohc'] = 1.0\n",
    "eng.workspace['cihc'] = 1.0\n",
    "eng.workspace['species'] = 2.0 # human\n",
    "\n",
    "sponts = [1.,2.,3.] # simulate LSR, MSR and HSR ANF fibers\n",
    "T = (total_length+1) / fs_tl # make simulation duration 1 sample longer to avoid errors\n",
    "eng.workspace['T'] = T\n",
    "\n",
    "window_length = int(total_length / ds_factor) # window length of the downsampled outputs\n",
    "ihc_target = np.zeros((len(L),window_length,N_cf))\n",
    "anfh_target = np.zeros((len(L), window_length, N_cf))\n",
    "anfm_target = np.zeros((len(L), window_length, N_cf))\n",
    "anfl_target = np.zeros((len(L), window_length, N_cf))\n",
    "for i in range(len(L)):\n",
    "    stim_in = matlab.double(stim[i,:].tolist())\n",
    "    eng.workspace['stim'] = stim_in\n",
    "    eng.workspace['f'] = float(f_tone)\n",
    "    ihc_out = eng.eval('model_IHC(stim,f,nrep,1/Fs,T,cohc,cihc,species);', nargout = 1) # simulate the IHC output\n",
    "    eng.workspace['ihc'] = ihc_out\n",
    "    ihc_out = np.array(ihc_out[0])\n",
    "    ihc_target[i,:,0] = sp_sig.resample_poly(ihc_out, fs_connear, fs_tl)[:window_length] # downsample\n",
    "\n",
    "    for sponti, spont in enumerate(sponts):\n",
    "        eng.workspace['spont'] = float(spont)\n",
    "        anf_out = eng.eval('model_Synapse(ihc,f,nrep,1/Fs,spont,noiseType,implnt);', nargout = 1) # simulate the mean ANF firing rate\n",
    "        anf_out = np.array(anf_out[0])[:total_length]\n",
    "        \n",
    "        if sponti == 0:\n",
    "            anfl_target[i,:,0] = sp_sig.resample_poly(anf_out, fs_connear, fs_tl)\n",
    "        elif sponti == 1:\n",
    "            anfm_target[i,:,0] = sp_sig.resample_poly(anf_out, fs_connear, fs_tl)\n",
    "        else:\n",
    "            anfh_target[i,:,0] = sp_sig.resample_poly(anf_out, fs_connear, fs_tl)\n",
    "        \n",
    "        if simulate_psth: # simulate psth\n",
    "            if sponti == 0:\n",
    "                anf_out = anfl_target[i,:,0]\n",
    "            elif sponti == 1:\n",
    "                anf_out = anfm_target[i,:,0]\n",
    "            else:\n",
    "                anf_out = anfh_target[i,:,0]\n",
    "            anf_out = np.tile(anf_out,nrep_spike)\n",
    "            anf_out = matlab.double(anf_out.tolist())\n",
    "            eng.workspace['rate'] = anf_out\n",
    "            psTH = eng.eval('model_SpikeGen(rate,f,nrep_spike,1/Fs_psth,spont,noiseType,implnt);', nargout = 1)\n",
    "            psTH = np.array(psTH[0])\n",
    "\n",
    "            pr = np.sum(np.reshape(psTH,(psthbins,int(psTH.size/psthbins)),order='F'),axis=0)/nrep_spike # pr of spike in each bin\n",
    "            psTH = pr/psthbinwidth # psth in units of /s\n",
    "\n",
    "            if sponti == 0:\n",
    "                anfl_target[i,:,0] = psTH\n",
    "            elif sponti == 1:\n",
    "                anfm_target[i,:,0] = psTH\n",
    "            else:\n",
    "                anfh_target[i,:,0] = psTH\n",
    "\n",
    "# remove context from the outputs (and last dimension)\n",
    "anfh_target = anfh_target[:, context_left:-context_right, 0]\n",
    "anfm_target = anfm_target[:, context_left:-context_right, 0]\n",
    "anfl_target = anfl_target[:, context_left:-context_right, 0]\n",
    "# apply proper scaling for feeding to the CoNNear\n",
    "ihc_target = ihc_target * ihc_scaling # scaling for the ihc input\n",
    "\n",
    "# compute the fft of the response\n",
    "N = stim_reg.size\n",
    "anfh_target_fft = (1/N)*(np.abs(np.fft.fft(anfh_target[:,stim_reg])))\n",
    "anfm_target_fft = (1/N)*(np.abs(np.fft.fft(anfm_target[:,stim_reg])))\n",
    "anfl_target_fft = (1/N)*(np.abs(np.fft.fft(anfl_target[:,stim_reg])))\n",
    "# divide the modulation frequency component of the Fourier transform by the DC component (1st bin) to get the vector strengh\n",
    "anfh_target_vs = anfh_target_fft[:,int(np.ceil(f_m/fs_connear*N))] / anfh_target_fft[:,0]\n",
    "anfm_target_vs = anfm_target_fft[:,int(np.ceil(f_m/fs_connear*N))] / anfm_target_fft[:,0]\n",
    "anfl_target_vs = anfl_target_fft[:,int(np.ceil(f_m/fs_connear*N))] / anfl_target_fft[:,0]\n",
    "\n",
    "############ CNN model #################\n",
    "# check the time dimension size\n",
    "if ihc_target.shape[1] % 2**Nenc: # input size needs to be a multiple of 16384\n",
    "    Npad = int(np.ceil(ihc_target.shape[1]/(2**Nenc)))*(2**Nenc)-ihc_target.shape[1]\n",
    "    ihc_target = np.pad(ihc_target,((0,0),(0,Npad),(0,0))) # zero-pad\n",
    "# simulate\n",
    "anfh_connear = anfh.predict(ihc_target, verbose=1)\n",
    "anfm_connear = anfm.predict(ihc_target, verbose=1)\n",
    "anfl_connear = anfl.predict(ihc_target, verbose=1)\n",
    "\n",
    "# scale back to the original ANF values\n",
    "anfh_connear = anfh_connear / an_scaling\n",
    "anfm_connear = anfm_connear / an_scaling\n",
    "anfl_connear = anfl_connear / an_scaling\n",
    "\n",
    "if simulate_psth: # simulate psth\n",
    "    # omit the negative values\n",
    "    anfh_connear = anfh_connear.clip(0)\n",
    "    anfm_connear = anfm_connear.clip(0)\n",
    "    anfl_connear = anfl_connear.clip(0)\n",
    "    \n",
    "    for i in range(len(L)):\n",
    "        for sponti, spont in enumerate(sponts):\n",
    "            eng.workspace['spont'] = float(spont)\n",
    "            if sponti == 0:\n",
    "                anf_out = anfl_connear[i,:,0]\n",
    "            elif sponti == 1:\n",
    "                anf_out = anfm_connear[i,:,0]\n",
    "            else:\n",
    "                anf_out = anfh_connear[i,:,0]\n",
    "            anf_out = np.tile(anf_out,nrep_spike)\n",
    "            anf_out = matlab.double(anf_out.tolist())\n",
    "            eng.workspace['rate'] = anf_out\n",
    "            psTH = eng.eval('model_SpikeGen(rate,f,nrep_spike,1/Fs,spont,noiseType,implnt);', nargout = 1)\n",
    "            psTH = np.array(psTH[0])\n",
    "\n",
    "            pr = np.sum(np.reshape(psTH,(psthbins,int(psTH.size/psthbins)),order='F'),axis=0)/nrep_spike # pr of spike in each bin\n",
    "            psTH = pr/psthbinwidth # psth in units of /s\n",
    "\n",
    "            if sponti == 0:\n",
    "                anfl_connear[i,:,0] = psTH\n",
    "            elif sponti == 1:\n",
    "                anfm_connear[i,:,0] = psTH\n",
    "            else:\n",
    "                anfh_connear[i,:,0] = psTH\n",
    "                \n",
    "    # remove context from the outputs\n",
    "    anfh_connear = anfh_connear[:, context_left:-context_right, :]\n",
    "    anfm_connear = anfm_connear[:, context_left:-context_right, :]\n",
    "    anfl_connear = anfl_connear[:, context_left:-context_right, :]\n",
    "\n",
    "# remove last dimension \n",
    "anfh_connear = anfh_connear[:, :, 0]\n",
    "anfm_connear = anfm_connear[:, :, 0]\n",
    "anfl_connear = anfl_connear[:, :, 0]\n",
    "\n",
    "# compute the fft of the response\n",
    "N = stim_reg.size\n",
    "anfh_connear_fft = (1/N)*(np.abs(np.fft.fft(anfh_connear[:,stim_reg])))\n",
    "anfm_connear_fft = (1/N)*(np.abs(np.fft.fft(anfm_connear[:,stim_reg])))\n",
    "anfl_connear_fft = (1/N)*(np.abs(np.fft.fft(anfl_connear[:,stim_reg])))\n",
    "# divide the modulation frequency component of the Fourier transform by the DC component (1st bin) to get the vector strengh\n",
    "anfh_connear_vs = anfh_connear_fft[:,int(np.ceil(f_m/fs_connear*N))] / anfh_connear_fft[:,0]\n",
    "anfm_connear_vs = anfm_connear_fft[:,int(np.ceil(f_m/fs_connear*N))] / anfm_connear_fft[:,0]\n",
    "anfl_connear_vs = anfl_connear_fft[:,int(np.ceil(f_m/fs_connear*N))] / anfl_connear_fft[:,0]\n",
    "\n",
    "eng.quit()\n",
    "os.chdir(parent_dir) # go back to the original folder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "############ Plots ###################\n",
    "# Plot the synchrony-level functions for the reference model\n",
    "plt.plot(anfh_target_vs.T,'ro-')\n",
    "plt.plot(anfm_target_vs.T,'bo-')\n",
    "plt.plot(anfl_target_vs.T,'co-')\n",
    "plt.xlim(0,10), plt.grid(which='both'), \n",
    "plt.ylim(0,1)\n",
    "plt.xticks(ticks=L/10 , labels=L.astype(int))\n",
    "plt.xlabel('Stimulus level (dB SPL)')\n",
    "plt.ylabel('Synchrony to fm')\n",
    "plt.title('Zilany2014 ANF model')\n",
    "plt.legend(['HSR','MSR','LSR'], frameon=False)\n",
    "plt.show()\n",
    "\n",
    "# Plot the synchrony-level functions for the CoNNear model\n",
    "plt.plot(anfh_connear_vs.T,'ro-')\n",
    "plt.plot(anfm_connear_vs.T,'bo-')\n",
    "plt.plot(anfl_connear_vs.T,'co-')\n",
    "plt.xlim(0,10), plt.grid(which='both'), \n",
    "plt.ylim(0,1)\n",
    "plt.xticks(ticks=L/10 , labels=L.astype(int))\n",
    "plt.xlabel('Stimulus level (dB SPL)')\n",
    "plt.ylabel('Synchrony to fm')\n",
    "plt.title('Zilany2014 ANF model - CNN')\n",
    "plt.legend(['HSR','MSR','LSR'], frameon=False)\n",
    "plt.show()\n",
    "\n",
    "del anfh, anfm, anfl # remove the connear models variables to free-up some memory"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "connear_notebook.ipynb",
   "provenance": [],
   "toc_visible": true
  },
  "kernelspec": {
   "display_name": "evaluation_env",
   "language": "python",
   "name": "evaluation_env"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}