run_all.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import json\n",
    "import pandas as pd\n",
    "import xgboost as xgb\n",
    "from sklearn.model_selection import train_test_split\n",
    "from configs.dataloaders import create_data_loaders\n",
    "from runner_static import static_training\n",
    "from runner_dynamic import dynamic_training\n",
    "from models.static_models.static_model_multibranch import StaticModelMultibranch\n",
    "from models.dynamic_models.lstm_model import LSTMDynamicModel\n",
    "from trainers.cross_validation import k_fold_cross_validation_static, k_fold_cross_validation_dynamic\n",
    "from losses.losses import CrossEntropy, CrossEntropyWithLasso, CrossEntropyWithTemporalSmoothness"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Static Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The main goal of this first part is to try different architectures and select the one that performs better for the **static protein localization problem**.\n",
    "\n",
    "We start by initializing all the indispensable variables and loading hyperparameters from `hyperparameters.json`. From now on, all the  __bold_variables__ are to be considered as taken from this JSON file."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define dimensions of the problem\n",
    "seed = 30026\n",
    "num_classes = 15\n",
    "num_timesteps = 5\n",
    "num_features = 34 # both static and dynamic\n",
    "\n",
    "# Decide whever to be verbose with display outputs and set device\n",
    "verbose = True\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define hyperparameters\n",
    "with open('hyperparameters/hyperparameters.json', 'r') as file:\n",
    "    params = json.load(file)\n",
    "\n",
    "static_params = params[\"static_model\"]\n",
    "\n",
    "# Load file\n",
    "data = pd.read_csv(params[\"file_path\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since our models will be combined together for creating more complex models, we split data in train and validation sets once and for all before creating all models: in this way, we avoid the risk of using the same samples for train and for validate at the same time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split dataset in train and test sets keeping balance for the static label localization\n",
    "data_train, data_test = train_test_split(data, test_size=params[\"test_size\"], random_state=seed, stratify = data['static_localization'])\n",
    "\n",
    "# Create dataloaders\n",
    "dataloader_train, dataloader_test = create_data_loaders(data_train, data_test, params[\"batch_size\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## First Model: Multi-Layer-Perceptron"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our first try is a simple MLP with just a hidden layer of size __intermediate_dim__, which takes as input the ESM-embeddings (you can choose between small and big embeddings with __long_embeddings__ boolean variable, depending on computational availabilities).\n",
    "\n",
    "We are using techniques for regularization which will be present in all our architectures: dropout, weight_decay, early stopping, layer normalization, weight initialization, learning rate scheduler.\n",
    "\n",
    "In this first phase, we only try some reasonable hyperparameters choices: because of high compuational costs, we cannot do cross-validation on all the hyperparameters for all the models, but we will do it for our best static model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting training of StaticModelMLP\n",
      "Epoch [1/100],\tTraining Loss: 1.7238, Training accuracy: 42.01,\tTest Loss: 1.4745, Test Accuracy: 49.79,\tLearning rate: 0.01000\n",
      "Epoch [2/100],\tTraining Loss: 1.4223, Training accuracy: 52.03,\tTest Loss: 1.3675, Test Accuracy: 51.20,\tLearning rate: 0.01000\n",
      "Epoch [3/100],\tTraining Loss: 1.3148, Training accuracy: 55.24,\tTest Loss: 1.3329, Test Accuracy: 52.61,\tLearning rate: 0.01000\n",
      "Epoch [4/100],\tTraining Loss: 1.2444, Training accuracy: 56.16,\tTest Loss: 1.3337, Test Accuracy: 53.31,\tLearning rate: 0.01000\n",
      "Epoch [5/100],\tTraining Loss: 1.2013, Training accuracy: 58.48,\tTest Loss: 1.2791, Test Accuracy: 57.69,\tLearning rate: 0.01000\n",
      "Epoch [6/100],\tTraining Loss: 1.1229, Training accuracy: 60.00,\tTest Loss: 1.2971, Test Accuracy: 54.02,\tLearning rate: 0.01000\n",
      "Epoch [7/100],\tTraining Loss: 1.0860, Training accuracy: 62.05,\tTest Loss: 1.3177, Test Accuracy: 56.42,\tLearning rate: 0.01000\n",
      "Epoch [8/100],\tTraining Loss: 1.0424, Training accuracy: 62.86,\tTest Loss: 1.2576, Test Accuracy: 58.11,\tLearning rate: 0.01000\n",
      "Epoch [9/100],\tTraining Loss: 1.0067, Training accuracy: 65.08,\tTest Loss: 1.3371, Test Accuracy: 55.99,\tLearning rate: 0.01000\n",
      "Epoch [10/100],\tTraining Loss: 0.9536, Training accuracy: 65.61,\tTest Loss: 1.2796, Test Accuracy: 56.28,\tLearning rate: 0.01000\n",
      "Epoch [11/100],\tTraining Loss: 0.9148, Training accuracy: 66.63,\tTest Loss: 1.4631, Test Accuracy: 52.33,\tLearning rate: 0.01000\n",
      "Epoch [12/100],\tTraining Loss: 0.8772, Training accuracy: 68.08,\tTest Loss: 1.3280, Test Accuracy: 55.85,\tLearning rate: 0.01000\n",
      "Epoch [13/100],\tTraining Loss: 0.8247, Training accuracy: 69.56,\tTest Loss: 1.3768, Test Accuracy: 57.83,\tLearning rate: 0.01000\n",
      "Epoch [14/100],\tTraining Loss: 0.8158, Training accuracy: 70.09,\tTest Loss: 1.3943, Test Accuracy: 57.97,\tLearning rate: 0.01000\n",
      "Epoch [15/100],\tTraining Loss: 0.7743, Training accuracy: 71.57,\tTest Loss: 1.4184, Test Accuracy: 56.70,\tLearning rate: 0.01000\n",
      "Epoch [16/100],\tTraining Loss: 0.7523, Training accuracy: 71.71,\tTest Loss: 1.3809, Test Accuracy: 57.69,\tLearning rate: 0.01000\n",
      "Epoch [17/100],\tTraining Loss: 0.7156, Training accuracy: 73.16,\tTest Loss: 1.4500, Test Accuracy: 56.84,\tLearning rate: 0.01000\n",
      "Epoch [18/100],\tTraining Loss: 0.6586, Training accuracy: 75.38,\tTest Loss: 1.4960, Test Accuracy: 55.99,\tLearning rate: 0.01000\n",
      "Epoch [19/100],\tTraining Loss: 0.6357, Training accuracy: 77.14,\tTest Loss: 1.5790, Test Accuracy: 56.28,\tLearning rate: 0.01000\n",
      "Epoch [20/100],\tTraining Loss: 0.6102, Training accuracy: 76.68,\tTest Loss: 1.4785, Test Accuracy: 57.12,\tLearning rate: 0.00100\n",
      "Epoch [21/100],\tTraining Loss: 0.4601, Training accuracy: 83.25,\tTest Loss: 1.5190, Test Accuracy: 58.11,\tLearning rate: 0.00100\n",
      "Epoch [22/100],\tTraining Loss: 0.4146, Training accuracy: 85.40,\tTest Loss: 1.5423, Test Accuracy: 58.67,\tLearning rate: 0.00100\n",
      "Epoch [23/100],\tTraining Loss: 0.3773, Training accuracy: 86.03,\tTest Loss: 1.5790, Test Accuracy: 58.67,\tLearning rate: 0.00100\n",
      "Epoch [24/100],\tTraining Loss: 0.3682, Training accuracy: 87.48,\tTest Loss: 1.5739, Test Accuracy: 57.69,\tLearning rate: 0.00100\n",
      "Epoch [25/100],\tTraining Loss: 0.3598, Training accuracy: 87.41,\tTest Loss: 1.5995, Test Accuracy: 58.82,\tLearning rate: 0.00100\n",
      "Epoch [26/100],\tTraining Loss: 0.3376, Training accuracy: 87.58,\tTest Loss: 1.6268, Test Accuracy: 57.26,\tLearning rate: 0.00100\n",
      "Epoch [27/100],\tTraining Loss: 0.3327, Training accuracy: 88.71,\tTest Loss: 1.6495, Test Accuracy: 57.69,\tLearning rate: 0.00100\n",
      "Epoch [28/100],\tTraining Loss: 0.3244, Training accuracy: 89.07,\tTest Loss: 1.6505, Test Accuracy: 57.12,\tLearning rate: 0.00100\n",
      "Early stopping triggered after 28 epochs. Best test loss: 1.2576\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Setting seed\n",
    "torch.manual_seed(30026)\n",
    "if(torch.cuda.is_available()):\n",
    "    torch.cuda.manual_seed(30026)\n",
    "    torch.cuda.manual_seed_all(30026)\n",
    "\n",
    "# Running the first model\n",
    "train_loss, train_accuracy, test_loss, test_accuracy, static_model_dict = static_training(num_classes, static_params, dataloader_train, dataloader_test, verbose, device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Even though we have used some technique like __dropout__, __weight_decay__ and early stopping (set with __patience__ parameter), this model appears to strongly overfit. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Second Model: XGBoost"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Commento di Riki aaaaaa\n",
    "\n",
    "Since XGBoost has a different architecture, we will use an appositely create JSON file `hyperparametersxgb.json`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define hyperparameters\n",
    "with open('hyperparameters/hyperparametersxgb.json', 'r') as file:\n",
    "    xgb_params = json.load(file)\n",
    "\n",
    "xgb_params = xgb_params[\"static_model\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting training of XGBoost\n",
      "[0]\ttrain-mlogloss:2.69687\ttrain-merror:0.40625\ttest-mlogloss:2.70327\ttest-merror:0.71875\n",
      "[25]\ttrain-mlogloss:2.44302\ttrain-merror:0.40625\ttest-mlogloss:2.59052\ttest-merror:0.71875\n",
      "[50]\ttrain-mlogloss:2.23168\ttrain-merror:0.37500\ttest-mlogloss:2.51728\ttest-merror:0.75000\n",
      "[75]\ttrain-mlogloss:2.04491\ttrain-merror:0.28125\ttest-mlogloss:2.48198\ttest-merror:0.71875\n",
      "[100]\ttrain-mlogloss:1.88214\ttrain-merror:0.28125\ttest-mlogloss:2.45011\ttest-merror:0.71875\n",
      "[125]\ttrain-mlogloss:1.74620\ttrain-merror:0.25000\ttest-mlogloss:2.42626\ttest-merror:0.71875\n",
      "[150]\ttrain-mlogloss:1.63589\ttrain-merror:0.25000\ttest-mlogloss:2.40447\ttest-merror:0.71875\n",
      "[175]\ttrain-mlogloss:1.54249\ttrain-merror:0.25000\ttest-mlogloss:2.38184\ttest-merror:0.75000\n",
      "[200]\ttrain-mlogloss:1.46518\ttrain-merror:0.25000\ttest-mlogloss:2.36062\ttest-merror:0.75000\n",
      "[225]\ttrain-mlogloss:1.40105\ttrain-merror:0.25000\ttest-mlogloss:2.33938\ttest-merror:0.75000\n",
      "[250]\ttrain-mlogloss:1.35274\ttrain-merror:0.25000\ttest-mlogloss:2.31879\ttest-merror:0.75000\n",
      "[275]\ttrain-mlogloss:1.32623\ttrain-merror:0.25000\ttest-mlogloss:2.29497\ttest-merror:0.75000\n",
      "[300]\ttrain-mlogloss:1.30568\ttrain-merror:0.25000\ttest-mlogloss:2.27392\ttest-merror:0.75000\n",
      "[325]\ttrain-mlogloss:1.28825\ttrain-merror:0.25000\ttest-mlogloss:2.25579\ttest-merror:0.75000\n",
      "[350]\ttrain-mlogloss:1.27339\ttrain-merror:0.25000\ttest-mlogloss:2.24007\ttest-merror:0.75000\n",
      "[375]\ttrain-mlogloss:1.26064\ttrain-merror:0.25000\ttest-mlogloss:2.22638\ttest-merror:0.75000\n",
      "[400]\ttrain-mlogloss:1.24965\ttrain-merror:0.25000\ttest-mlogloss:2.21440\ttest-merror:0.75000\n",
      "[425]\ttrain-mlogloss:1.24015\ttrain-merror:0.25000\ttest-mlogloss:2.20388\ttest-merror:0.75000\n",
      "[450]\ttrain-mlogloss:1.23189\ttrain-merror:0.25000\ttest-mlogloss:2.19461\ttest-merror:0.75000\n",
      "[475]\ttrain-mlogloss:1.22468\ttrain-merror:0.25000\ttest-mlogloss:2.18640\ttest-merror:0.75000\n",
      "[500]\ttrain-mlogloss:1.21837\ttrain-merror:0.25000\ttest-mlogloss:2.17912\ttest-merror:0.75000\n",
      "[525]\ttrain-mlogloss:1.21283\ttrain-merror:0.25000\ttest-mlogloss:2.17264\ttest-merror:0.75000\n",
      "[550]\ttrain-mlogloss:1.20795\ttrain-merror:0.25000\ttest-mlogloss:2.16686\ttest-merror:0.75000\n",
      "[575]\ttrain-mlogloss:1.20363\ttrain-merror:0.25000\ttest-mlogloss:2.16169\ttest-merror:0.75000\n",
      "[600]\ttrain-mlogloss:1.19981\ttrain-merror:0.25000\ttest-mlogloss:2.15707\ttest-merror:0.75000\n",
      "[625]\ttrain-mlogloss:1.19641\ttrain-merror:0.25000\ttest-mlogloss:2.15292\ttest-merror:0.75000\n",
      "[650]\ttrain-mlogloss:1.19339\ttrain-merror:0.25000\ttest-mlogloss:2.14919\ttest-merror:0.75000\n",
      "[675]\ttrain-mlogloss:1.19069\ttrain-merror:0.25000\ttest-mlogloss:2.14583\ttest-merror:0.75000\n",
      "[700]\ttrain-mlogloss:1.18828\ttrain-merror:0.25000\ttest-mlogloss:2.14280\ttest-merror:0.75000\n",
      "[725]\ttrain-mlogloss:1.18612\ttrain-merror:0.25000\ttest-mlogloss:2.14007\ttest-merror:0.75000\n",
      "[750]\ttrain-mlogloss:1.18417\ttrain-merror:0.25000\ttest-mlogloss:2.13760\ttest-merror:0.75000\n",
      "[775]\ttrain-mlogloss:1.18241\ttrain-merror:0.25000\ttest-mlogloss:2.13537\ttest-merror:0.75000\n",
      "[800]\ttrain-mlogloss:1.18081\ttrain-merror:0.25000\ttest-mlogloss:2.13335\ttest-merror:0.75000\n",
      "[825]\ttrain-mlogloss:1.17934\ttrain-merror:0.25000\ttest-mlogloss:2.13151\ttest-merror:0.75000\n",
      "[850]\ttrain-mlogloss:1.17800\ttrain-merror:0.25000\ttest-mlogloss:2.12985\ttest-merror:0.75000\n",
      "[875]\ttrain-mlogloss:1.17678\ttrain-merror:0.25000\ttest-mlogloss:2.12834\ttest-merror:0.75000\n",
      "[900]\ttrain-mlogloss:1.17565\ttrain-merror:0.25000\ttest-mlogloss:2.12697\ttest-merror:0.75000\n",
      "[925]\ttrain-mlogloss:1.17461\ttrain-merror:0.25000\ttest-mlogloss:2.12572\ttest-merror:0.75000\n",
      "[950]\ttrain-mlogloss:1.17366\ttrain-merror:0.25000\ttest-mlogloss:2.12459\ttest-merror:0.75000\n",
      "[975]\ttrain-mlogloss:1.17277\ttrain-merror:0.25000\ttest-mlogloss:2.12355\ttest-merror:0.75000\n",
      "[999]\ttrain-mlogloss:1.17199\ttrain-merror:0.25000\ttest-mlogloss:2.12265\ttest-merror:0.75000\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Setting seed\n",
    "torch.manual_seed(30026)\n",
    "if(torch.cuda.is_available()):\n",
    "    torch.cuda.manual_seed(30026)\n",
    "    torch.cuda.manual_seed_all(30026)\n",
    "\n",
    "# Running the second model\n",
    "train_loss, train_accuracy, test_loss, test_accuracy, static_model_dict = static_training(num_classes, xgb_params, dataloader_train, dataloader_test, verbose, device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Third Model: StaticModelMultibranch"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The last static models is created with more complex architecture: since from biological analysis we expect the first and last amino acids of each sequence to be the most representative for our task, this neural networks takes as input also the strings of first and last 20 amino acids of each sequence. These strings are embedded by some transformer layers and then given to some MLP layer together with ESM-embeddings."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Change model_type for running the right model (from 0 to 2, for more info look at static_training inside runner_static.py)\n",
    "static_params[\"model_type\"] +=2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting training of StaticModelMultibranch\n",
      "Epoch [1/100],\tTraining Loss: 2.2329, Training accuracy: 32.80,\tTest Loss: 1.6835, Test Accuracy: 41.04,\tLearning rate: 0.01000\n",
      "Epoch [2/100],\tTraining Loss: 1.6708, Training accuracy: 42.47,\tTest Loss: 1.5014, Test Accuracy: 48.94,\tLearning rate: 0.01000\n",
      "Epoch [3/100],\tTraining Loss: 1.5690, Training accuracy: 46.77,\tTest Loss: 1.4128, Test Accuracy: 50.63,\tLearning rate: 0.01000\n",
      "Epoch [4/100],\tTraining Loss: 1.4871, Training accuracy: 49.45,\tTest Loss: 1.4161, Test Accuracy: 50.35,\tLearning rate: 0.01000\n",
      "Epoch [5/100],\tTraining Loss: 1.4626, Training accuracy: 50.55,\tTest Loss: 1.4205, Test Accuracy: 52.47,\tLearning rate: 0.01000\n",
      "Epoch [6/100],\tTraining Loss: 1.4302, Training accuracy: 51.92,\tTest Loss: 1.3975, Test Accuracy: 52.75,\tLearning rate: 0.01000\n",
      "Epoch [7/100],\tTraining Loss: 1.3863, Training accuracy: 51.96,\tTest Loss: 1.4016, Test Accuracy: 52.89,\tLearning rate: 0.01000\n",
      "Epoch [8/100],\tTraining Loss: 1.3317, Training accuracy: 54.78,\tTest Loss: 1.3868, Test Accuracy: 50.49,\tLearning rate: 0.01000\n",
      "Epoch [9/100],\tTraining Loss: 1.3050, Training accuracy: 55.84,\tTest Loss: 1.3604, Test Accuracy: 52.47,\tLearning rate: 0.01000\n",
      "Epoch [10/100],\tTraining Loss: 1.2836, Training accuracy: 55.45,\tTest Loss: 1.3313, Test Accuracy: 55.15,\tLearning rate: 0.01000\n",
      "Epoch [11/100],\tTraining Loss: 1.2715, Training accuracy: 56.23,\tTest Loss: 1.3872, Test Accuracy: 52.19,\tLearning rate: 0.01000\n",
      "Epoch [12/100],\tTraining Loss: 1.2500, Training accuracy: 57.18,\tTest Loss: 1.3534, Test Accuracy: 55.43,\tLearning rate: 0.01000\n",
      "Epoch [13/100],\tTraining Loss: 1.1867, Training accuracy: 58.94,\tTest Loss: 1.3226, Test Accuracy: 54.02,\tLearning rate: 0.01000\n",
      "Epoch [14/100],\tTraining Loss: 1.1953, Training accuracy: 59.40,\tTest Loss: 1.4220, Test Accuracy: 54.02,\tLearning rate: 0.01000\n",
      "Epoch [15/100],\tTraining Loss: 1.1740, Training accuracy: 59.08,\tTest Loss: 1.3474, Test Accuracy: 53.46,\tLearning rate: 0.01000\n",
      "Epoch [16/100],\tTraining Loss: 1.1401, Training accuracy: 60.07,\tTest Loss: 1.5239, Test Accuracy: 53.88,\tLearning rate: 0.01000\n",
      "Epoch [17/100],\tTraining Loss: 1.1469, Training accuracy: 60.04,\tTest Loss: 1.3448, Test Accuracy: 55.99,\tLearning rate: 0.01000\n",
      "Epoch [18/100],\tTraining Loss: 1.1271, Training accuracy: 61.59,\tTest Loss: 1.3481, Test Accuracy: 55.71,\tLearning rate: 0.01000\n",
      "Epoch [19/100],\tTraining Loss: 1.0979, Training accuracy: 61.41,\tTest Loss: 1.4255, Test Accuracy: 53.88,\tLearning rate: 0.01000\n",
      "Epoch [20/100],\tTraining Loss: 1.0649, Training accuracy: 62.72,\tTest Loss: 1.4189, Test Accuracy: 56.28,\tLearning rate: 0.00100\n",
      "Epoch [21/100],\tTraining Loss: 0.9677, Training accuracy: 65.26,\tTest Loss: 1.3658, Test Accuracy: 58.25,\tLearning rate: 0.00100\n",
      "Epoch [22/100],\tTraining Loss: 0.8907, Training accuracy: 68.18,\tTest Loss: 1.3983, Test Accuracy: 57.26,\tLearning rate: 0.00100\n",
      "Epoch [23/100],\tTraining Loss: 0.8902, Training accuracy: 67.58,\tTest Loss: 1.4064, Test Accuracy: 58.39,\tLearning rate: 0.00100\n",
      "Epoch [24/100],\tTraining Loss: 0.8704, Training accuracy: 68.43,\tTest Loss: 1.4293, Test Accuracy: 57.26,\tLearning rate: 0.00100\n",
      "Epoch [25/100],\tTraining Loss: 0.8400, Training accuracy: 69.95,\tTest Loss: 1.4111, Test Accuracy: 57.12,\tLearning rate: 0.00100\n",
      "Epoch [26/100],\tTraining Loss: 0.8247, Training accuracy: 69.91,\tTest Loss: 1.4417, Test Accuracy: 58.25,\tLearning rate: 0.00100\n",
      "Epoch [27/100],\tTraining Loss: 0.8162, Training accuracy: 69.63,\tTest Loss: 1.4412, Test Accuracy: 57.40,\tLearning rate: 0.00100\n",
      "Epoch [28/100],\tTraining Loss: 0.7977, Training accuracy: 70.51,\tTest Loss: 1.5046, Test Accuracy: 57.55,\tLearning rate: 0.00100\n",
      "Epoch [29/100],\tTraining Loss: 0.7895, Training accuracy: 71.08,\tTest Loss: 1.5180, Test Accuracy: 57.97,\tLearning rate: 0.00100\n",
      "Epoch [30/100],\tTraining Loss: 0.7734, Training accuracy: 71.29,\tTest Loss: 1.5153, Test Accuracy: 57.26,\tLearning rate: 0.00100\n",
      "Epoch [31/100],\tTraining Loss: 0.7667, Training accuracy: 71.57,\tTest Loss: 1.5150, Test Accuracy: 57.69,\tLearning rate: 0.00100\n",
      "Epoch [32/100],\tTraining Loss: 0.7538, Training accuracy: 71.89,\tTest Loss: 1.5534, Test Accuracy: 58.39,\tLearning rate: 0.00100\n",
      "Epoch [33/100],\tTraining Loss: 0.7309, Training accuracy: 73.26,\tTest Loss: 1.5492, Test Accuracy: 56.98,\tLearning rate: 0.00100\n",
      "Early stopping triggered after 33 epochs. Best test loss: 1.3226\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Setting seed\n",
    "torch.manual_seed(30026)\n",
    "if(torch.cuda.is_available()):\n",
    "    torch.cuda.manual_seed(30026)\n",
    "    torch.cuda.manual_seed_all(30026)\n",
    "\n",
    "# Running the third model\n",
    "train_loss, train_accuracy, test_loss, test_accuracy, static_model_dict = static_training(num_classes, static_params, dataloader_train, dataloader_test, verbose, device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By analyzing our main metrics (validation loss and validation accuracy) and the behavior in terms of overfitting, we can conclude that this is the best static model we have created.\n",
    "For this reason, we perform k-fold cross-validation for finding the best hyperparameters of our model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### K-Fold Cross validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Setting seed\n",
    "torch.manual_seed(30026)\n",
    "if(torch.cuda.is_available()):\n",
    "    torch.cuda.manual_seed(30026)\n",
    "    torch.cuda.manual_seed_all(30026)\n",
    "\n",
    "# Choose parameters for grid cross-validation\n",
    "parameter_grid = {\n",
    "    \"dropout\": [0, 0.25, 0.5],\n",
    "    \"learning_rate\": [0.001, 0.01],\n",
    "    \"weight_decay\": [0, 0.0001, 0.01],\n",
    "    \"scheduler\": ['CosineAnnealingLR', 'StepLR', 'ExponentialLR']\n",
    "}\n",
    "\n",
    "# Setting principal fixed parameters for k-fold (i.e. number of folds and metric to use for choosing best model)\n",
    "k_folds = 3\n",
    "cross_validation_on_loss = True\n",
    "\n",
    "best_params, results = k_fold_cross_validation_static(\n",
    "    StaticModelMultibranch,\n",
    "    data,\n",
    "    CrossEntropy,\n",
    "    parameter_grid,\n",
    "    static_params[\"num_epochs\"],\n",
    "    static_params[\"patience\"],\n",
    "    params[\"batch_size\"],\n",
    "    device,\n",
    "    k_folds=k_folds,\n",
    "    seed=seed,\n",
    "    cross_validation_on_loss=cross_validation_on_loss,\n",
    "    verbose=True\n",
    ")\n",
    "\n",
    "# We save best results in a json file for using them in a second moment\n",
    "results_file = 'hyperparameters/cross_val_results_static'\n",
    "with open(results_file, 'w') as f:\n",
    "    json.dump(results, f, indent=4)\n",
    "\n",
    "best_file = 'hyperparameters/best_results_static'\n",
    "with open(best_file, 'w') as f:\n",
    "    json.dump(best_params, f, indent=4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Starting from this parameters, we create a new `hyperparameters_cross_validation.json` file for running our best model on the hyperparameters obtained with 3-fold cross-validation. The result of this training will be our best static architecture and will be used in the second part of the project."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load hyperparameters\n",
    "with open('hyperparameters/hyperparameters_cross_val.json', 'r') as file:\n",
    "    params = json.load(file)\n",
    "\n",
    "static_params = params[\"static_model\"]\n",
    "\n",
    "# Set embedding dimensions based on json parameter\n",
    "if static_params[\"long_embeddings\"]:\n",
    "    embedding_dim = 640\n",
    "else:\n",
    "    embedding_dim = 320"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting training of StaticModelMultibranch\n",
      "Epoch [1/100],\tTraining Loss: 2.3190, Training accuracy: 30.09,\tTest Loss: 1.7884, Test Accuracy: 39.92,\tLearning rate: 0.00100\n",
      "Epoch [2/100],\tTraining Loss: 1.8186, Training accuracy: 37.43,\tTest Loss: 1.6091, Test Accuracy: 46.12,\tLearning rate: 0.00100\n",
      "Epoch [3/100],\tTraining Loss: 1.6574, Training accuracy: 43.74,\tTest Loss: 1.5300, Test Accuracy: 47.95,\tLearning rate: 0.00100\n",
      "Epoch [4/100],\tTraining Loss: 1.5609, Training accuracy: 46.07,\tTest Loss: 1.4175, Test Accuracy: 51.06,\tLearning rate: 0.00100\n",
      "Epoch [5/100],\tTraining Loss: 1.4742, Training accuracy: 50.90,\tTest Loss: 1.4578, Test Accuracy: 49.51,\tLearning rate: 0.00100\n",
      "Epoch [6/100],\tTraining Loss: 1.4318, Training accuracy: 51.64,\tTest Loss: 1.3743, Test Accuracy: 52.89,\tLearning rate: 0.00100\n",
      "Epoch [7/100],\tTraining Loss: 1.3779, Training accuracy: 53.30,\tTest Loss: 1.3855, Test Accuracy: 53.46,\tLearning rate: 0.00100\n",
      "Epoch [8/100],\tTraining Loss: 1.3247, Training accuracy: 54.39,\tTest Loss: 1.3506, Test Accuracy: 54.16,\tLearning rate: 0.00100\n",
      "Epoch [9/100],\tTraining Loss: 1.2821, Training accuracy: 56.44,\tTest Loss: 1.3315, Test Accuracy: 54.87,\tLearning rate: 0.00100\n",
      "Epoch [10/100],\tTraining Loss: 1.2237, Training accuracy: 57.60,\tTest Loss: 1.3132, Test Accuracy: 54.72,\tLearning rate: 0.00100\n",
      "Epoch [11/100],\tTraining Loss: 1.2033, Training accuracy: 58.77,\tTest Loss: 1.3172, Test Accuracy: 54.44,\tLearning rate: 0.00100\n",
      "Epoch [12/100],\tTraining Loss: 1.1621, Training accuracy: 60.11,\tTest Loss: 1.3010, Test Accuracy: 55.99,\tLearning rate: 0.00100\n",
      "Epoch [13/100],\tTraining Loss: 1.1296, Training accuracy: 61.45,\tTest Loss: 1.2999, Test Accuracy: 56.42,\tLearning rate: 0.00100\n",
      "Epoch [14/100],\tTraining Loss: 1.0913, Training accuracy: 61.16,\tTest Loss: 1.3038, Test Accuracy: 56.42,\tLearning rate: 0.00100\n",
      "Epoch [15/100],\tTraining Loss: 1.0540, Training accuracy: 62.68,\tTest Loss: 1.3187, Test Accuracy: 55.99,\tLearning rate: 0.00100\n",
      "Epoch [16/100],\tTraining Loss: 1.0400, Training accuracy: 62.96,\tTest Loss: 1.3128, Test Accuracy: 58.53,\tLearning rate: 0.00100\n",
      "Epoch [17/100],\tTraining Loss: 1.0308, Training accuracy: 63.49,\tTest Loss: 1.2955, Test Accuracy: 57.55,\tLearning rate: 0.00100\n",
      "Epoch [18/100],\tTraining Loss: 0.9774, Training accuracy: 65.43,\tTest Loss: 1.3123, Test Accuracy: 56.98,\tLearning rate: 0.00100\n",
      "Epoch [19/100],\tTraining Loss: 0.9611, Training accuracy: 65.33,\tTest Loss: 1.2962, Test Accuracy: 57.69,\tLearning rate: 0.00100\n",
      "Epoch [20/100],\tTraining Loss: 0.9236, Training accuracy: 67.13,\tTest Loss: 1.3248, Test Accuracy: 56.28,\tLearning rate: 0.00010\n",
      "Epoch [21/100],\tTraining Loss: 0.8274, Training accuracy: 70.51,\tTest Loss: 1.3086, Test Accuracy: 58.25,\tLearning rate: 0.00010\n",
      "Epoch [22/100],\tTraining Loss: 0.7997, Training accuracy: 71.11,\tTest Loss: 1.3159, Test Accuracy: 58.11,\tLearning rate: 0.00010\n",
      "Epoch [23/100],\tTraining Loss: 0.7651, Training accuracy: 72.49,\tTest Loss: 1.3307, Test Accuracy: 58.39,\tLearning rate: 0.00010\n",
      "Epoch [24/100],\tTraining Loss: 0.7514, Training accuracy: 73.19,\tTest Loss: 1.3238, Test Accuracy: 58.67,\tLearning rate: 0.00010\n",
      "Epoch [25/100],\tTraining Loss: 0.7597, Training accuracy: 71.92,\tTest Loss: 1.3381, Test Accuracy: 58.53,\tLearning rate: 0.00010\n",
      "Epoch [26/100],\tTraining Loss: 0.7389, Training accuracy: 73.69,\tTest Loss: 1.3432, Test Accuracy: 57.97,\tLearning rate: 0.00010\n",
      "Epoch [27/100],\tTraining Loss: 0.7509, Training accuracy: 73.79,\tTest Loss: 1.3462, Test Accuracy: 58.11,\tLearning rate: 0.00010\n",
      "Epoch [28/100],\tTraining Loss: 0.7174, Training accuracy: 73.83,\tTest Loss: 1.3660, Test Accuracy: 58.82,\tLearning rate: 0.00010\n",
      "Epoch [29/100],\tTraining Loss: 0.7462, Training accuracy: 72.80,\tTest Loss: 1.3573, Test Accuracy: 58.82,\tLearning rate: 0.00010\n",
      "Epoch [30/100],\tTraining Loss: 0.7360, Training accuracy: 73.97,\tTest Loss: 1.3543, Test Accuracy: 59.10,\tLearning rate: 0.00010\n",
      "Epoch [31/100],\tTraining Loss: 0.7014, Training accuracy: 74.14,\tTest Loss: 1.3593, Test Accuracy: 59.10,\tLearning rate: 0.00010\n",
      "Epoch [32/100],\tTraining Loss: 0.7084, Training accuracy: 73.93,\tTest Loss: 1.3696, Test Accuracy: 59.10,\tLearning rate: 0.00010\n",
      "Epoch [33/100],\tTraining Loss: 0.7135, Training accuracy: 73.47,\tTest Loss: 1.3655, Test Accuracy: 58.11,\tLearning rate: 0.00010\n",
      "Epoch [34/100],\tTraining Loss: 0.7138, Training accuracy: 74.64,\tTest Loss: 1.3668, Test Accuracy: 57.97,\tLearning rate: 0.00010\n",
      "Epoch [35/100],\tTraining Loss: 0.7030, Training accuracy: 74.74,\tTest Loss: 1.3820, Test Accuracy: 58.11,\tLearning rate: 0.00010\n",
      "Epoch [36/100],\tTraining Loss: 0.7034, Training accuracy: 73.79,\tTest Loss: 1.3767, Test Accuracy: 58.53,\tLearning rate: 0.00010\n",
      "Epoch [37/100],\tTraining Loss: 0.7014, Training accuracy: 74.36,\tTest Loss: 1.3698, Test Accuracy: 58.96,\tLearning rate: 0.00010\n",
      "Early stopping triggered after 37 epochs. Best test loss: 1.2955\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Setting seed\n",
    "torch.manual_seed(30026)\n",
    "if(torch.cuda.is_available()):\n",
    "    torch.cuda.manual_seed(30026)\n",
    "    torch.cuda.manual_seed_all(30026)\n",
    "\n",
    "# Running the third model\n",
    "train_loss, train_accuracy, test_loss, test_accuracy, static_model_dict = static_training(num_classes, static_params, dataloader_train, dataloader_test, verbose, device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<All keys matched successfully>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Save best model\n",
    "best_static_model = StaticModelMultibranch(num_classes=num_classes, embedding_dim=embedding_dim, extremities_dim=20, char_vocab_size=20, char_embed_dim=16, intermediate_dim=static_params[\"intermediate_dim\"], dropout=static_params[\"dropout\"])\n",
    "\n",
    "# Load the weights from the dictionary\n",
    "best_static_model.load_state_dict(static_model_dict)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dynamic Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this second part, we try to solve the dynamic counterpart of the localization problem: every protein can now change position during eukaryotic cell cycle, which consists of 5 different timesteps."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## First Model: LSTMDynamicModel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The first model consists of a LSTM architecture which takes as input the dynamic data and the output of the StaticModelMultibranch static model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load hyperparameters\n",
    "dynamic_params = params[\"dynamic_model\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting training of LSTMDynamicModel\n",
      "Epoch [1/100],\tTraining Loss: 1.9174, Training accuracy: 38.40,\tTest Loss: 1.6963, Test Accuracy: 45.33,\tLearning rate: 0.00100\n",
      "Epoch [2/100],\tTraining Loss: 1.2839, Training accuracy: 58.29,\tTest Loss: 1.4687, Test Accuracy: 55.94,\tLearning rate: 0.00100\n",
      "Epoch [3/100],\tTraining Loss: 1.1109, Training accuracy: 66.02,\tTest Loss: 1.4595, Test Accuracy: 55.91,\tLearning rate: 0.00100\n",
      "Epoch [4/100],\tTraining Loss: 1.0604, Training accuracy: 68.28,\tTest Loss: 1.4300, Test Accuracy: 57.38,\tLearning rate: 0.00100\n",
      "Epoch [5/100],\tTraining Loss: 0.9892, Training accuracy: 69.99,\tTest Loss: 1.4493, Test Accuracy: 57.63,\tLearning rate: 0.00100\n",
      "Epoch [6/100],\tTraining Loss: 0.9634, Training accuracy: 70.86,\tTest Loss: 1.4570, Test Accuracy: 57.32,\tLearning rate: 0.00100\n",
      "Epoch [7/100],\tTraining Loss: 0.9487, Training accuracy: 70.48,\tTest Loss: 1.3936, Test Accuracy: 57.01,\tLearning rate: 0.00100\n",
      "Epoch [8/100],\tTraining Loss: 0.9204, Training accuracy: 71.86,\tTest Loss: 1.4134, Test Accuracy: 57.40,\tLearning rate: 0.00100\n",
      "Epoch [9/100],\tTraining Loss: 0.9324, Training accuracy: 71.85,\tTest Loss: 1.4366, Test Accuracy: 58.39,\tLearning rate: 0.00100\n",
      "Epoch [10/100],\tTraining Loss: 0.9194, Training accuracy: 72.51,\tTest Loss: 1.3548, Test Accuracy: 58.17,\tLearning rate: 0.00100\n",
      "Epoch [11/100],\tTraining Loss: 0.8705, Training accuracy: 73.20,\tTest Loss: 1.4263, Test Accuracy: 58.19,\tLearning rate: 0.00100\n",
      "Epoch [12/100],\tTraining Loss: 0.8818, Training accuracy: 72.35,\tTest Loss: 1.4208, Test Accuracy: 58.00,\tLearning rate: 0.00100\n",
      "Epoch [13/100],\tTraining Loss: 0.8896, Training accuracy: 72.97,\tTest Loss: 1.3749, Test Accuracy: 58.42,\tLearning rate: 0.00100\n",
      "Epoch [14/100],\tTraining Loss: 0.8539, Training accuracy: 73.26,\tTest Loss: 1.3478, Test Accuracy: 58.45,\tLearning rate: 0.00100\n",
      "Epoch [15/100],\tTraining Loss: 0.8628, Training accuracy: 73.42,\tTest Loss: 1.3494, Test Accuracy: 58.34,\tLearning rate: 0.00100\n",
      "Epoch [16/100],\tTraining Loss: 0.8580, Training accuracy: 73.26,\tTest Loss: 1.3646, Test Accuracy: 58.73,\tLearning rate: 0.00100\n",
      "Epoch [17/100],\tTraining Loss: 0.8472, Training accuracy: 73.43,\tTest Loss: 1.3846, Test Accuracy: 58.00,\tLearning rate: 0.00100\n",
      "Epoch [18/100],\tTraining Loss: 0.8284, Training accuracy: 74.24,\tTest Loss: 1.4260, Test Accuracy: 59.69,\tLearning rate: 0.00100\n",
      "Epoch [19/100],\tTraining Loss: 0.8432, Training accuracy: 73.64,\tTest Loss: 1.3420, Test Accuracy: 58.50,\tLearning rate: 0.00100\n",
      "Epoch [20/100],\tTraining Loss: 0.8320, Training accuracy: 73.83,\tTest Loss: 1.4416, Test Accuracy: 59.04,\tLearning rate: 0.00010\n",
      "Epoch [21/100],\tTraining Loss: 0.8167, Training accuracy: 74.34,\tTest Loss: 1.3790, Test Accuracy: 58.90,\tLearning rate: 0.00010\n",
      "Epoch [22/100],\tTraining Loss: 0.8179, Training accuracy: 74.48,\tTest Loss: 1.3707, Test Accuracy: 58.65,\tLearning rate: 0.00010\n",
      "Epoch [23/100],\tTraining Loss: 0.8071, Training accuracy: 75.03,\tTest Loss: 1.3788, Test Accuracy: 58.93,\tLearning rate: 0.00010\n",
      "Epoch [24/100],\tTraining Loss: 0.8224, Training accuracy: 74.26,\tTest Loss: 1.3673, Test Accuracy: 58.93,\tLearning rate: 0.00010\n",
      "Epoch [25/100],\tTraining Loss: 0.7985, Training accuracy: 74.74,\tTest Loss: 1.3659, Test Accuracy: 58.87,\tLearning rate: 0.00010\n",
      "Epoch [26/100],\tTraining Loss: 0.8046, Training accuracy: 74.35,\tTest Loss: 1.3668, Test Accuracy: 58.96,\tLearning rate: 0.00010\n",
      "Epoch [27/100],\tTraining Loss: 0.8262, Training accuracy: 73.95,\tTest Loss: 1.3557, Test Accuracy: 58.82,\tLearning rate: 0.00010\n",
      "Epoch [28/100],\tTraining Loss: 0.7910, Training accuracy: 75.68,\tTest Loss: 1.3634, Test Accuracy: 59.24,\tLearning rate: 0.00010\n",
      "Epoch [29/100],\tTraining Loss: 0.7869, Training accuracy: 75.60,\tTest Loss: 1.3747, Test Accuracy: 59.35,\tLearning rate: 0.00010\n",
      "Epoch [30/100],\tTraining Loss: 0.7979, Training accuracy: 74.63,\tTest Loss: 1.3866, Test Accuracy: 59.21,\tLearning rate: 0.00010\n",
      "Epoch [31/100],\tTraining Loss: 0.8262, Training accuracy: 73.68,\tTest Loss: 1.3677, Test Accuracy: 58.93,\tLearning rate: 0.00010\n",
      "Epoch [32/100],\tTraining Loss: 0.8041, Training accuracy: 75.16,\tTest Loss: 1.3660, Test Accuracy: 58.73,\tLearning rate: 0.00010\n",
      "Epoch [33/100],\tTraining Loss: 0.7915, Training accuracy: 74.21,\tTest Loss: 1.3726, Test Accuracy: 58.98,\tLearning rate: 0.00010\n",
      "Epoch [34/100],\tTraining Loss: 0.8168, Training accuracy: 74.80,\tTest Loss: 1.3723, Test Accuracy: 59.10,\tLearning rate: 0.00010\n",
      "Epoch [35/100],\tTraining Loss: 0.7938, Training accuracy: 74.80,\tTest Loss: 1.3784, Test Accuracy: 59.10,\tLearning rate: 0.00010\n",
      "Epoch [36/100],\tTraining Loss: 0.7905, Training accuracy: 75.08,\tTest Loss: 1.3740, Test Accuracy: 59.24,\tLearning rate: 0.00010\n",
      "Epoch [37/100],\tTraining Loss: 0.7979, Training accuracy: 74.86,\tTest Loss: 1.3686, Test Accuracy: 59.49,\tLearning rate: 0.00010\n",
      "Epoch [38/100],\tTraining Loss: 0.7779, Training accuracy: 75.84,\tTest Loss: 1.3699, Test Accuracy: 59.41,\tLearning rate: 0.00010\n",
      "Epoch [39/100],\tTraining Loss: 0.8202, Training accuracy: 74.20,\tTest Loss: 1.3654, Test Accuracy: 59.21,\tLearning rate: 0.00010\n",
      "Early stopping triggered after 39 epochs. Best test loss: 1.3420\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA04AAAIjCAYAAAA0vUuxAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAi09JREFUeJzs3Xd8U9X/x/FX0l26aAu0rLL3FAQBQZSNVgH9OnCACwe4UH9OlvMrivp1b3CBG1ygDAUUQQQpIiCy94ZSSunM/f1xaEtpaUvb5Kbt+/l43EeS25vkk9MU8s459xyHZVkWIiIiIiIiclpOuwsQERERERHxdgpOIiIiIiIiRVBwEhERERERKYKCk4iIiIiISBEUnERERERERIqg4CQiIiIiIlIEBScREREREZEiKDiJiIiIiIgUQcFJRERERESkCApOIiKV1PDhw6lXr16J7jt+/HgcDkfZFuRltmzZgsPhYMqUKXaXIiIiXkDBSUTEyzgcjmJt8+fPt7vUSq9evXrF+l2VVfh66qmnmDFjRrGOzQ5+zz33XJk8t4hIZedrdwEiIpLXhx9+mOf2Bx98wJw5c/Ltb968eame5+2338blcpXovo8++igPPvhgqZ6/InjxxRdJTk7OuT1z5kymTZvGCy+8QHR0dM7+rl27lsnzPfXUU1x22WUMGjSoTB5PRESKT8FJRMTLXHPNNXluL1myhDlz5uTbf6qUlBSCg4OL/Tx+fn4lqg/A19cXX1/9F3JqgNmzZw/Tpk1j0KBBJR4GKSIi3klD9UREyqGePXvSqlUrli9fTo8ePQgODubhhx8G4Ouvv+bCCy+kZs2aBAQE0LBhQx5//HGysrLyPMap5zidPLTrrbfeomHDhgQEBHD22Wfzxx9/5LlvQec4ORwORo0axYwZM2jVqhUBAQG0bNmSH374IV/98+fPp2PHjgQGBtKwYUPefPPNYp839csvv/Cf//yHunXrEhAQQJ06dbjnnns4fvx4vtcXEhLCzp07GTRoECEhIVSrVo377rsvX1skJiYyfPhwwsPDiYiIYNiwYSQmJhZZS3F99NFHdOjQgaCgICIjI7nyyivZvn17nmPWr1/PpZdeSkxMDIGBgdSuXZsrr7ySI0eOAKZ9jx07xvvvv58zBHD48OGlrm3fvn3ceOON1KhRg8DAQNq2bcv777+f77hPPvmEDh06EBoaSlhYGK1bt+Z///tfzs8zMjKYMGECjRs3JjAwkKioKM4991zmzJlT6hpFRLyBvi4UESmnDh48yIABA7jyyiu55pprqFGjBgBTpkwhJCSE0aNHExISwk8//cTYsWNJSkri2WefLfJxp06dytGjR7nllltwOBxMnDiRIUOGsGnTpiJ7qX799Ve++uorbr/9dkJDQ3nppZe49NJL2bZtG1FRUQCsWLGC/v37Exsby4QJE8jKyuKxxx6jWrVqxXrdn3/+OSkpKdx2221ERUWxdOlSXn75ZXbs2MHnn3+e59isrCz69etH586dee6555g7dy6TJk2iYcOG3HbbbQBYlsUll1zCr7/+yq233krz5s2ZPn06w4YNK1Y9RXnyyScZM2YMl19+OTfddBP79+/n5ZdfpkePHqxYsYKIiAjS09Pp168faWlp3HHHHcTExLBz506+++47EhMTCQ8P58MPP+Smm26iU6dOjBgxAoCGDRuWqrbjx4/Ts2dPNmzYwKhRo6hfvz6ff/45w4cPJzExkbvuuguAOXPmcNVVV9GrVy+eeeYZANauXcuiRYtyjhk/fjxPP/10To1JSUksW7aMP//8kz59+pSqThERr2CJiIhXGzlypHXqP9fnnXeeBVhvvPFGvuNTUlLy7bvlllus4OBgKzU1NWffsGHDrLi4uJzbmzdvtgArKirKOnToUM7+r7/+2gKsb7/9NmffuHHj8tUEWP7+/taGDRty9q1cudICrJdffjlnX3x8vBUcHGzt3LkzZ9/69estX1/ffI9ZkIJe39NPP205HA5r69ateV4fYD322GN5jm3fvr3VoUOHnNszZsywAGvixIk5+zIzM63u3btbgDV58uQia8r27LPPWoC1efNmy7Isa8uWLZaPj4/15JNP5jlu1apVlq+vb87+FStWWID1+eefF/r4VapUsYYNG1asWrJ/n88+++xpj3nxxRctwProo49y9qWnp1tdunSxQkJCrKSkJMuyLOuuu+6ywsLCrMzMzNM+Vtu2ba0LL7ywWLWJiJRHGqonIlJOBQQEcP311+fbHxQUlHP96NGjHDhwgO7du5OSksI///xT5ONeccUVVK1aNed29+7dAdi0aVOR9+3du3eeXpA2bdoQFhaWc9+srCzmzp3LoEGDqFmzZs5xjRo1YsCAAUU+PuR9fceOHePAgQN07doVy7JYsWJFvuNvvfXWPLe7d++e57XMnDkTX1/fnB4oAB8fH+64445i1VOYr776CpfLxeWXX86BAwdytpiYGBo3bszPP/8MQHh4OAA//vgjKSkppX7e4po5cyYxMTFcddVVOfv8/Py48847SU5OZsGCBQBERERw7NixQofdRUREsHr1atavX+/2ukVE7KDgJCJSTtWqVQt/f/98+1evXs3gwYMJDw8nLCyMatWq5UwskX2+TGHq1q2b53Z2iDp8+PAZ3zf7/tn33bdvH8ePH6dRo0b5jitoX0G2bdvG8OHDiYyMzDlv6bzzzgPyv77AwMB8QwBPrgdg69atxMbGEhISkue4pk2bFquewqxfvx7LsmjcuDHVqlXLs61du5Z9+/YBUL9+fUaPHs0777xDdHQ0/fr149VXXy3W76s0tm7dSuPGjXE6834cyJ6xcevWrQDcfvvtNGnShAEDBlC7dm1uuOGGfOeuPfbYYyQmJtKkSRNat27N/fffz19//eXW+kVEPEnnOImIlFMn97xkS0xM5LzzziMsLIzHHnuMhg0bEhgYyJ9//skDDzxQrOnHfXx8CtxvWZZb71scWVlZ9OnTh0OHDvHAAw/QrFkzqlSpws6dOxk+fHi+13e6ejzF5XLhcDiYNWtWgbWcHNYmTZrE8OHD+frrr5k9ezZ33nknTz/9NEuWLKF27dqeLDuf6tWrk5CQwI8//sisWbOYNWsWkydP5rrrrsuZSKJHjx5s3Lgxp/533nmHF154gTfeeIObbrrJ1vpFRMqCgpOISAUyf/58Dh48yFdffUWPHj1y9m/evNnGqnJVr16dwMBANmzYkO9nBe071apVq/j33395//33ue6663L2l2bmtri4OObNm0dycnKeILNu3boSP2a2hg0bYlkW9evXp0mTJkUe37p1a1q3bs2jjz7Kb7/9Rrdu3XjjjTd44oknAIo16+CZiIuL46+//sLlcuXpdcoe0hkXF5ezz9/fn/j4eOLj43G5XNx+++28+eabjBkzJqe3MDIykuuvv57rr7+e5ORkevTowfjx4xWcRKRC0FA9EZEKJLtX4+QenvT0dF577TW7SsrDx8eH3r17M2PGDHbt2pWzf8OGDcyaNatY94e8r8+yrDzTYp+pgQMHkpmZyeuvv56zLysri5dffrnEj5ltyJAh+Pj4MGHChHy9bpZlcfDgQQCSkpLIzMzM8/PWrVvjdDpJS0vL2VelSpUynSZ94MCB7Nmzh08//TRnX2ZmJi+//DIhISE5QyCz68zmdDpp06YNQE59px4TEhJCo0aN8tQvIlKeqcdJRKQC6dq1K1WrVmXYsGHceeedOBwOPvzwwzIbKlcWxo8fz+zZs+nWrRu33XYbWVlZvPLKK7Rq1YqEhIRC79usWTMaNmzIfffdx86dOwkLC+PLL78s1vlXpxMfH0+3bt148MEH2bJlCy1atOCrr74qk/OLGjZsyBNPPMFDDz3Eli1bGDRoEKGhoWzevJnp06czYsQI7rvvPn766SdGjRrFf/7zH5o0aUJmZiYffvghPj4+XHrppTmP16FDB+bOncvzzz9PzZo1qV+/Pp07dy60hnnz5pGamppv/6BBgxgxYgRvvvkmw4cPZ/ny5dSrV48vvviCRYsW8eKLLxIaGgrATTfdxKFDh7jggguoXbs2W7du5eWXX6Zdu3Y550O1aNGCnj170qFDByIjI1m2bBlffPEFo0aNKnU7ioh4AwUnEZEKJCoqiu+++457772XRx99lKpVq3LNNdfQq1cv+vXrZ3d5gPnwP2vWLO677z7GjBlDnTp1eOyxx1i7dm2Rs/75+fnx7bff5pz/ExgYyODBgxk1ahRt27YtUT1Op5NvvvmGu+++m48++giHw8HFF1/MpEmTaN++fYke82QPPvggTZo04YUXXmDChAkA1KlTh759+3LxxRcD0LZtW/r168e3337Lzp07CQ4Opm3btsyaNYtzzjkn57Gef/55RowYwaOPPsrx48cZNmxYkcHphx9+KHAR4nr16tGqVSvmz5/Pgw8+yPvvv09SUhJNmzZl8uTJeRbXveaaa3jrrbd47bXXSExMJCYmhiuuuILx48fnDPG78847+eabb5g9ezZpaWnExcXxxBNPcP/995e2CUVEvILD8qavIUVEpNIaNGiQprMWERGvpXOcRETE444fP57n9vr165k5cyY9e/a0pyAREZEiqMdJREQ8LjY2luHDh9OgQQO2bt3K66+/TlpaGitWrKBx48Z2lyciIpKPznESERGP69+/P9OmTWPPnj0EBATQpUsXnnrqKYUmERHxWupxEhERERERKYLOcRIRERERESmCgpOIiIiIiEgRKt05Ti6Xi127dhEaGorD4bC7HBERERERsYllWRw9epSaNWvmrEt3OpUuOO3atYs6derYXYaIiIiIiHiJ7du3U7t27UKPqXTBKTQ0FDCNExYWZnM1kJGRwezZs+nbty9+fn52l1Phqb09T23ueWpzz1J7e57a3PPU5p6nNveMpKQk6tSpk5MRClPpglP28LywsDCvCU7BwcGEhYXpj8ID1N6epzb3PLW5Z6m9PU9t7nlqc89Tm3tWcU7h0eQQIiIiIiIiRVBwEhERERERKYKCk4iIiIiISBEq3TlOIiIiIiKFsSyLzMxMsrKybKshIyMDX19fUlNTba2jIvDz88PHx6fUj6PgJCIiIiJyQnp6Ort37yYlJcXWOizLIiYmhu3bt2vt0VJyOBzUrl2bkJCQUj2OgpOIiIiICOByudi8eTM+Pj7UrFkTf39/20KLy+UiOTmZkJCQIhdmldOzLIv9+/ezY8cOGjduXKqeJwUnERERERFMb5PL5aJOnToEBwfbWovL5SI9PZ3AwEAFp1KqVq0aW7ZsISMjo1TBSb8FEREREZGTKKhULGXVa6h3hYiIiIiISBEUnERERERERIqg4CQiIiIiIvnUq1ePF1980e4yvIaCk4iIiIhIOeZwOArdxo8fX6LH/eOPPxgxYkSpauvZsyd33313qR7DW2hWPRERERGRcmz37t051z/99FPGjh3LunXrcvadvH6RZVlkZWXh61t0DKhWrVrZFlrOqcdJREREROQ0LAuOHbNns6zi1RgTE5OzhYeH43A4cm7/888/hIaGMmvWLDp06EBAQAC//vorGzdu5JJLLqFGjRqEhIRw9tlnM3fu3DyPe+pQPYfDwTvvvMPgwYMJDg6mcePGfPPNN6Vq3y+//JKWLVsSEBBAvXr1mDRpUp6fv/baazRu3JjAwEBq1KjBZZddlvOzL774gtatWxMUFERUVBS9e/fm2LFjpaqnMOpxEhERERE5jZQUOKnDxoOc7NgB4eFl82gPPvggzz33HA0aNKBq1aps376dgQMH8uSTTxIQEMAHH3xAfHw869ato27duqd9nAkTJjBx4kSeffZZXn75Za6++mq2bt1KZGTkGde0fPlyLr/8csaPH88VV1zBb7/9xu23305UVBTDhw9n2bJl3HnnnXz44Yd07dqVQ4cO8csvvwCml+2qq65i4sSJDB48mKNHj/LLL79gFTdtloCCk4iIiIhIBffYY4/Rp0+fnNuRkZG0bds25/bjjz/O9OnT+eabbxg1atRpH2f48OFcddVVADz11FO89NJLLF26lP79+59xTc8//zy9evVizJgxADRp0oQ1a9bw7LPPMnz4cLZt20aVKlW46KKLCA0NJS4ujvbt2wMmOGVmZjJkyBDi4uIAaN269RnXcCYUnGy0ZQv88YeDjRsjGTjQ7mpERERE5FTBwZCc7PnndblcZGaW3eN17Ngxz+3k5GTGjx/P999/nxNCjh8/zrZt2wp9nDZt2uRcr1KlCmFhYezbt69ENa1du5ZLLrkkz75u3brx4osvkpWVRZ8+fYiLi6NBgwb079+f/v375wwTbNu2Lb169aJ169b069ePvn37ctlll1G1atUS1VIcOsfJRjNmwOWX+/L99/XtLkVERERECuBwQJUq9mwOR9m9jipVquS5fd999zF9+nSeeuopfvnlFxISEmjdujXp6emFPo6fn98p7ePA5XKVXaEnCQ0N5c8//2TatGnExsYyduxY2rZtS2JiIj4+PsyZM4dZs2bRokULXn75ZZo2bcrmzZvdUgsoONkqe6KSpKQAewsRERERkUpl0aJFDB8+nMGDB9O6dWtiYmLYsmWLR2to3rw5ixYtyldXkyZN8PHxAcDX15fevXszceJE/vrrL7Zs2cJPP/0EmNDWrVs3JkyYwIoVK/D392f69Oluq1dD9WwUHW0uk5L87S1ERERERCqVxo0b89VXXxEfH4/D4WDMmDFu6znav38/CQkJefbFxsZy7733cvbZZ/P4449zxRVXsHjxYl555RVee+01AL777js2bdpEjx49qFq1KjNnzsTlctG0aVN+//135s2bR9++falevTq///47+/fvp3nz5m55DaDgZCsFJxERERGxw/PPP88NN9xA165diY6O5oEHHiApKcktzzV16lSmTp2aZ9/jjz/Oo48+ymeffcbYsWN5/PHHiY2N5bHHHmP48OEARERE8NVXXzF+/HhSU1Np3Lgx06ZNo2XLlqxdu5aFCxfy4osvkpSURFxcHJMmTWLAgAFueQ2g4GSr3KF6/liWexK+iIiIiFQew4cPzwkeAD179ixwiu569erlDHnLNnLkyDy3Tx26V9DjJCYmFlrP/PnzC/35pZdeyqWXXlrgz84999zT3r958+b88MMPhT52WdM5TjbK7nHKzPTh6FF7axERERERkdNTcLJRcDAEBZnkfuCAzcWIiIiIiMhpKTjZLHu43sGDZTjfpIiIiIiIlCkFJ5tFRZnL/fvtrUNERERERE5Pwclm0dEaqiciIiIi4u0UnGyWPUGEhuqJiIiIiHgvBSebZfc4aaieiIiIiIj3UnCyWfY5TupxEhERERHxXgpONsueVU/nOImIiIiIeC8FJ5tFRZmhegcP2lyIiIiIiIicloKTzbInh9i/X0P1REREROTMORyOQrfx48eX6rFnzJhRZseVZ752F1DZZU8OoR4nERERESmJ3bt351z/9NNPGTt2LOvWrcvZFxISYkdZFY56nGyW3eN0+DBkZtpbi4iIiIicwrIg85g9m2UVq8SYmJicLTw8HIfDkWffJ598QvPmzQkMDKRZs2a89tprOfdNT09n1KhRxMbGEhgYSFxcHE8//TQA9erVA2Dw4ME4HI6c22fK5XLx2GOPUbt2bQICAmjXrh0//PBDsWqwLIvx48dTt25dAgICqFmzJnfeeWeJ6igt9TjZLDLSXFqWg0OHoHp1e+sRERERkZNkpcBnnu+xcQL03QGEl+pxPv74Y8aOHcsrr7xC+/btWbFiBTfffDNVqlRh2LBhvPTSS3zzzTd89tln1K1bl+3bt7N9+3YA/vjjD6pXr87kyZPp378/Pj4+Jarhf//7H5MmTeLNN9+kffv2vPfee1x88cWsXr2axo0bF1rDl19+yQsvvMAnn3xCy5Yt2bNnDytXrixVm5SUgpPNfH0hNDSdo0f9OXBAwUlEREREys64ceOYNGkSQ4YMAaB+/fqsWbOGN998k2HDhrFt2zYaN27Mueeei8PhIC4uLue+1U5M/xwREUFMTEyJa3juued44IEHuPLKKwF45pln+Pnnn3nxxRd59dVXC61h27ZtxMTE0Lt3b/z8/Khbty6dOnUqcS2loeDkBU4OTiIiIiLiRXyC4fJkjz+ty+WCY6U7j+PYsWNs3LiRG2+8kZtvvjlnf2ZmJuHhpidr+PDh9OnTh6ZNm9K/f38uuugi+vbtW6rnPVlSUhK7du2iW7duefZ369Ytp+eosBr+85//8OKLL9KgQQP69+/PwIEDiY+Px9fX8zFGwckLhIWlsWtXCPv3212JiIiIiOThcIBvFc8/r8sFjqRSPURysgl8b7/9Np07d87zs+xhd2eddRabN29m1qxZzJ07l8svv5zevXvzxRdflOq5z0RhNdSpU4d169Yxd+5c5syZw+23386zzz7LggUL8PPz81iNoMkhvEJYWDqgRXBFREREpOzUqFGDmjVrsmnTJho1apRnq1+/fs5xYWFhXHHFFbz99tt8+umnfPnllxw6dAgAPz8/srKySlxDWFgYNWvWZNGiRXn2L1q0iBYtWhSrhqCgIOLj43nppZeYP38+ixcvZtWqVSWuqaTU4+QFFJxERERExB0mTJjAnXfeSXh4OP379yctLY1ly5Zx+PBhRo8ezfPPP09sbCzt27fH6XTy+eefExMTQ0REBGBm1ps3bx7dunUjICCAqlWrnva5Nm/eTEJCQp59jRs35v7772fcuHE0bNiQdu3aMXnyZBISEvj4448BCq1hypQpZGVl0blzZ4KDg/noo48ICgrKcx6Upyg4eYHs4KSheiIiIiJSlm666SaCg4N59tlnuf/++6lSpQqtW7fm7rvvBiA0NJSJEyeyfv16fHx8OPvss5k5cyZOpxmYNmnSJEaPHs3bb79NrVq12LJly2mfa/To0fn2/fLLL9x5550cOXKEe++9l3379tGiRQu++eYbGjduXGQNERER/Pe//2X06NFkZWXRunVrvv32W6Kiosq8rYqi4OQFwsLSAPU4iYiIiEjpDB8+nOHDh+fZN3ToUIYOHVrg8TfffHOeiSNOFR8fT3x8fJHPaxWx5tS4ceMYN27cGdcwaNAgBg0aVOTze4LOcfICGqonIiIiIuLdFJy8gIbqiYiIiIh4NwUnL6CheiIiIiIi3k3ByQtoqJ6IiIiIiHdTcPIC2cEpJcVsIiIiImKfoiY6kPKlrH6fCk5eICgoE39/8wtVr5OIiIiIPfz8/ABI0TfZFUp6uumk8PHxKdXjaDpyL+BwQHQ07NplglPdunZXJCIiIlL5+Pj4EBERwb59+wAIDg7G4XDYUovL5SI9PZ3U1NScNZXkzLlcLvbv309wcDC+vqWLPgpOXiIqygQnzawnIiIiYp+YmBiAnPBkF8uyOH78OEFBQbaFt4rC6XRSt27dUrejgpOXiI62AIeG6omIiIjYyOFwEBsbS/Xq1cnIyLCtjoyMDBYuXEiPHj1yhhBKyfj7+5dJr52Ck5eIjjaXCk4iIiIi9vPx8Sn1OTGlff7MzEwCAwMVnLyEBkx6CdPjpKF6IiIiIiLeSMHJS0RFmUv1OImIiIiIeB8FJy9RrZq5VHASEREREfE+Ck5eIipKQ/VERERERLyVrcFp4cKFxMfHU7NmTRwOBzNmzCjyPh9//DFt27YlODiY2NhYbrjhBg4ePOj+Yt1Mk0OIiIiIiHgvW4PTsWPHaNu2La+++mqxjl+0aBHXXXcdN954I6tXr+bzzz9n6dKl3HzzzW6u1P2yJ4dQcBIRERER8T62Tkc+YMAABgwYUOzjFy9eTL169bjzzjsBqF+/PrfccgvPPPOMu0r0mOwep4MHweUCLRAtIiIiIuI9ytU6Tl26dOHhhx9m5syZDBgwgH379vHFF18wcODA094nLS2NtLS0nNtJSUmAWVTMzkXNsmXXEBaWAfiRlQX792cQGWlvXRVVdnt7w+++slCbe57a3LPU3p6nNvc8tbnnqc0940za12FZluXGWorN4XAwffp0Bg0aVOhxn3/+OTfccAOpqalkZmYSHx/Pl19+edqFwcaPH8+ECRPy7Z86dSrBwcFlUXqZGTp0ICkpfrz66lxq1TpmdzkiIiIiIhVaSkoKQ4cO5ciRI4SFhRV6bLkKTmvWrKF3797cc8899OvXj927d3P//fdz9tln8+677xZ4n4J6nOrUqcOBAweKbBxPyMjIYM6cOfTp04fWrYPYtMnB/PmZdO3qFb+WCufk9tYq3J6hNvc8tblnqb09T23ueWpzz1Obe0ZSUhLR0dHFCk7laqje008/Tbdu3bj//vsBaNOmDVWqVKF79+488cQTxMbG5rtPQEAAAQEB+fb7+fl51ZvQz8+P6GgHmzZBYqIvXlRaheRtv//KQG3ueWpzz1J7e57a3PPU5p6nNnevM2nbcjUFQUpKCs5TZk3w8fEBwEs6zkpFi+CKiIiIiHgnW4NTcnIyCQkJJCQkALB582YSEhLYtm0bAA899BDXXXddzvHx8fF89dVXvP7662zatIlFixZx55130qlTJ2rWrGnHSyhT2TPraRFcERERERHvYutQvWXLlnH++efn3B49ejQAw4YNY8qUKezevTsnRAEMHz6co0eP8sorr3DvvfcSERHBBRdcUCGmIwctgisiIiIi4q1sDU49e/YsdIjdlClT8u274447uOOOO9xYlX00VE9ERERExDuVq3OcKjoN1RMRERER8U4KTl5EQ/VERERERLyTgpMX0VA9ERERERHvpODkRTRUT0RERETEOyk4eZHs4HT0KKSl2VuLiIiIiIjkUnDyIhERcGI9Xw4etLUUERERERE5iYKTF3E6ISrKXNdwPRERERER76Hg5GU0s56IiIiIiPdRcPIymllPRERERMT7KDh5Gc2sJyIiIiLifRScvIyG6omIiIiIeB8FJy+joXoiIiIiIt5HwcnLaKieiIiIiIj3UXDyMhqqJyIiIiLifRScvIyG6omIiIiIeB8FJy+jHicREREREe+j4ORlTj7HybLsrUVERERERAwFJy+THZwyMuDoUXtrERERERERQ8HJywQHmw00XE9ERERExFsoOHkhTUkuIiIiIuJdFJy8kGbWExERERHxLgpOXkgz64mIiIiIeBcFJy+koXoiIiIiIt5FwckLaaieiIiIiIh3UXDyQhqqJyIiIiLiXRScvJCG6omIiIiIeBcFJy+koXoiIiIiIt5FwckLaaieiIiIiIh3UXDyQhqqJyIiIiLiXRScvIFl5bmZPVTv8GHIzLShHhERERERyUPByU6bP8T3x3a0yPgwz+6qVcHhMNcPHbKhLhERERERyUPByU6uTBxJa6ia9U+e3b6+JjyBhuuJiIiIiHgDBSc7RXcBIMK1AVwZeX6kmfVERERERLyHgpOdwppg+UXgSzocWZXnR5pZT0RERETEeyg42cnhxIrqDIDz4JI8P9LMeiIiIiIi3kPByWZWZCcAHAd/z7NfQ/VERERERLyHgpPNrKhzgPzBSUP1RERERES8h4KTzayoTlg4cBzbBKn7cvZrqJ6IiIiIiPdQcLKbXzhHHbXN9QO55zlpqJ6IiIiIiPdQcPICh32amisHFufs01A9ERERERHvoeDkBQ45s4NTbo+ThuqJiIiIiHgPBScvkNPjdHApuDIBDdUTEREREfEmCk5e4KijNpZvGGSlQKJZCDe7x+n4cUhJsbE4ERERERFRcPIKDidWlFnPiRML4YaEgL+/2aXheiIiIiIi9lJw8hJWVGdzZb+ZIMLh0HA9ERERERFvoeDkJazIE8HpYP4JIhScRERERETspeDkJXKG6h1dD6kmKWlmPRERERER76Dg5C38IyEse3a93wEN1RMRERER8RYKTt4kuou5PLEQrobqiYiIiIh4BwUnb3Ka4KSheiIiIiIi9lJw8iZR55jLg0vBlaWheiIiIiIiXkLByZuEtwTfEMhMhiOrNVRPRERERMRLKDh5E6cPnLQQrobqiYiIiIh4BwUnb3PSeU4aqiciIiIi4h0UnLxN9InznA7k9jgdPAgul30liYiIiIhUdgpO3iZ7goikf4gKPQSY0HT4sI01iYiIiIhUcgpO3iYwGkIbA+Cf9Dvh4Wa3huuJiIiIiNhHwckbReUfrqfgJCIiIiJiHwUnb1Qtd4IIzawnIiIiImI/BSdvlLMQ7u9Ur2ZmhVCPk4iIiIiIfRScvFFEa/AJhowkWsetBRScRERERETspODkjZy+EHU2AG1rLQE0VE9ERERExE4KTt7qxEK4zaIXA+pxEhERERGxk63BaeHChcTHx1OzZk0cDgczZswo8j5paWk88sgjxMXFERAQQL169XjvvffcX6ynnQhOdYIUnERERERE7OZr55MfO3aMtm3bcsMNNzBkyJBi3efyyy9n7969vPvuuzRq1Ijdu3fjcrncXKkNos0EEVWdawgPTuTAgQh76xERERERqcRsDU4DBgxgwIABxT7+hx9+YMGCBWzatInIyEgA6tWr56bqbBZYHUIaQPImOjVcyob9fe2uSERERESk0rI1OJ2pb775ho4dOzJx4kQ+/PBDqlSpwsUXX8zjjz9OUFBQgfdJS0sjLS0t53ZSUhIAGRkZZGRkeKTuwmTXUFAtPpGdcCZv4pxGS1gytw8ZGZmeLq/CKay9xT3U5p6nNvcstbfnqc09T23ueWpzzziT9i1XwWnTpk38+uuvBAYGMn36dA4cOMDtt9/OwYMHmTx5coH3efrpp5kwYUK+/bNnzyY4ONjdJRfbnDlz8u2rnxFKG6BL48U8Pt3B11//gJ9fBRyWaIOC2lvcS23ueWpzz1J7e57a3PPU5p6nNnevlJSUYh/rsCzLcmMtxeZwOJg+fTqDBg067TF9+/bll19+Yc+ePYSHhwPw1Vdfcdlll3Hs2LECe50K6nGqU6cOBw4cICwsrMxfx5nKyMhgzpw59OnTBz8/vzw/cxxaju+8LhxKrkr0rQfYtCmLWrVsKrSCKKy9xT3U5p6nNvcstbfnqc09T23ueWpzz0hKSiI6OpojR44UmQ3KVY9TbGwstWrVyglNAM2bN8eyLHbs2EHjxo3z3ScgIICAgIB8+/38/LzqTVhgPdU6gE8QkSGHaRLzL0eONKOintLlad72+68M1Oaepzb3LLW356nNPU9t7nlqc/c6k7YtV+s4devWjV27dpGcnJyz799//8XpdFK7dm0bK3MTpx9EdgTMcD1NSS4iIiIiYg9bg1NycjIJCQkkJCQAsHnzZhISEti2bRsADz30ENddd13O8UOHDiUqKorrr7+eNWvWsHDhQu6//35uuOGG004OUe6dmJb8nEZL2L/f5lpERERERCopW4PTsmXLaN++Pe3btwdg9OjRtG/fnrFjxwKwe/funBAFEBISwpw5c0hMTKRjx45cffXVxMfH89JLL9lSv0ecWAhXPU4iIiIiIvax9Rynnj17UtjcFFOmTMm3r1mzZpVrdpETPU6tav/NdxuPAqH21iMiIiIiUgmVq3OcKqWgWA5nxOF0WoSkLbW7GhERERGRSknBqRzY7zK9TtV9lthciYiIiIhI5aTgVA4cDTDnOdWrstjmSkREREREKicFp3IgM8IEp6bRS8A71isWEREREalUFJzKgYCYdqSmB1A1+CAc3WB3OSIiIiIilY6CUzkQVc2f5Vs6AGAd0HA9ERERERFPU3AqB6KjYckGM0FE+i5NECEiIiIi4mkKTuVAUBCs2GbOc0I9TiIiIiIiHqfgVE5sSDQ9Tv4pf0FGss3ViIiIiIhULgpO5URWQG22H6yNAxccWmZ3OSIiIiIilYqCUzkRHQ2L12u4noiIiIiIHRScyomTJ4jggCaIEBERERHxJAWncqJatVN6nLQQroiIiIiIxyg4lRPR0bBia3syXX6Qth+Obba7JBERERGRSkPBqZyIjoa0jEA2Hj7L7Niv85xERERERDxFwamcqFbNXCZszz7PScFJRERERMRTFJzKiehoc5lzntNBTRAhIiIiIuIpCk7lRHZwmvfXieB0eCVkpthXkIiIiIhIJaLgVE5kB6e/N9XBCowFKxMOLbe3KBERERGRSkLBqZyIjASHA8BBWqgWwhURERER8SQFp3LCx8eEJ4BEXy2EKyIiIiLiSQpO5Uj2cL3d6VoIV0RERETEkxScypHsKck3J3UAhy+k7oFjW+0tSkRERESkElBwKkeye5z2HQyCqu3MDQ3XExERERFxOwWnciQ7OO3fD0RrgggREREREU9RcCpHsofqHTgARJ+YIEIL4YqIiIiIuJ2CUzmS3eNkglP2QrgrICvVtppERERERCoDBadyJM9QvSr1ILA6uDLg0J92liUiIiIiUuEpOJUjeYbqORw6z0lERERExEMUnMqRPEP1AKKyF8JVcBIRERERcScFp3Lk5KF6lsVJPU6aIEJERERExJ0UnMqR7KF6qamQkgJEdQSHDxzfCce221qbiIiIiEhFpuBUjlSpAgEB5vqBA4BvFYhoY3ZoWnIREREREbdRcCpHHI5TZtaD3OF6+3Wek4iIiIiIuyg4lTN5ZtYDLYQrIiIiIuIBCk7lTL6Z9bJ7nA4th6w0W2oSEREREanoFJzKmXxD9UIaQkA0uNLh8Arb6hIRERERqcgUnMqZfEP1HI6T1nPScD0REREREXdQcCpn8g3VA6iWvZ6TJogQEREREXEHBadypsDgpB4nERERERG38rW7ADkz2UP1cs5xAog6GxxOSNkG8y+EiLZQta1Z4ym0MTj1axYRERERKQ19oi5nCuxx8guFGr1gzxzYNdNs2XwCIbxl3jAV0QYCIj1at4iIiIhIeabgVM4UGJwAzvsODi6FxL8gcSUc/guOrILMY2aq8kPL8x4fXNuEqYg26p0SERERESmCPiWXM9lD9Q4eBJcLnNlnqfn4Q/VzzZbNckHyJji88qRAtRKObYGUHWbb9X3u8Sf3TtW7GmIu8NTLEhERERHxagpO5UxUlLl0ueDw4dzbBXI4IbSR2epemrs//Qgkriq8d2rTe1BnCLSfBCH13PmS3C/tECStg6PrcB5eTau0fyD9HPCrYXdlIiIiIlJOKDiVM35+EB4OR46Y4XqFBqfT8Q8vvHdqzxzY+A5s/8qcL9X8/6DFA+AbXGavo8y5MuDoRji6zoSkE0GJpHWQljuu0QdoCLiWXAPn/wBOH9tKFhEREZHyQ8GpHKpWzQSn/fuhadMyetBTe6eajILld8Len+Hvx2DTZDhrEtS5zCy6awfLgrT9+YNR0joT+qzM0983uDaENiWrSkOsTe/ju3cu/PUItPuv5+oXERERkXJLwakcio6GDRsKmCCiLEW0ggvmmV6nP0ebqc5/vRyq94SOL0FEazc++UkyjsK2z2DzR3A4ATIST3+sbxUIbQJhTSG0qbkMa2r2+YUA4MrIIGFnGGenPQdrnoHIjlD3Mo+8FBEREREpvxScyqHTzqxX1hwO0/tUcwCsfRbW/Bf2zYdZ7aDx7dB6gnumNbcs2P+rOc9q62eQlXJyUVCl7olg1Cw3HIU1haBaxeoN2+V7LllxLnz+fR6WDIew5hDRsuxfh4iIiIhUGApO5VCBi+C6k28wtB4HDYbDn/fC9i/h31dg6zRo+xQ0uLFszhVK2QWbPzCB6ej63P1hTaHBDRDb30yZ7htU6qdytX4CnyMrYe88+GUw9FsK/hGlflwRERERqZicRR8i3sZjPU6nqhIH3b+AC+aaacvTDsLSW+DHs2H/opI9ZlY6bJ8O8+Ph6zqw8iETmnyrmLDUZxFcuBZa/B9UbVMmoQkw61V1+wSC65rn++1aM0GGiFQOGUfhxy7wy6VmchkREZEiKDiVQ7YFp2wxvWDACujwP/ALh8MrYM65sOhqSNlZvMdIXG16r2bUhl+GwK7vTHCpdi50fg8G74Fz3oVqXd03GUVgNPT4CpwB5vlXPeae5/EkV6Y+BIoUx7qX4OCS3PM4RUREiqDgVA55fKheQZx+0PROiF8PDW8GHLB1KnzXFFY/DVlp+e+TfgQ2vAU/ngMzW8E/z5tZ8gJjzHTnF/0DfX6BhtfnTObgdpEdoNOb5vrfE2DHt5553rKWsgP+vA++iIRZZ5k1uUSkYOlH4J9Jubf/fQU2TravHhERKRcUnMoh23ucThZYDTq/Bf3/gOgu5gP7yofh+5aw8zsz0cPeBbB4GEyPNUP7Dv4ODl+oPQjO+xYGbTfTgoeV1dzqZ6jBMGg80lxffA0k/WtPHSWR+DcsHg5f1zcfBDOPwpG/4e8n7K5MxHutewnSD5uJYVqNM/v+uBUO/G5vXSIi4tU0OUQ55FXBKVtkB3M+0paPIeH/IHkjLIiHgOg8C9AS1hwa3gj1roGgGvbVe6qznofEBHOu1i+Doe8S8Au1u6qCWRbs/wXWTIRd3+fur94Tqp9nes7WPmfaWLMFiuSVnpjb29RqLMRdDokrYccMM2y4/3IIirGzQhER8VLqcSqHvGKoXkEcDqh/DVy0zgy9c/qZ0OQbaobz9V0MF66G5vd6V2gC8PGHc7+AoFg4sgaW3GACijdxZZnzMWZ3gbnnnQhNDrMocd/foffP0GY81LrYLAa87HbveA3HthKVtcpMwpGZUvTxUjGlHSp4CK+n/fMiZByB8BZQ9z9m8e8uH5jbx3fBr5eZSWtEREROoR6ncii7xyk5GVJTITDQ3nry8Qs1Q+8a3QpJ/0D17maWPG8XFAPnfgnzzoPtX8DaiSYA2i0r1UzTvva53GnanQHQ4HoTQkMb5T2+40uwZy7sW2ju12CY52vOlrwZ3zkdOTfjCPwwxuzzrwrBtc26WydfBteG4BPX/SLcNymIeJZlmXOI/rwHIs825zE6bfqvJ/0wrHvBXG81LncZBb9Q6D4jd4bQ5XdBp9ftqVFERLyWglM5FB4Ovr6QmQkHD0KtWnZXdBoh9cxWnlTrAh1egj9uM+dqVT0LYvvYU0vaIVj/Ovz7EqTuM/v8q5rzsZqMOn2vXZU4s+5WwgOw4j6odREERHmu7myuLFh8LY6MI6RTBT8fcGQdMx9e0w9D4qrT39cnODdEnRysqsRBTB/wCfDc65CSy0qHZSNh4zvm9sElsP41M7GMHf55ATKSILwV1L0s78/CGkPXqbDgItjwBkSeBY1utqdOERHxSgpO5ZDDYXqd9uwxw/W8NjiVV41ugYN/mIV4F10J/ZdBSH3PPf+xbeYD3sa3c2fHC64LzUab88OKM+Ngs3tMb9OR1ZDwkJnAw9PWPgP7F2H5hrLA7zl6XjgcP46bGQCP7zSXea7vhOM7zPpgWSmmd+3khZCzRXWGnt/bEwal+FL3mzWS9v8COKDmhWba/7/GQN0rPD9cN+2QGaYH0Hq8GaJ3qloDoe0TsPIRE/jCW5olEURERFBwKreyg5NXTRBRUTgccParpkfk0B/mhPE+i8A32L3Pe/gvWPssbJ0GVpbZF9HWLP5b9z/mnLHicvrB2a/D3B4mgDUY7tkPgAeXwV9mtrKs9i+QsibatKtfOPiHFz5pReZxc67JqcHq+E7Y85OZlXFOdzj/R6hSx0MvSM7I4ZWw8BI4thX8wqDrNIjtB7PPgUPLzAQyXd73bE3Zs05GtIE6g09/XIuH4NAKM1z3l0vNZBHBNT1Xp4iIeC1NDlFOeeXMehWJTyB0/xICqsHhBDONujsmWnBlwc6Z8PMAmNUWtnxkQlONXiYYDFgB9YaeWWjKVr27OQ8KzFTLnloYN/MY/Ha1maCi7n+w4q49s/v7BkFoQ6hxHtS/2gTHji+Z30ffRWbIXtJamNMVjqx1z2uQkts+HeZ0M6EppJGZobLWQHM+UcdXAYfpDd2/yHM1pR4wU5DD6XubsjkccM5kM5wvdY8JT94wqYWIiNhOwamc8tqZ9SqSKnXg3M/A4WMCzb8vl91jH9sGf42Hb+rDggth9w/mw1zdK8zQwF5zIbZv6SdIaDcR/CNN71n2B0d3+/M+OPovBNWEs98o20kewluY3r+wpqYXas65WnvHW1gWrHrc9NBmHoOY3tB/KYQ3zz0mupMZbgrwx0hwZXqmtn8mQWYyVG1v1o8ril8I9Jhhzik8uMQM2/OGGSpFRMRWCk7llHqcPKRGT2j/rLn+52gzU11JuTLMdOI/D4Sv65n1llK2m2DT9G6IXw/nfmLWxCorgdHQfqK5vmocHNtedo9dkJ3fmRPrwQzFCogs++eoUhd6/2pmaEs/BPMugF0/lv3zSPFlHoNFV8CqseZ2kzuh5ywTPE7V9mmzP3ElrH/D/bWl7s/90qP1+OIH+dCGZoihwwkb3819X4uISKVla3BauHAh8fHx1KxZE4fDwYwZM4p930WLFuHr60u7du3cVp83U3DyoKZ3Q9xQM4Tu1/+Yno4zcXQDJDwIM+qYYT+7ZwEW1LjAfDAbvAs6vAAhDdxRvRmuV62b+XC7/C73PAfA8b1m/SuApveYHgd3CYyGXj+ZGfayUsxMaFumue/55PSObTPnnG373Awp7fQ2dPzf6accD4yGtk+Z6389at437rT2WfPej+wAteLP7L41+5mgB7DsTtj3S9nXJyIi5YatwenYsWO0bduWV1999Yzul5iYyHXXXUevXr3cVJn301A9D3I4oPPb5qTy1H3wy2VFn/OQlWo+yM+7AL5tDGuegdS9EFgDWjxoepd6zYN6V7p/am2H00wU4fCBHdNNr1BZsyz4/UZI2w8RraHdU2X/HKfyC4HzvjPDG61M+G2o54YjirH/N7P20eEV5nzAC+ZBo5uKvl/Dm81U/xlHYOWD7qsvdR/8e+L/lzPpbTpZ8/tz32O/XnbmX5yIiEiFYeusegMGDGDAgAFnfL9bb72VoUOH4uPjU2QvVVpaGmlpuR9yk5KSAMjIyCAjw0Mnyxciu4YzrSUiwgH4sn+/i4yMLDdUVjGVtL3BD7p8hu/cLjgO/o5r6UiyOhawQGbSGpyb3sO59SMc6YcAsHBgxfTFVf8GrJoX5U704Mn3X5VmOJvchc+657H+uIPMyO5lOkugc+Nb+Oz6HsvpT2anKeDyyZmMouRtXhwO6PQ+Tv8ofDa8BsvvIitlD66W4yv1ArrubXPDseUDfJbfjsOVjhXehsxuX5p1tor5nI72/8P3p+6waQqZ9W7AijqnzGt0/v1ffLJScFXtSFa1viX/m+vwBr5H1uA4sgrXgsFknf+TmUDmBE+0t+SlNvc8tbnnqc0940za12FZ3nHGq8PhYPr06QwaNKjQ4yZPnszrr7/Ob7/9xhNPPMGMGTNISEg47fHjx49nwoQJ+fZPnTqV4GA3Ty/tRgkJ1Rg/vitxcUf43//m211OpVE980/OSXscBxYJ/rez1a8vPlYqNTMXEZc5hyjXPznHHndEsdW3N9t8e3HcWd3Gqg0f6zgXHL+DYOsA//pdylr/M5zt7jRCXDs57/g9+JLOKv8b2OR3cZk87hmxLJpkfEbzDDNcb4tvX1b632J62aRMOawsWqS/T6PMbwDY5XMOfwbcRZYj6Iwfq13ay8RlziPRWZ8Fgc+V6e8rwHWY3sdvwZd0FgeMYZ9v6c4dDHbt5bzj9+HPUbb5ns8K/zsrdTgXEakoUlJSGDp0KEeOHCEsLKzQY8tVcFq/fj3nnnsuv/zyC02aNGH8+PFFBqeCepzq1KnDgQMHimwcT8jIyGDOnDn06dMHP7/iTzmdkACdOvkRE2OxbZuHZqaqAEra3idzrv0vPn+PxXL6Y9W9EseOGTgyTU+m5fDBih2Iq8GNWDH9vO6Du2PnN/j+dhmWw5fMvssgrEXpHtCVjs9PPXAe/hNX9V5k9fg+31TPZdHmxeXc+BbOP+/AgYWr1iCyOn+Qp2egsnBbm6cn4rPkGpx7ZwOQ1eJRXC0eLXx678Kk7cd3VkscGYlktf8frka3lVmpzoT78Vn/P1yRnci64JcyCTmOvT/hs3AgDlxktXsBV+ORgGff42KozT1Pbe55anPPSEpKIjo6uljBqdwsgJuVlcXQoUOZMGECTZo0Kfb9AgICCAjIfw6Jn5+fV70Jz7Se2FhzeeCAA19fP33xeYZK9ftv/SgkrsCxYzqOLR+YfSENoOFNOBoMxxEU673TVda7FLbG49j5LX4r7oRe80v3gXLleDj8J/hXxdn1fZz+pz9fyyN/c81GQnAN+O1qnDtn4Fx0iZlW2s/+L0nsUKZtnrQOFlxsppr3CYYu7+NT9zJK9dWAX01o+yQsG4nP3+PwqX8VBFYrfa3Hd8OmNwFwtnkMp79/6R8ToHY/M8vminvxWXkfPlHtzMybJ3jb/yuVgdrc89Tmnqc2d68zaVuv/Xx3qqNHj7Js2TJGjRqFr68vvr6+PPbYY6xcuRJfX19++uknu0v0qOxZ9TIz4cRpW+IpDoeZarvufyDuKrhgrpnsoeVDEBRrd3VF6/iy+eC7b6FZiLSk9v0Cq0/MONbpLQiuVTb1lVbdy8xU2L4hsPdnmNvT/TO3VXS7foQfO5vQFFzHLERc97KyeexGt5j1lTISzeyTZWH1f80ELdFdzHpoZanZPVDv6txZNo9tK9vHFxERr1VuglNYWBirVq0iISEhZ7v11ltp2rQpCQkJdO7c2e4SPSowEEJCzHXNrGcDv1CzOG63qRDTq+RDlexQJQ5an1hvZ8V9kHbozB8j/QgsvhawoMHwsvsQXVZiLoDe881Mb4dXmIVykzfbXVX5Y1nwzwuwYKCZAa9aN+j3B1RtV3bP4fSBjidmvtv0HuxfXLrHS9kJG0xvE20eK/vzkBwO80VB1faQdgAWDoas42X7HCKllZ4IO2dCwsOw5EbY8XXOhD0iUnK2DtVLTk5mw4YNObc3b95MQkICkZGR1K1bl4ceeoidO3fywQcf4HQ6adWqVZ77V69encDAwHz7K4voaEhONms5NWpkdzVSrjQbDZs/hCOrzbf8nd86s/svuwOObYUq9aHD/9xTY2lFdoA+i+DnvpC8AWZ3hfN/hKpt7K7M+7kyYMc3sP512DvP7GtwA5z9mnumz6/Wxaw3tmkyLBsF/ZaaQFUSa/4LrjSodi7UcNOSFb7B0GM6/NARDv+Jz7LbwPqPe55LpDhSdppRAPt/hf2/QOIq4KRT2De9Z75IqneN+bJL/w6KlIitX5MvW7aM9u3b0759ewBGjx5N+/btGTvWfBu+e/dutm3TMIjT0SK4UmJOP7O2E8DGt8/sW/6tn8KWD00vW9cPvfv8obDGJjyFt4LUPTC3hxYxLUzyJkh4yCzW/OtlJjQ5fOCsF6HzO+5dc6zdf8Evwpwzt/EMg3y2lB2w4cR9W09w76x3VeLg3M/B4YNz21QaZH7rvucSOZllwZG15r3+23XwdX2YURt+uwrWvwqJfwEWhDY2X3g0vRsCY8w6e+tegFltYdZZsO5lSNUHCJEzYWuPU8+ePSlsUr8pU6YUev/x48czfvz4si2qHNEiuFIq1bvnfsv/x63Qfzk4i/gnIWUHLL3VXG/xsBm65e2Ca0KfhbAgHvaf6IHq9hnUjre7Mu+QlQ47v4YNb8OeObn7A2uYD12NbjKTn7hbYHVo8zgsvwNWPgJ1LjvziSJWPwWudKjeA2qc7546T1ajJ5z1PCy/i1bpU7AW7oCIlhDeAsJbQnhz8I9wfx1gFuU+uh6S1poP1UlrIWU7hDWH6K7mbzW0saZQL49cGXDoz9zepP2/QtrBvMc4nGb4aLXupre12rkQVCP35+2fhd0/wqYp5u/98ApYvgJW3Au14s3/BbH9i/4/QPJyZUHqbjMC49hWc87jsa1mwezA6hBQ3VxmbwHVISBK7VyOleg3t337dhwOB7Vr1wZg6dKlTJ06lRYtWjBixIgyLVBOTz1OUmrtJpqx74l/wbqXoPno0x9ruWDxMHMSf+TZuedJlQf+VeH82fDrFbDrO/hlMHR60wSDyvpB8ugGE5Y2TTbfRAPggNh+0Ohm82HK6eFZnBrfChvfgcSVsPIh08tVXMe2mfsCtHbDuU2n0+QOXIcScG6ejGPvbDgxVXuOoJomSIW1OBGoTmwBUSV7voyjucHo5JCUvMlMWHGq/Yty2yWgGlTrmhukIjtUyun6vV5GEhz4PTcoHViS/zw6nyCIPudESOpurvuFnv4xnb5Q60KzpR2ELdNg8xQ4tBy2f2W2wBonhvJdb74AEMhMMV9AnBqMUrIvd5iQdEYc5u//1GBVUMjyrWp6GMVrlCg4DR06lBEjRnDttdeyZ88e+vTpQ8uWLfn444/Zs2dPzlA7cS8FJym1wGhoPxF+vwlWjTUzBVapU/Cx/7wAe38yM/J1/cjzH6pLyzcYenwFv98Mm983r3nzh9DuGYiuJJPLZKXDjhlmiE/2uUtgZoNscAM0vAlC6tlVnflwd/arZjKPje9Cw5uL/7tZ/ZT5Zr7G+VDjPPfWeTKHg6wOb/Dr7mZ0bRmGb/I6OLIGktaYD1XHd5ltz9y89wusbnql8gWqE71saftzQ9HJQSllx+lr8QszPUzhzc1lUC3zpciBRXBwmXnMHV+bDcDpb8JTtW4Q3c2EqkD7F+uuVCzL9BQeWHxi+w0S/ybP+UkA/pEmJFXvboJS1fbgU8Jp9gOioOkosx3+y/x7uOUjSN0L/0wyW2RHcy5U3FUQEFnaV1l2LMvUmbwpd8tMPjFBk+PEpfPEFyfOvPsL3HfSdVem+ftKOSkgpRVjSI/DF4Jrm+G7wXXNpdPP3Dd1n9nSsi8PApaZWCbtALCm0If2Ay4kEJ8f6poZTYNrmecKrg1BJy6Da0FAdPmapKocK1Fw+vvvv+nUqRMAn332Ga1atWLRokXMnj2bW2+9VcHJQzRUT8pEg+th43vmP+w/74buX+Y/5vBKWPmwud7hBQgr/lpqXsXpB+dMNkPPVj8F+xbA7HOgzhBo8ySEN7O7QvdI+tf0Omyakrd3qeYAE05qXeg9QbhaN6g/zHyYWzYS+v5e9EQRyVvMye9gzm3yNIeDwz5NsRoMhJPXA0k/ciL4rMndktaYD2TZH6j2/pz3sQKizIfD9EJmuwyMyQ1HeYJSbAE9bVebi6w0M9zrwCLTC3XgN/P82R/Yec4cF9LI/A6qdTVhKry5PpCVpcxjcPAP0+b7f4ODS058gD5Flfrm91D9xNC7sGbu+T1UbQNVJ5lzDHfNMj3QO7+DQ8vM9udoqH0J1B8O0R4Y/gqQeRyObc4bjnK2zZCV4pk6svmG5A1FVeKgyknXA2OLP5mNK9OEp+wgdWqwOvV2ZjK+pJqlII7+e/rHdfqbL0qCTwpTOcHqxOYXZv4dcJ3Yskp46cowQ5ADa5yyVXfvebBeokTBKSMjI2dR2blz53LxxRcD0KxZM3bv3l121Umh1OMkZcLhhE5vwKz2ZrjGzu/NB+lsWanw29Xm3JFaF5sP2uWZw2GGGTa4HlaNMx/Qt39lvoVvcAO0Huc9a1KVgtPKwLHtU9jyXt4P50E1oeGNZqsSZ1+BhWn3jOkZO7TcBL7GtxR+/OonT/Q29TIfNL2Ff7gZQhV9Tt79GcmQ9E9ukMoOVcmbTjp3xQFV6uWGopMv/aueeS0+AWb2wmpdoPl9JpwlbzQf3rPD1JHVZgbK5A3m7wLMhB3RXUxvmE8gOAPMYzkDTI+HM+CUfQHmQ1y+fSf2+wSYIYWuTDPEyZVx4jITrAxzmb0v++f5js3AkZlKTOa/cLwd+Hnp+9iyTEg+8FtuUEpcmX9IpTMAojqaIZTRXcwWFOPZWp1+UPtis6Xuhy0fmxCV+Bds+xy2fY5vYCydMurgs2Qq+AWb4YI+QeZ9kXP91NsFXPcNym2bk0PRsROXx4v4HOlwmt6XkAYmYPpHnBjO5irk0gVYeS9P/RkO8+/jyaEouK75eyurob9OX3Pu2cnnnxUi4/gRFvwwlZ6dGuObvudEj9gOOL7DzKSYssP0wLnSTdg8ZvOyG37hBQSqE6/31H2+wfbWWkIlCk4tW7bkjTfe4MILL2TOnDk8/vjjAOzatYuoqBKO25YzpuAkZSaitVnYc+1zZjroGufn/qOW8KD5QBVYw5xzUlHOCapSB855D5rda3rTdn5jZhjc8iE0vQtaPFCyD6h2cmXCgcU4t35J35T38P39qNnvcELsAGg0AmoO9P4Tk4NqmDWYlt9lznWqc6kZVlqQ5M2mJw2gjQ29TSXhF2I+KEd1zLs/8zgcXWeuhzY1HzDdxeGA0EZma3Cd2Zd+2JxPs/9EkDq41JzTuHuW2byIL9AZ4Lv/mg/R0edA1DkmdES2t+fcrYzkE0MjTwpKqXvyHxdUK/dcs+gupRt25w6B1aDZ3WY7nAAbJ8PWj3Gk7iaW3bB9qftr8AuDkIYmHJ26Bdf1rvZyJ99gjjlrYlU/L29v9smy0s0EFSknhamccLUjd8hwdmB3+Ob/UqOgLzwK+vLDJwAcfqZHPHWfCW2pe00PmSvDrPeXcaTw3rGc1xZiPld0/Sj/l0terET/ez7zzDMMHjyYZ599lmHDhtG2bVsAvvnmm5whfOJ+GqonZarVODPV+LEt5hv8tk/C7tmw7sQ6TZ3fO/NZzsqDiJZw3tfmg2LCA+ZyzTPmPKAWD0GTUe79AFta6UfMbFk7v4VdMyH9ED6AD2AF1cbR8EZoeIP5FrU8aXy7Oc8p8S8TbE+31tjfT5heiJi+5WOWx8L4BpXt4sJnyr+qGb5Zc4C57coww3QP/GZOkM8zZCf9NMN40gsZ4pOe9/mcfuZDXM6lr/lQ5vQtcr8LH47uX0+YaxuOlO2wbbvpGcl+3KrtTwSpE2GqSlzZfOljucz5L0nrTMhN+sdcT1oHx3fmP97he2K2u5OC0unOI/VGVdtBx/9B+2fJ3DmLv5f+SOsWjfAhwwT9rOwt9TTXT7mdeRxcqaYdg+sWHIxCGpRtL09F5+Of20N2Oq4s8/fn9C/5GnmFsSzzxUt2kMqz7cu/LyvVnJuWnGwCWTlSouDUs2dPDhw4QFJSElWr5n4jO2LECIKDy2fXW3mkHicpU34h0OElM+Pc2mdND8WS4eZnjW+HWgNtLc/tqnWD3r+Y8f0rHzRDpxL+D/59yczSVv869/yHUxLJm2DHtyYs7VuQd1Yn/6q4YvqzdG99Ogx8FL8ALw59hXH6QsdXYW53M1yv0c0QdXbeY45uzB1SVl56m8oTp1/BPWMlZVkmjDl8Sv23lJWRwfyZMxnYpzt+R1eanrIDS0xPT9p+01t2cKn5+wXzzXZOr9Q5ZvIDv5DTP0FGUm4gyg5HR9eZiRyyUk9/v+znyQ5JkR29+4uX4vLxx4odyFY/aNl4ID6n6/0oLstSMPIkpw843dyDHRBptvDmhR9rWZB5FI6fCFFhTd1XlxuUKDgdP34cy7JyQtPWrVuZPn06zZs3p1+/fmVaoJxednBKTISMjNP34ooUW+1LzDTUO7+FeeebD+RhzcwaIJWBw2HWd6o50AzZ+2us+ab99xvgn+eg7VPmPC9P/4fvyoKDv5vhhDu/NaHuZGFNze+tVjxEdyUry2LvzJnePySvKNXPhXrXmt/FH7dD3yV5P3CvfsIMP4ntX66GelRaDkfZD7HyCz0xk+KJiQssy/SaH1icG6YOrzAf0E6eUdDhhIg2JkhFdTwRlE7qPSpoiF02p5+ZRCOsmfnbC2tqhlaGNfWuGei8mUJT5eVwmKGYfmFmkfpypkT/q15yySUMGTKEW2+9lcTERDp37oyfnx8HDhzg+eef57bbbivrOqUAVauC0wkuFxw8CDEePpdUKiCHw/Q67ZlrhlU4fKHrx+X2JM4Sc/qcmIr3Svj3VTN08cgaWDjIfJPc7hnzod6dMo6aoZI7v4Vd3+edecvhY6Ykzg5Lp/7nk5Xh3to8qf1Es2DnoWWw6V1znhZA0nrY/IG5bsdMeuKdHA4IqW+2ekPNvszjJjxlh6mDS8x5H4cTzLbhNI8VGJM/GIU1NZN2lPcvJUSkREr0l//nn3/ywgsvAPDFF19Qo0YNVqxYwZdffsnYsWMVnDzExwciI81QvQMHFJykjITUM8Fg+Z2mpynyLLsrso9PIDS/18xAt2YirHvRnO8x90RoafsURLQqu+c7tvWkIXjz854T4hdhzj2pFQ81+5e/iStKKijGDJX8825IODFRREAU/P24OU+i5oUQrXNrpRC+QeYco2pdc/el7DCLzB5cYqZp94/MH5L8w+2rWUS8UomCU0pKCqGhZoXq2bNnM2TIEJxOJ+eccw5bt24t0wKlcNHRucFJpMw0vcNM113YOQCViX8EtHvKTBTx9wQzacHOb835UPWvM+cxuFJzT3zOOmXLs+944T87WUgjE5RqX2zOwfKWtZY8rclI09uUuApWPmJmgNz6sflZ6/G2liblVHBtqFsb6l5qdyUiUo6UKDg1atSIGTNmMHjwYH788UfuueceAPbt20dYWFiZFiiFq1YN/vlHM+uJGyg05RdcEzq9Cc1Gmw/w2780kxNkT1BQWg6nWXQ0ZwheU50LACcmingF5p5nZjs8vNL0NtWKL7uJC0RERIpQouA0duxYhg4dyj333MMFF1xAly5dANP71L59+zItUAqnmfVEbBDWFLp/YYb6rH/N9Bb5BJ5YIDQw93rOoo8F/eyU/b5BZvidn758KlD1HlDvGtjykRleBeptEhERjypRcLrssss499xz2b17d84aTgC9evVi8ODBZVacFE3BScRG0Z3NJp7RfqKZFS3zKNQeVLnPvxMREY8r8bQwMTExxMTEsGPHDgBq166txW9tkL0IroKTiFR4QbFwzruw7uXKM0W+iIh4DWdJ7uRyuXjssccIDw8nLi6OuLg4IiIiePzxx3G5XGVdoxQiu8dJ5ziJSKVQ9z/QZyGENrK7EhERqWRK1OP0yCOP8O677/Lf//6Xbt26AfDrr78yfvx4UlNTefLJJ8u0SDk9DdUTEREREXG/EgWn999/n3feeYeLL744Z1+bNm2oVasWt99+u4KTB2monoiIiIiI+5VoqN6hQ4do1qxZvv3NmjXj0KFDpS5Kik9D9URERERE3K9Ewalt27a88sor+fa/8sortGnTptRFSfGdPFTPsuytRURERESkoirRUL2JEydy4YUXMnfu3Jw1nBYvXsz27duZOXNmmRYohcseqpeaCikpUKWKvfWIiIiIiFREJepxOu+88/j3338ZPHgwiYmJJCYmMmTIEFavXs2HH35Y1jVKIYKDITDQXNdwPRERERER9yjxOk41a9bMNwnEypUreffdd3nrrbdKXZgUj8Nhhuvt2GGG69WrZ3dFIiIiIiIVT4l6nMS7aGY9ERERERH3UnCqADSznoiIiIiIeyk4VQBaBFdERERExL3O6BynIUOGFPrzxMTE0tQiJaSheiIiIiIi7nVGwSk8PLzIn1933XWlKkjOnIbqiYiIiIi41xkFp8mTJ7urDikFDdUTEREREXEvneNUAWionoiIiIiIeyk4VQAaqiciIiIi4l4KThWAhuqJiIiIiLiXglMFkD1U79AhyMqytxYRERERkYpIwakCiIw0ly4XHD5sby0iIiIiIhWRglMF4OcHERHmuobriYiIiIiUPQWnCiIuzlyuXGlvHSIiIiIiFZGCUwXRt6+5nDnT3jpERERERCoiBacKYuBAczlrljnXSUREREREyo6CUwXRrRuEhZm1nJYvt7saEREREZGKRcGpgvDzyx2u9/339tYiIiIiIlLRKDhVINnD9XSek4iIiIhI2VJwqkAGDDCXf/wBe/faW4uIiIiISEWi4FSBxMRAhw7m+g8/2FuLiIiIiEhFouBUwVx4obnUeU4iIiIiImVHwamCyT7PafZsyMiwtxYRERERkYpCwamC6dgRoqPhyBFYvNjuakREREREKgYFpwrGxyd3kggN1xMRERERKRsKThWQpiUXERERESlbCk4VUL9+4HTC33/Dtm12VyMiIiIiUv4pOFVAVatC167munqdRERERERKT8GpgsoerqfznERERERESk/BqYLKXs9p3jxITbW3FhERERGR8k7BqYJq3Rpq14bjx2HBArurEREREREp3xScKiiHQ8P1RERERETKioJTBXZycLIse2sRERERESnPFJwqsF69wN8fNm2Cf/+1uxoRERERkfJLwakCCwmB884z1zUtuYiIiIhIySk4VXDZw/UUnERERERESk7BqYLLnpZ8wQI4etTeWkREREREyisFpwqucWNo1AgyMsyaTiIiIiIicuYUnCoBTUsuIiIiIlI6Ck6VQPZwvZkzNS25iIiIiEhJ2BqcFi5cSHx8PDVr1sThcDBjxoxCj//qq6/o06cP1apVIywsjC5duvDjjz96pthyrEcPCA6GXbtg5Uq7qxERERERKX9sDU7Hjh2jbdu2vPrqq8U6fuHChfTp04eZM2eyfPlyzj//fOLj41mxYoWbKy3fAgOhd29zXbPriYiIiIicOV87n3zAgAEMGDCg2Me/+OKLeW4/9dRTfP3113z77be0b9++jKurWAYOhG++McHp4YftrkZEREREpHyxNTiVlsvl4ujRo0RGRp72mLS0NNLS0nJuJyUlAZCRkUFGRobbayxKdg3ursX0OPmxeLHFnj2ZREW59em8lqfaW3KpzT1Pbe5Zam/PU5t7ntrc89TmnnEm7euwLO+YLsDhcDB9+nQGDRpU7PtMnDiR//73v/zzzz9Ur169wGPGjx/PhAkT8u2fOnUqwcHBJS23XLrrrp5s3RrO6NHL6NFjp93liIiIiIjYKiUlhaFDh3LkyBHCwsIKPbbc9jhNnTqVCRMm8PXXX582NAE89NBDjB49Oud2UlISderUoW/fvkU2jidkZGQwZ84c+vTpg5+fn1uf69dfnTz3HOzefRYDB7Z163N5K0+2txhqc89Tm3uW2tvz1Oaepzb3PLW5Z2SPRiuOchmcPvnkE2666SY+//xzemfPenAaAQEBBAQE5Nvv5+fnVW9CT9Rz8cXw3HMwe7YTp9OJj49bn86redvvvzJQm3ue2tyz1N6epzb3PLW556nN3etM2rbcreM0bdo0rr/+eqZNm8aF2QsUSbF06QIREXDwICxdanc1IiIiIiLlh63BKTk5mYSEBBISEgDYvHkzCQkJbNu2DTDD7K677rqc46dOncp1113HpEmT6Ny5M3v27GHPnj0cOXLEjvLLHV9f6NfPXNe05CIiIiIixWdrcFq2bBnt27fPmUp89OjRtG/fnrFjxwKwe/funBAF8NZbb5GZmcnIkSOJjY3N2e666y5b6i+PBg40lwpOIiIiIiLFZ+s5Tj179qSwSf2mTJmS5/b8+fPdW1Al0L8/OBzw55+wezfExtpdkYiIiIiI9yt35zhJ6VSvDmefba7PmmVvLSIiIiIi5YWCUyWUPVzv++/trUNEREREpLxQcKqEsicjnDMH0tPtrUVEREREpDxQcKqEzjrLDNk7ehQWLbK7GhERERER76fgVAk5nTBggLmu4XoiIiIiIkVTcKqksofraVpyEREREZGiKThVUn36gI8PrF0LmzfbXY2IiIiIiHdTcKqkIiLg3HPNdfU6iYiIiIgUTsGpEtO05CIiIiIixaPgVIllB6eff4aUFHtrERERERHxZgpOlVjLllC3LqSmwvz5dlcjIiIiIuK9FJwqMYdDw/VERERERIpDwamSO3lacsuytxYREREREW+l4FTJnX8+BATAli1manIREREREclPwamSq1IFevY01zUtuYiIiIhIwRScJGe4ns5zEhEREREpmIKT5EwQ8euvcOSIvbWIiIiIiHgjBSehYUNo2hQyM2HuXLurERERERHxPgpOAmhachERERGRwig4CZAbnGbNApfL3lpERERERLyNgpMA0L07hITAnj2wYoXd1YiIiIiIeBcFJwHMWk69e5vrmpZcRERERCQvBSfJoWnJRUREREQKpuAkOQYMMJdLl8L+/fbWIiIiIiLiTRScJEetWtC2LVgW/Pij3dWIiIiIiHgPBSfJI3u43rvvQkaGvbWIiIiIiHgLBSfJY+hQ8PeH+fPhqqsUnkREREREQMFJTtGyJUyfbsLTl18qPImIiIiIgIKTFGDgwLzhaehQhScRERERqdwUnKRAAwfCV1+Z8PTFF3D11QpPIiIiIlJ5KTjJaV14YW54+vxzhScRERERqbwUnKRQF15ohuv5+eWGp8xMu6sSEREREfEsBScp0kUXmZ6n7PA0dKjCk4iIiIhULgpOUiynhif1PImIiIhIZaLgJMV20UW5w/Y++wyuuUbhSUREREQqBwUnOSPx8bnh6dNPFZ5EREREpHJQcJIzFh9vpijPDk/XXqvwJCIiIiIVm4KTlMjFF5tznfz84JNPFJ5EREREpGJTcJISu+SSvOHpuusUnkRERESkYlJwklLJDk++vjBtGgwbpvAkIiIiIhWPgpOU2iWXmHOefH1h6lSFJxERERGpeBScpEyc3POUHZ6ysuyuSkRERESkbCg4SZkZNMis75Qdnq67TuFJRERERCoGBScpU4MH5w1P6nkSERERkYpAwUnK3ODBZn0nX1/4+GOYONHuikRERERESkfBSdxiyBB4801zfexYWL7c3npEREREREpDwUnc5vrr4bLLzAx7V18NKSl2VyQiIiIiUjIKTuI2DofpdapVC9atg3vvtbsiEREREZGSUXASt4qMhPffN9ffeAO+/dbeekRERERESkLBSdyuV6/c3qYbb4S9e+2tR0RERETkTCk4iUc8+SS0aQP798MNN4Bl2V2RiIiIiEjxKTiJRwQEmHWdAgJg5kx47TW7KxIRERERKT4FJ/GYli1z13S67z5Ys8beekREREREikvBSTzqjjugXz9ITTVTlKen212RiIiIiEjRFJzEoxwOmDwZoqMhIQHGjLG7IhERERGRoik4icfFxsI775jrzz4L8+fbWo6IiIiISJEUnMQWl1wCN99sZte79lo4fNjuikRERERETk/BSWzz/PPQqBHs2AG33aYpykVERETEeyk4iW1CQuDjj8HHBz79FD76yO6KREREREQKpuAkturUCcaPN9dHjoTNm20tR0RERESkQApOYruHHoJu3eDoUXO+U2am3RWJiIiIiOSl4CS28/GBDz+E0FBYtAieecbuikRERERE8lJwEq9Qvz68+qq5Pn48/PGHreWIiIiIiORha3BauHAh8fHx1KxZE4fDwYwZM4q8z/z58znrrLMICAigUaNGTJkyxe11imdccw1cfrkZqnf11ZCcbHdFIiIiIiKGrcHp2LFjtG3bllezuxqKsHnzZi688ELOP/98EhISuPvuu7npppv48ccf3VypeILDAW+8AbVrw/r1MHq03RWJiIiIiBi+dj75gAEDGDBgQLGPf+ONN6hfvz6TJk0CoHnz5vz666+88MIL9OvXz11ligdVrQoffAC9esHbb8PAgTBokN1ViYiIiEhlZ2twOlOLFy+md+/eefb169ePu++++7T3SUtLIy0tLed2UlISABkZGWRkZLilzjORXYM31OItzj0X7rnHyfPP+3DTTRZnnZVJbGzZPLba2/PU5p6nNvcstbfnqc09T23ueWpzzziT9i1XwWnPnj3UqFEjz74aNWqQlJTE8ePHCQoKynefp59+mgkTJuTbP3v2bIKDg91W65maM2eO3SV4lS5dnNSv353NmyMYNOgQY8YswVmGA0vV3p6nNvc8tblnqb09T23ueWpzz1Obu1dKSkqxjy1XwakkHnroIUafdLJMUlISderUoW/fvoSFhdlYmZGRkcGcOXPo06cPfn5+dpfjVZo0gc6dLVasqMGWLRcxapSr1I+p9vY8tbnnqc09S+3teWpzz1Obe57a3DOyR6MVR7kKTjExMezduzfPvr179xIWFlZgbxNAQEAAAQEB+fb7+fl51ZvQ2+rxBm3awHPPwahR8NBDPvTt60PLlmXz2Gpvz1Obe57a3LPU3p6nNvc8tbnnqc3d60zatlyt49SlSxfmzZuXZ9+cOXPo0qWLTRWJu91+OwwYAGlp0L07PPWUpikXEREREc+zNTglJyeTkJBAQkICYKYbT0hIYNu2bYAZZnfdddflHH/rrbeyadMm/u///o9//vmH1157jc8++4x77rnHjvLFAxwOmDzZ9D4dPgyPPAINGsCkSXD8uN3ViYiIiEhlYWtwWrZsGe3bt6d9+/YAjB49mvbt2zN27FgAdu/enROiAOrXr8/333/PnDlzaNu2LZMmTeKdd97RVOQVXI0a8Oef8NFH0KgR7N8P990HDRvCK6+Y3igREREREXey9Rynnj17YlnWaX8+ZcqUAu+zYsUKN1Yl3sjHB66+Gq64Aj78ECZMgK1b4Y47YOJEGDMGhg8HDQEWEREREXcoV+c4ifj6wvXXw7//wuuvQ61asH07jBgBzZqZxXOzsuyuUkREREQqGgUnKZf8/eHWW2HDBnjxRTOcb9MmGDYMWraETz4BV+lnLxcRERERARScpJwLDIS77oKNG+GZZyAyEtatg6uugnbtYPp0KGQ0qIiIiIhIsSg4SYVQpQr83//B5s3w2GMQHg6rVsGQIdCxI8ycqQAlIiIiIiWn4CQVSliYmShi82YzdXlIiJmR78ILoWtX+OknhwKUiIiIiJwxBSepkKpWhSeeMAHq/vshKAiWLIH+/X155JFuvPmmk5NmuhcRERERKZSCk1Ro0dFmuvKNG83U5f7+FmvWRHPHHT7ExUHr1vDgg/DLL5CZaXe1IiIiIuKtFJykUoiNhZdegrVrM7nmmjV07erC6YS//zaTSvToAdWrm0klPvoIDhywu2IRERER8SYKTlKp1KkDl122nvnzs9i3Dz7+GIYONbPxHT5spjG/9loTorp2hSefhIQETSwhIiIiUtkpOEmlFRVlQtPHH8O+ffDrr/DQQ9CmjQlKixfDo49C+/ZQuzbcfDPMmAHJyXZXLiIiIiKepuAkAvj4QLdu8NRTsHIlbNsGb7wBF18MwcGwaxe88w4MHmwCV9++8L//mf0iIiIiUvEpOIkUoE4duOUW+PprOHgQfvjBTC7RoAGkp8OcOXD33dCwoemVOnrU7opFRERExJ0UnESKEBgI/fqZySU2bIC1a+G556BzZ0hNNedBNW5seqSysuyuVkRERETcQcFJ5Aw4HNCsGdx7rzkHavp0aNQI9u4150C1bw9z59pdpYiIiIiUNQUnkRJyOGDQIFi9Gl54wSy6u2oV9OkDF11keqZEREREpGJQcBIpJX9/c77Thg1w113g6wvff28W1x05Evbvt7tCERERESktBSeRMhIZCS++aHqgBg0y5zu99poZyvfss5CW5t7nz8iAP/4wswK6XO59LhEREZHKRsFJpIw1aWLOffr5Z3POU1IS/N//QfPm8PnnZbeY7rFjMG8ejB8PvXpBeDh06gTt2kG1anDZZfD66/Dvv1rAV0RERKS0fO0uQKSi6tkTli2DDz+Ehx+GzZvh8suha1d4/nkzK9+ZOHTILNL7yy9mW74cMjPzHhMZaaZLP3QIvvzSbGAW8L3gAhOwLrjA3BYRERGR4lNwEnEjpxOGDTO9P889BxMnwm+/wTnnwFVXwdNPQ1xcwffdudMEpIULzeXff+c/pk4d6N7dbD16mBn/srLMkL2ffjI9Ur/9Bjt2wAcfmA1Mr1h2iDr/fLOor4iIiIicnoKTiAdUqQLjxpkpyx99FKZMgWnT4KuvYPRoePBB2LMnb1DavDn/4zRrljcoFRS6nE7Tq9W1q3mu48dh0SITon76yfSC/fuv2V5/3cwO2K5dbo9U9+4QEuLuFhEREREpXxScRDyoZk147z244w6zFtTPP5tep4kT8y+e63SaQNOjhwkz554L1auf+XMGBUHv3mYDSEyEBQtye6RWr4YVK8w2aZKZFbBzZzOt+h13mOF/IiIiIpWdgpOIDdq3N6Hl22/h/vtN709AgJncIbs3qUsXCAsr++eOiIBLLjEbmJ6un3829cybB1u2mB6qRYvMZBbz5kGNGmVfh4iIiEh5ouAkYhOHAy6+GAYMgI0boV49CAz0fB0xMeZ8q6uuMrc3bzZhadw40xt13nnmdq1anq9NRERExFtoOnIRm/n5mXOX7AhNBalfH266yQznq1MH1q0z4WnbNrsrExEREbGPgpOIFKhRIzNRRf36pkfsvPMKnrBCREREpDJQcBKR06pXz4Snxo3NuU89esD69XZXJSIiIuJ5Ck4iUqjatc2wvebNzXpQPXrA2rV2VyUiIiLiWQpOIlKk2FiYPx9atzaz8J13HqxaZXdVIiIiIp6j4CQixVK9upm2/KyzYP9+6NkT/vzT7qpEREREPEPBSUSKLSrKTE3euTMcOgS9esHSpXZXJSIiIuJ+Ck4ickYiImD2bDj3XEhMhN69zWK5IiIiIhWZgpOInLGwMJg1C84/H44ehX79zDlQ3syyzOQWc+fCK6/AtGmQlWV3VSIiIlJe+NpdgIiUTyEh8N13MHiw6YEaOBC+/hr69LG3rtRU2LAB/vkn/3bsWN5j33gDPvwQ6ta1p1YREREpPxScRKTEgoNNWLrsMvj+e4iPhy+/hAsvdP9zHzhQcDjavBlcroLv4+NjFvZt3NhMdLFwIbRpYwLUlVe6v2YREREpvxScRKRUAgPhq69M8Jg+3fRAffqpuSwLLhesWQOLF5uJKNauNQHp4MHT3yc8HJo1y781aAD+/uaYDRvgmmvg99/hqqtg5kwzhC8srGzqFhERkYpFwUlESs3f34Sla681l//5D0ydCpdffuaPdeSICTOLF8Nvv5nrR44UfGy9etC0af6AVKMGOByFP0+jRvDLL/D44/Dkk2bI3q+/wkcfQdeuZ163iIiIVGwKTiJSJvz84OOPTYj68EPTi5OeDldccfr7WBasX28CUnZQWr3a7D9ZlSrQqRN06WKG1jVrZobbBQeXvubHHoO+fU3o27wZuneHRx+FMWPAV/9CioiIyAn6WCAiZcbHByZPNuHp3Xfhuuvg+HEH1aubnx87Bn/8kRuUFi8ueMhdgwYmJHXtai5bt3ZviDn3XEhIgFGjTI/TY4+ZCS8++ggaNnTf84qIiEj5oeAkImXKxwfeegsCAuC112DECF+6d+/AhAm+/PVX/inAAwLg7LNzg9I550BMjOfrDg83PWUDB8Jtt8GSJdCuHbz8MgwbVvTQPxEREanYFJxEpMw5nWaihYAAeOEF+OWX2jk/q107tyepa1cTTrInbPAGV11l6rr2WnMO1PXXm4kj3ngDIiPtrk5ERETsouAkIm7hcMCkSVCvXhbz5m3miivq0b27L3Xq2F1Z0eLizHTlzzwD48bB55+bYYUffGAW/RUREZHKR8FJRNzG4YDbbnMRF7eagQPj8POzu6Li8/GBhx82C/pefbWZxKJXL7j/fjMTX1n0kmVmmmnRV62Cv/4y53vVqGGGKsbG5l7WqOFdvXIiIiKVkYKTiEghzj4b/vwT7rkH3nkHJk6EuXPNDILNmhXvMSwLdu82Aenkbc0aSEsr3mNEReUGqZND1amXWodKRETEPRScRESKEBICb79tJo646SYTpM46C55/Hm65Je/EEcnJ8PffueHor7/M5aFDBT92cDC0amWmWY+JgX37YM8eE7R27zbXMzNNb9TBg2a69sIEBUFMjC9RUZ1JSHDStasJfxERZdYcIiIilZKCk4hIMQ0eDJ07m1n25s41s+99/z20bZsblDZvLvi+TqdZe6pNGzO9evZWv7752em4XCZ0nRqmTr7Mvp6UBMePw+bNDjZvjmHZstzHadrU1J69tWlDuRo6KSIiYjcFJxGRM1CzJvz4I/zvf/Dgg/Ddd2Y7WWxs3nDUujU0b256g86U0wnR0WZr1arwY1NSTIDasiWTqVPXkpzckj/+cLJpE6xbZ7YPPjDHBgaaXrNOnXLDVL16mnZdRETkdBScRETOkNNpznm64AJ49lkTiE4OSdHR9tQVHGwWD65Tx+Lo0U0MHNgMPz8n+/fD0qVm+/13c3n4sFmI+Lffcu9frVreXikN8RMREcml4CQiUkJt28JHH9ldRdGqVYMLLzQbmMkq1q83ISo7SCUkwP79+XvQWrY0a1tdey3UrWtL+SIiIl5BwUlEpJJxOKBJE7Nde63Zl5pqwtPJYWrjRjMZxaOPwpgxZjr24cPNuV7BwZ6v2+WC5cvhn39Mz16rVuCr/8VERMRD9F+OiIgQGAjnnGO2bAcOmMkv3n/fLAg8d67ZQkPh8stNiOrWzb3nRR09CnPmmDq+/x727s39WZUq5hytLl3Mds459g2TFBGRik/BSUREChQdbWYQHDYMtmwxE0tMmWJmDnz3XbM1bGgC1LXXQlxc2Tzvxo0mJH33HcyfDxkZuT8LCTEzAv79t5lF8OefzZatcWMTorp2NZctW5rFjEVEREpLwUlERIpUrx6MHWuG7f36qwlQn39uQs6YMWa74AITooYMMb1BxZWRYSapyD6/6p9/8v68YUOIj4eLLoLu3cHfH7KyYO1aWLw4d/vnH3Pu1vr1ubMHhobm75WKjCyjRhERkUpFwUlERIrN6YQePcz28svw1VcmRP30U+52++3wn/+YENW9e8FD+Q4cgB9+MEHphx/gyJHcn/n6mvtddJGZ0KJJk/yP4eNjznFq1QpuvtnsO3TInJ+VHaSWLDFD/ebNM1u2pk1zg9T555teKhERkaIoOImISIlUqWKG6F17LWzdCh9+aELUxo0webLZGjSA664zW3Jybq/SkiVmsods0dEwYIAJS337lmwa9MhI8xgDBpjbWVlmcouTe6X+/Td3TaspU8xxTZvCxRebrUsXDe0TEZGCKTiJiEipxcWZYXyPPAKLFplQ8tlnsGkTjB9vtlO1aWOC0kUXmeF0ZR1YfHzMc7RpA7fcYvYdPGhC2+LFps5Fi0yIevZZs0VHm16uiy82AS4kpGxrEu+TlGSGdGrx58K5XObLCJHKzGl3ASIiUnE4HHDuufDOO7Bnj1nnqndvsz8w0ISS1183PVQrV8KTT3q2lycqytTwxBNmUon9++HTT+Hqq6FqVTOE8P334dJLzbEDBph6d+zwTH3iGZYFs2ebc+ciIszQ0E2b7K7KOyUlwcMPQ3i4Lzfc0I/HHnPmmd1SpDJRcBIREbcIDjaBZM4c09Nz6JAZpnfrrd6zmG54uJla/aOPYN8+M4vf6NFmQor0dHP+1e23Q5060KEDTJgAf/5pPnhXdCkp8O235vc1ciSsWWN3RaWXnAyvvQYtWkC/fub9aFmm57FtW3jvvcrxuy2OzEx46y1zDuDTT0NamoMjRwJ54gkf6taF6683X36IlERWVvn8skLBSURE3K5qVQgKsruKwvn6wnnnwaRJZma+NWvgv/81U5s7HCYwjR9vAlTduiZQ/fADpKWV/rldLvM4R4+aD6x22bPH9BZefLEZtnjxxfDmmyZstGxpFj9eutS++kpq40a45x6oVcuEwH/+McPz7rzThOUePUyouvFGMyvk/v12V2yv2bOhfXszxHXfPhOePv88k/vu+4NOnVykp5vhuO3amdk0v/km7zmLIoVJTja9+l26mKUuyhOd4yQiInIKhwOaNzfbAw+YD4/ff296YH780Qzde/11s4WE+NKsWWc++MCHzEwzvXp6et7t1H2n3j753JGgIDjrLOjc2WydOplzyNxxDo5lmTWxvvnGvLbff8/787p1zXC2PXvMDIozZpjtggvM8K0LLvDec4MsyyzY/NJL5neX3ZPUpAnccYdZnyw01Oz76ScTmB991Ly+334zvU8XXmhb+bZYswbuuw9mzTK3q1Y1Xxbceis4HBZ+frt46ql2LF/u5MUX4YsvctdSa9TIBNHrry9f5wa6XOZ3/tNP0KuX+bJAE8S4z/bt5t+UlSshIABWrTLLXZQbViVz5MgRC7COHDlidymWZVlWenq6NWPGDCs9Pd3uUioFtbfnqc09T23uXsePW9b331vWLbdYVmysZZmP5O7dqle3rPh4y3r8ccuaPduyDh8uef1paZY1Z45l3XGHZdWrl/+5zj7bPE9CgmW5XLn3W7PGsoYNsyxf37zHTp9uWVlZpWzUM1TYe/zoUct69VXLat487+saONCyZs0qvNaEBMtq2TL3PrfcYlnJyW58IV5i3z7Luu02y/LxMa/b19ey7rnHsg4ezD2moDbfutWy/u//LCsiIrfNwsMt6777LGvLFs+/jjPhclnWV19ZVtu2ed8njRtb1htvWFZKit0VVrx/y5cutayYmNx/0377ze6KjDPJBupxEhEROQOBgTBwoNleew2WLs3k3XfX0KpVS4KCfPD3p8DNz6/g/Sf/zM8Ptm0zPT/Z28qVpsfr22/Nlq1ZM9Mbld0z1bq1eYyCHDpkehG++cYML0xKyvt6evc237RfeCHUrFnwYzRvboZnTZhgemfefhv++MMM32vRAh58EK680rwGO2zcCK++anqKstcFCw01PSAjR5qepqK0bQvLlpnZIZ9/3gxT/OknM9V+587urd8OqammR+7JJ3PfE4MHwzPPFG99s7p1zbFjxphFp1980Qxzfe45035Dhpghkl26eE/PpGWZv4Px4yEhwewLDYVLLjHnvK1fb3rYxowxPZO3324mipHS+fxzsyxFaqpZf+/bb8tZT1M2DwQ5r6Iep8pN7e15anPPU5t7lrvb+/hx883sCy9Y1pVXWlaDBgX3SgUEWFaXLpZ1992WNW2aZf3xh2VNmmRZ552X25Nwcg/WjTda1owZJe9R2bvXsh5+2LLCwnIfNy7Osl55xf3f1me3eVpaujV7tmVddJFlORy5dTRpYlkvvWRZpfmvft48y6pd2zyej49ljRtnWRXlT8rlsqzPPrOs+vVz26x9e8v6+efT36c47/OsLMv67jvL6tUr7/utUyfLmjrV3vZzuSzr66/N68yuKyTEsh55JLdnLSnJ/J3VrZt7THCwZY0aZVmbNnm+5orwb7nLZXqwT+759ZKP4DnOJBt4RXB65ZVXrLi4OCsgIMDq1KmT9fvvvxd6/AsvvGA1adLECgwMtGrXrm3dfffd1vHjx4v1XApOlZva2/PU5p6nNvcsO9p73z7zAXXsWMvq18+yqlYterhfq1Ym6CxeXLZD6xITLevpp00YOzmYPf20+VlZy8qyrC1b0q1bbkmwmjVz5XmNAwYUPRzvTBw6ZFlXXZU3APz7b9k8tl2WLLGsrl1zX1PNmpY1ZUrRbXam7/OVKy3rhhtMoM9+rlq1zPvi5CGA7uZyWdY331jWWWflDUwPPWRZBw4UfJ/0dMv6+GPLatcu9z5Op2VdcYVlLVvmudrL+7/lx49b1tVX57bh3XdbVmam3VXlV66C0yeffGL5+/tb7733nrV69Wrr5ptvtiIiIqy9e/cWePzHH39sBQQEWB9//LG1efNm68cff7RiY2Ote+65p1jPp+BUuam9PU9t7nlqc8/yhvZ2ucwH+g8/NN+On3226Qnq1cuyXnzRsjZudH8NKSmmtykuLu/5Lg8/bHqniistzbI2bLCsn36yrMmTLWv8eMu6/nrLuuACy2rY0LL8/fMGwtBQy7rzTstat85NL8wyvSXZ5/EEB5tzYE4+/6s82LrVsoYOzduTMn588XscS/o+37vXsiZMsKwaNXKfOzDQ/D4fe8yyFi60rNTUErygIrhc5suFjh1zn7dKFct68EHL2r+/+I8xZ45l9e2b9z13/vmWNXOm+98D3vBvS0nt3Zsb0H18zN+MtypXwalTp07WyJEjc25nZWVZNWvWtJ5++ukCjx85cqR1wQUX5Nk3evRoq1u3bsV6PgWnyk3t7Xlqc89Tm3uW2juv9HTLev/9vJMzBAWZySi2bjXDoVatMh9qX33Vsh54wAxB7NLF9H6cPOTudJuPj8uqW/eI9cILmR4b9rNtm/mwn13DhRda1u7dnnnu0khKMuE1MNDU7XBY1vDhlrVjx5k9Tmnf56mppmfr1MkYTg1SCxaULki5XGbylrPPzhuYHnig+IGpIAkJlnXNNXmHvbZqZV5TWlrJH7cw5fXfllWrcieeiYgw4dOblZvJIdLT01m+fDkPPfRQzj6n00nv3r1ZvHhxgffp2rUrH330EUuXLqVTp05s2rSJmTNncu211xZ4fFpaGmknLbKRdOLsx4yMDDIyMsrw1ZRMdg3eUEtloPb2PLW556nNPUvtnd9VV8EVV8C33zp45hkny5Y5efllePnl4t0/MNCibl2IizOXdetaJzZzvXr1DH7++Wf69OmDn58fnmj6mBiYORNeftnJo486+f57B61bW7z+ehaXXGK5v4AzsHMn/P67g99/d/Dxx0727TMzM/To4eLZZ7No394cdybtVtr3udMJQ4ea98batfDLL04WLHCwcKGDffsc/PSTmYgDzO//nHMsevQwW6dOFoGBhT++ZcHs2Q4ef9zJ0qVmmdLgYItbb3Vx770uqlXLfh0lKp8WLczEI+PHm/fAu+86+ftvB8OHwyOPWIwa5eKmm1yEh5fs8QtS0jY/cgQ2bHDw77/mcv16B8HBcM01Lrp1s9w6UccPPzi4+mofjh510KiRxfTpmTRtWvJ294QzaV+HZVm2/bXv2rWLWrVq8dtvv9GlS5ec/f/3f//HggUL+P3UBSVOeOmll7jvvvuwLIvMzExuvfVWXn/99QKPHT9+PBMmTMi3f+rUqQQHB5fNCxEREZECWRb89Vc1vviiMatWmU+vISHpVKuWQrVqx6lePYXoaHNZrdpxqlVLITw83WtmYSvI1q2hvPBCB7ZsMZ+Se/Xayk03/U1QkOdXL05Lc7JpUwTr1lXl33+rsm5dJAcP5l1tOjY2meHDV9Op0x6va1fLgh07Qvj772j+/jua1aujSEzMm5L8/LJo2vQQrVodpFWrAzRpchh/f1fO/RMSqvHJJ81Yty4SAH//TAYO3MygQRuIiEh3S93JyX78+GM9vvuuAYcPm3qDgzPo23cLPXrsIDw8nSpVMggIyHJLm6elOdmzpwq7doWwe3cVdu4Mybl+avudrHbto/Ttu4WePbcTFlZ2acay4PvvG/Dee61wuRy0bHmABx5YWqbP4S4pKSkMHTqUI0eOEBYWVuix5S44zZ8/nyuvvJInnniCzp07s2HDBu666y5uvvlmxowZk+/4gnqc6tSpw4EDB4psHE/IyMhgzpw5Od+aiXupvT1Pbe55anPPUnsX3549UKVK7sKzJeUNbZ6WBuPHO3n+eSeW5aBBA4tXX82ibVuLyEjTw1LWLAs2bTK9SUuXmh6llSsdZGbm/WTudFq0bg2dO7vo2tXissus005VX1yeanPLgnXrYOHC3B6pvXvzvr6AAIvOnS26drWYP9/BkiWmsYOCLG65xfQw1ajhthLzSEuDadMcTJrkw7p1+ROSn59F1aoQEQEREXmvR0SYRYarVrUID+fEz3KP8fPL4JNPfqN69XPZssWP9eth/XrTg7R9O1jW6RNZTIxFo0YWjRtDo0YWGzc6+PRTB8eOmfsEBFgMGWJx000uzj23dL1QGRkwerSTN980KwcPH+7ilVeySv2e85SkpCSio6OLFZxsHaoXHR2Nj48Pe/fuzbN/7969xMTEFHifMWPGcO2113LTTTcB0Lp1a44dO8aIESN45JFHcJ7yL1VAQAABAQH5HsfPz8+r/oPztnoqOrW356nNPU9t7llq76LVqVO2j2dnm/v5mfWKLr7YrE+zaZODAQPMxyofH7P2T7VqZqteveDr2bdPF7SSksxaWUuW5G4HDuQ/rkYNs1bSOeeYrUMHByEhAD5ueN3ub/PWrc02cqQJUv/+C/Pn52579phAtXChOT4wEG67Df7v/xzExPjgjtd9On5+cPPNcOON8P33Zi2rv/6Cw4chKwsyMhzs22fWYoMzTSd+QO/T/jQ83KxPlr01bpx7GRbmyPd8L7wA06aZ9clWrHAwbZqDadOcNG0KI0aY93F09JlVmJgIl18Oc+aYtbqeeQbuu8+Jw+GGbw7c5Ezez7YGJ39/fzp06MC8efMYNGgQAC6Xi3nz5jFq1KgC75OSkpIvHPn4mD8QGzvPREREpBLq0cMsUnzfffDll7kfmHM/LBfN6cwNWtlB6t9/YfVqExxO5u8PZ52VG5LOOccsROttQ/DKisMBTZua7ZZbcoPUggXw669mwea77oLYWHvrdDohPt5sYOpMTjbB4vBhs518vajbqanmcfz9s2jSxEnTpo484ahJExNyzuT3HhZm2nDECFi+HN56C6ZONT18994LDz0El15qfn7eeUU/9saNcNFF8M8/EBxsHuuSS8687coTW4MTwOjRoxk2bBgdO3akU6dOvPjiixw7dozrr78egOuuu45atWrx9NNPAxAfH8/zzz9P+/btc4bqjRkzhvj4+JwAJSIiIuIp4eHw9ttmy8gwvUL795vgtH9/4dcPHwaXK/f2mjV5H7tevbwhqV07KGAgTaVxcpAaMcLuak7P4TBDUkNDS9bTmpoKhw9nsGTJTC66aGCZ9vI5HNCxo9kmTcrthfrzT3N92jQTzEaMgGHDCu6FWrgQhgyBgwehdm345htyJh2pyGwPTldccQX79+9n7Nix7Nmzh3bt2vHDDz9Q48Tg1G3btuXpYXr00UdxOBw8+uij7Ny5k2rVqhEfH8+TTz5p10sQERERAczQrdjY4veAnBy0skPVwYPmw3bnzmY2P6l8AgNNYHHHuXInCw01AenUXqh//zW9qA8/bALSiBHQs6cJXVOmmNsZGXD22fD11/b3+HmK7cEJYNSoUacdmjd//vw8t319fRk3bhzjxo3zQGUiIiIi7nOmQUvEXTp0MD1Pzz0Hn3xiQtSyZeb6J5+YYYIdO5oeKYD//MeEqMo0SXX5OXNLRERERETcKjTUTHjxxx+mF+qWWyAkBNavzw1NY8aYMFWZQhMoOImIiIiISAHOOgveeAN27zbn8F10kQlMjz3m/mGE3sgrhuqJiIiIiIh3CgmBm24yW2VWCbOiiIiIiIjImVFwEhERERERKYKCk4iIiIiISBEUnERERERERIqg4CQiIiIiIlIEBScREREREZEiKDiJiIiIiIgUQcFJRERERESkCApOIiIiIiIiRVBwEhERERERKYKCk4iIiIiISBEUnERERERERIqg4CQiIiIiIlIEBScREREREZEiKDiJiIiIiIgUQcFJRERERESkCApOIiIiIiIiRVBwEhERERERKYKv3QV4mmVZACQlJdlciZGRkUFKSgpJSUn4+fnZXU6Fp/b2PLW556nNPUvt7Xlqc89Tm3ue2twzsjNBdkYoTKULTkePHgWgTp06NlciIiIiIiLe4OjRo4SHhxd6jMMqTryqQFwuF7t27SI0NBSHw2F3OSQlJVGnTh22b99OWFiY3eVUeGpvz1Obe57a3LPU3p6nNvc8tbnnqc09w7Isjh49Ss2aNXE6Cz+LqdL1ODmdTmrXrm13GfmEhYXpj8KD1N6epzb3PLW5Z6m9PU9t7nlqc89Tm7tfUT1N2TQ5hIiIiIiISBEUnERERERERIqg4GSzgIAAxo0bR0BAgN2lVApqb89Tm3ue2tyz1N6epzb3PLW556nNvU+lmxxCRERERETkTKnHSUREREREpAgKTiIiIiIiIkVQcBIRERERESmCgpOIiIiIiEgRFJxs9Oqrr1KvXj0CAwPp3LkzS5cutbukCmv8+PE4HI48W7Nmzewuq0JZuHAh8fHx1KxZE4fDwYwZM/L83LIsxo4dS2xsLEFBQfTu3Zv169fbU2wFUFR7Dx8+PN97vn///vYUW0E8/fTTnH322YSGhlK9enUGDRrEunXr8hyTmprKyJEjiYqKIiQkhEsvvZS9e/faVHH5Vpz27tmzZ773+a233mpTxeXf66+/Tps2bXIWXO3SpQuzZs3K+bne32WvqDbXe9y7KDjZ5NNPP2X06NGMGzeOP//8k7Zt29KvXz/27dtnd2kVVsuWLdm9e3fO9uuvv9pdUoVy7Ngx2rZty6uvvlrgzydOnMhLL73EG2+8we+//06VKlXo168fqampHq60YiiqvQH69++f5z0/bdo0D1ZY8SxYsICRI0eyZMkS5syZQ0ZGBn379uXYsWM5x9xzzz18++23fP755yxYsIBdu3YxZMgQG6suv4rT3gA333xznvf5xIkTbaq4/Ktduzb//e9/Wb58OcuWLeOCCy7gkksuYfXq1YDe3+5QVJuD3uNexRJbdOrUyRo5cmTO7aysLKtmzZrW008/bWNVFde4ceOstm3b2l1GpQFY06dPz7ntcrmsmJgY69lnn83Zl5iYaAUEBFjTpk2zocKK5dT2tizLGjZsmHXJJZfYUk9lsW/fPguwFixYYFmWeU/7+flZn3/+ec4xa9eutQBr8eLFdpVZYZza3pZlWeedd55111132VdUJVC1alXrnXfe0fvbg7Lb3LL0Hvc26nGyQXp6OsuXL6d37945+5xOJ71792bx4sU2VlaxrV+/npo1a9KgQQOuvvpqtm3bZndJlcbmzZvZs2dPnvd8eHg4nTt31nvejebPn0/16tVp2rQpt912GwcPHrS7pArlyJEjAERGRgKwfPlyMjIy8rzPmzVrRt26dfU+LwOntne2jz/+mOjoaFq1asVDDz1ESkqKHeVVOFlZWXzyySccO3aMLl266P3tAae2eTa9x72Hr90FVEYHDhwgKyuLGjVq5Nlfo0YN/vnnH5uqqtg6d+7MlClTaNq0Kbt372bChAl0796dv//+m9DQULvLq/D27NkDUOB7PvtnUrb69+/PkCFDqF+/Phs3buThhx9mwIABLF68GB8fH7vLK/dcLhd333033bp1o1WrVoB5n/v7+xMREZHnWL3PS6+g9gYYOnQocXFx1KxZk7/++osHHniAdevW8dVXX9lYbfm2atUqunTpQmpqKiEhIUyfPp0WLVqQkJCg97ebnK7NQe9xb6PgJJXCgAEDcq63adOGzp07ExcXx2effcaNN95oY2Ui7nHllVfmXG/dujVt2rShYcOGzJ8/n169etlYWcUwcuRI/v77b50r6SGna+8RI0bkXG/dujWxsbH06tWLjRs30rBhQ0+XWSE0bdqUhIQEjhw5whdffMGwYcNYsGCB3WVVaKdr8xYtWug97mU0VM8G0dHR+Pj45JuJZu/evcTExNhUVeUSERFBkyZN2LBhg92lVArZ72u95+3ToEEDoqOj9Z4vA6NGjeK7777j559/pnbt2jn7Y2JiSE9PJzExMc/xep+XzunauyCdO3cG0Pu8FPz9/WnUqBEdOnTg6aefpm3btvzvf//T+9uNTtfmBdF73F4KTjbw9/enQ4cOzJs3L2efy+Vi3rx5eca0ivskJyezceNGYmNj7S6lUqhfvz4xMTF53vNJSUn8/vvves97yI4dOzh48KDe86VgWRajRo1i+vTp/PTTT9SvXz/Pzzt06ICfn1+e9/m6devYtm2b3uclUFR7FyQhIQFA7/My5HK5SEtL0/vbg7LbvCB6j9tLQ/VsMnr0aIYNG0bHjh3p1KkTL774IseOHeP666+3u7QK6b777iM+Pp64uDh27drFuHHj8PHx4aqrrrK7tAojOTk5zzdgmzdvJiEhgcjISOrWrcvdd9/NE088QePGjalfvz5jxoyhZs2aDBo0yL6iy7HC2jsyMpIJEyZw6aWXEhMTw8aNG/m///s/GjVqRL9+/WysunwbOXIkU6dO5euvvyY0NDTnvI7w8HCCgoIIDw/nxhtvZPTo0URGRhIWFsYdd9xBly5dOOecc2yuvvwpqr03btzI1KlTGThwIFFRUfz111/cc8899OjRgzZt2thcffn00EMPMWDAAOrWrcvRo0eZOnUq8+fP58cff9T7200Ka3O9x72Q3dP6VWYvv/yyVbduXcvf39/q1KmTtWTJErtLqrCuuOIKKzY21vL397dq1aplXXHFFdaGDRvsLqtC+fnnny0g3zZs2DDLssyU5GPGjLFq1KhhBQQEWL169bLWrVtnb9HlWGHtnZKSYvXt29eqVq2a5efnZ8XFxVk333yztWfPHrvLLtcKam/Amjx5cs4xx48ft26//XaratWqVnBwsDV48GBr9+7d9hVdjhXV3tu2bbN69OhhRUZGWgEBAVajRo2s+++/3zpy5Ii9hZdjN9xwgxUXF2f5+/tb1apVs3r16mXNnj075+d6f5e9wtpc73Hv47Asy/JkUBMRERERESlvdI6TiIiIiIhIERScREREREREiqDgJCIiIiIiUgQFJxERERERkSIoOImIiIiIiBRBwUlERERERKQICk4iIiIiIiJFUHASEREREREpgoKTiIhIIRwOBzNmzLC7DBERsZmCk4iIeK3hw4fjcDjybf3797e7NBERqWR87S5ARESkMP3792fy5Ml59gUEBNhUjYiIVFbqcRIREa8WEBBATExMnq1q1aqAGUb3+uuvM2DAAIKCgmjQoAFffPFFnvuvWrWKCy64gKCgIKKiohgxYgTJycl5jnnvvfdo2bIlAQEBxMbGMmrUqDw/P3DgAIMHDyY4OJjGjRvzzTff5Pzs8OHDXH311VSrVo2goCAaN26cL+iJiEj5p+AkIiLl2pj/b+d+QqH74jiOf64QBkWDJhsLNaEof8rfhab8K1IjqUnDRv4kGyXy+BNLYWWKWJGpWSjJEJZKlPwpw46URNlIseG3eGpqep5+o9/jx3h6v1bnnnO753vv7tO55/z4IbvdruPjYzkcDjU1Ncnn80mSnp6eVFlZqYSEBB0cHMjj8Wh7ezsgGLlcLnV1damtrU2np6daXV1Venp6wByjo6NqbGzUycmJampq5HA49PDw4J//7OxMXq9XPp9PLpdLZrP58z4AAOBTGG9vb29fXQQAAL/T0tKixcVFRUVFBfQPDAxoYGBAhmGovb1dLpfLP1ZYWKjc3FzNzMxobm5OfX19ur6+lslkkiStr6+rtrZWNzc3SklJUWpqqlpbWzU+Pv7bGgzD0ODgoMbGxiT9DGOxsbHyer2qqqpSXV2dzGazFhYW/qevAAAIBexxAgCEtPLy8oBgJEmJiYn+dlFRUcBYUVGRjo6OJEk+n085OTn+0CRJJSUlen191cXFhQzD0M3NjWw227/WkJ2d7W+bTCbFx8fr7u5OktTR0SG73a7Dw0NVVFSovr5excXF/+ldAQChi+AEAAhpJpPpl1/nPkp0dPS77ouIiAi4NgxDr6+vkqTq6mpdXV1pfX1dW1tbstls6urq0sTExIfXCwD4OuxxAgB8a3t7e79cZ2RkSJIyMjJ0fHysp6cn//ju7q7CwsJktVoVFxentLQ07ezs/FENSUlJcjqdWlxc1PT0tGZnZ//oeQCA0MOKEwAgpL28vOj29jagLzw83H8Ag8fjUX5+vkpLS7W0tKT9/X3Nz89LkhwOh4aHh+V0OjUyMqL7+3t1d3erublZKSkpkqSRkRG1t7crOTlZ1dXVenx81O7urrq7u99V39DQkPLy8pSVlaWXlxetra35gxsA4O9BcAIAhLSNjQ1ZLJaAPqvVqvPzc0k/T7xzu93q7OyUxWLR8vKyMjMzJUkxMTHa3NxUT0+PCgoKFBMTI7vdrsnJSf+znE6nnp+fNTU1pd7eXpnNZjU0NLy7vsjISPX39+vy8lLR0dEqKyuT2+3+gDcHAIQSTtUDAHxbhmFoZWVF9fX1X10KAOAvxx4nAAAAAAiC4AQAAAAAQbDHCQDwbfG3OQDgs7DiBAAAAABBEJwAAAAAIAiCEwAAAAAEQXACAAAAgCAITgAAAAAQBMEJAAAAAIIgOAEAAABAEAQnAAAAAAjiH/4mFbSq+J5NAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy only on dynamic proteins: 0.3653061224489796\n"
     ]
    }
   ],
   "source": [
    "# Setting seed\n",
    "torch.manual_seed(30026)\n",
    "if(torch.cuda.is_available()):\n",
    "    torch.cuda.manual_seed(30026)\n",
    "    torch.cuda.manual_seed_all(30026)\n",
    "\n",
    "# Run training\n",
    "train_loss, train_accuracy, test_loss, test_accuracy, lstm_dynamic_model = dynamic_training(num_features, num_classes, num_timesteps, embedding_dim, dynamic_params, best_static_model, dataloader_train, dataloader_test, verbose, device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Second Model: TCNModel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Change model_type for running the right model (from 0 to 1, for more info look at dynamic_training inside runner_dynamic.py)\n",
    "dynamic_params[\"model_type\"] +=1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting training of TCNDynamicModel\n",
      "Epoch [1/100],\tTraining Loss: 1.5848, Training accuracy: 49.00,\tTest Loss: 1.4106, Test Accuracy: 53.09,\tLearning rate: 0.00100\n",
      "Epoch [2/100],\tTraining Loss: 1.2345, Training accuracy: 59.17,\tTest Loss: 1.3427, Test Accuracy: 55.63,\tLearning rate: 0.00100\n",
      "Epoch [3/100],\tTraining Loss: 1.1060, Training accuracy: 63.08,\tTest Loss: 1.3180, Test Accuracy: 57.74,\tLearning rate: 0.00100\n",
      "Epoch [4/100],\tTraining Loss: 1.0281, Training accuracy: 65.25,\tTest Loss: 1.2937, Test Accuracy: 58.36,\tLearning rate: 0.00100\n",
      "Epoch [5/100],\tTraining Loss: 0.9621, Training accuracy: 67.82,\tTest Loss: 1.3214, Test Accuracy: 58.76,\tLearning rate: 0.00100\n",
      "Epoch [6/100],\tTraining Loss: 0.9241, Training accuracy: 68.31,\tTest Loss: 1.2948, Test Accuracy: 58.84,\tLearning rate: 0.00100\n",
      "Epoch [7/100],\tTraining Loss: 0.8901, Training accuracy: 69.59,\tTest Loss: 1.2888, Test Accuracy: 57.94,\tLearning rate: 0.00100\n",
      "Epoch [8/100],\tTraining Loss: 0.8313, Training accuracy: 71.01,\tTest Loss: 1.3050, Test Accuracy: 59.04,\tLearning rate: 0.00100\n",
      "Epoch [9/100],\tTraining Loss: 0.8163, Training accuracy: 71.96,\tTest Loss: 1.3101, Test Accuracy: 58.59,\tLearning rate: 0.00100\n",
      "Epoch [10/100],\tTraining Loss: 0.7884, Training accuracy: 72.74,\tTest Loss: 1.3397, Test Accuracy: 60.00,\tLearning rate: 0.00100\n",
      "Epoch [11/100],\tTraining Loss: 0.7366, Training accuracy: 73.99,\tTest Loss: 1.3052, Test Accuracy: 58.62,\tLearning rate: 0.00100\n",
      "Epoch [12/100],\tTraining Loss: 0.7382, Training accuracy: 73.71,\tTest Loss: 1.3152, Test Accuracy: 58.48,\tLearning rate: 0.00100\n",
      "Epoch [13/100],\tTraining Loss: 0.7107, Training accuracy: 75.78,\tTest Loss: 1.3737, Test Accuracy: 59.89,\tLearning rate: 0.00100\n",
      "Epoch [14/100],\tTraining Loss: 0.6876, Training accuracy: 76.28,\tTest Loss: 1.3617, Test Accuracy: 59.27,\tLearning rate: 0.00100\n",
      "Epoch [15/100],\tTraining Loss: 0.6718, Training accuracy: 76.69,\tTest Loss: 1.4057, Test Accuracy: 57.49,\tLearning rate: 0.00100\n",
      "Epoch [16/100],\tTraining Loss: 0.6622, Training accuracy: 77.19,\tTest Loss: 1.3624, Test Accuracy: 60.65,\tLearning rate: 0.00100\n",
      "Epoch [17/100],\tTraining Loss: 0.6593, Training accuracy: 77.03,\tTest Loss: 1.3811, Test Accuracy: 57.52,\tLearning rate: 0.00100\n",
      "Epoch [18/100],\tTraining Loss: 0.6115, Training accuracy: 79.50,\tTest Loss: 1.3994, Test Accuracy: 59.07,\tLearning rate: 0.00100\n",
      "Epoch [19/100],\tTraining Loss: 0.5847, Training accuracy: 79.36,\tTest Loss: 1.4105, Test Accuracy: 58.50,\tLearning rate: 0.00100\n",
      "Epoch [20/100],\tTraining Loss: 0.5716, Training accuracy: 80.35,\tTest Loss: 1.4316, Test Accuracy: 60.17,\tLearning rate: 0.00010\n",
      "Epoch [21/100],\tTraining Loss: 0.5010, Training accuracy: 83.46,\tTest Loss: 1.4208, Test Accuracy: 60.48,\tLearning rate: 0.00010\n",
      "Epoch [22/100],\tTraining Loss: 0.4859, Training accuracy: 83.91,\tTest Loss: 1.4348, Test Accuracy: 60.65,\tLearning rate: 0.00010\n",
      "Epoch [23/100],\tTraining Loss: 0.4726, Training accuracy: 84.02,\tTest Loss: 1.4323, Test Accuracy: 60.42,\tLearning rate: 0.00010\n",
      "Epoch [24/100],\tTraining Loss: 0.4575, Training accuracy: 84.88,\tTest Loss: 1.4407, Test Accuracy: 60.51,\tLearning rate: 0.00010\n",
      "Epoch [25/100],\tTraining Loss: 0.4536, Training accuracy: 84.72,\tTest Loss: 1.4579, Test Accuracy: 60.48,\tLearning rate: 0.00010\n",
      "Epoch [26/100],\tTraining Loss: 0.4536, Training accuracy: 84.87,\tTest Loss: 1.4414, Test Accuracy: 60.68,\tLearning rate: 0.00010\n",
      "Epoch [27/100],\tTraining Loss: 0.4604, Training accuracy: 84.18,\tTest Loss: 1.4533, Test Accuracy: 60.45,\tLearning rate: 0.00010\n",
      "Early stopping triggered after 27 epochs. Best test loss: 1.2888\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy only on dynamic proteins: 0.3795918367346939\n"
     ]
    }
   ],
   "source": [
    "# Setting seed\n",
    "torch.manual_seed(30026)\n",
    "if(torch.cuda.is_available()):\n",
    "    torch.cuda.manual_seed(30026)\n",
    "    torch.cuda.manual_seed_all(30026)\n",
    "\n",
    "# Run training\n",
    "train_loss, train_accuracy, test_loss, test_accuracy, dynamic_model = dynamic_training(num_features, num_classes, num_timesteps, embedding_dim, dynamic_params, best_static_model, dataloader_train, dataloader_test, verbose, device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Third Model: SimpleModel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Change model_type for running the right model (from 1 to 2, for more info look at dynamic_training inside runner_dynamic.py)\n",
    "dynamic_params[\"model_type\"] +=1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting training of SimpleDynamicModel\n",
      "Epoch [1/100],\tTraining Loss: 2.0910, Training accuracy: 29.43,\tTest Loss: 1.8369, Test Accuracy: 37.86,\tLearning rate: 0.00100\n",
      "Epoch [2/100],\tTraining Loss: 1.7588, Training accuracy: 39.54,\tTest Loss: 1.7218, Test Accuracy: 40.25,\tLearning rate: 0.00100\n",
      "Epoch [3/100],\tTraining Loss: 1.6581, Training accuracy: 43.92,\tTest Loss: 1.6460, Test Accuracy: 44.15,\tLearning rate: 0.00100\n",
      "Epoch [4/100],\tTraining Loss: 1.5868, Training accuracy: 46.96,\tTest Loss: 1.5928, Test Accuracy: 45.44,\tLearning rate: 0.00100\n",
      "Epoch [5/100],\tTraining Loss: 1.5332, Training accuracy: 48.04,\tTest Loss: 1.5538, Test Accuracy: 46.63,\tLearning rate: 0.00100\n",
      "Epoch [6/100],\tTraining Loss: 1.4912, Training accuracy: 50.67,\tTest Loss: 1.5318, Test Accuracy: 47.56,\tLearning rate: 0.00100\n",
      "Epoch [7/100],\tTraining Loss: 1.4597, Training accuracy: 51.94,\tTest Loss: 1.4998, Test Accuracy: 48.63,\tLearning rate: 0.00100\n",
      "Epoch [8/100],\tTraining Loss: 1.4306, Training accuracy: 52.60,\tTest Loss: 1.4841, Test Accuracy: 48.91,\tLearning rate: 0.00100\n",
      "Epoch [9/100],\tTraining Loss: 1.4041, Training accuracy: 53.69,\tTest Loss: 1.4680, Test Accuracy: 48.55,\tLearning rate: 0.00100\n",
      "Epoch [10/100],\tTraining Loss: 1.3849, Training accuracy: 54.73,\tTest Loss: 1.4496, Test Accuracy: 50.04,\tLearning rate: 0.00100\n",
      "Epoch [11/100],\tTraining Loss: 1.3650, Training accuracy: 55.16,\tTest Loss: 1.4383, Test Accuracy: 49.90,\tLearning rate: 0.00100\n",
      "Epoch [12/100],\tTraining Loss: 1.3469, Training accuracy: 55.90,\tTest Loss: 1.4298, Test Accuracy: 50.92,\tLearning rate: 0.00100\n",
      "Epoch [13/100],\tTraining Loss: 1.3339, Training accuracy: 55.84,\tTest Loss: 1.4200, Test Accuracy: 51.68,\tLearning rate: 0.00100\n",
      "Epoch [14/100],\tTraining Loss: 1.3211, Training accuracy: 56.42,\tTest Loss: 1.4123, Test Accuracy: 52.10,\tLearning rate: 0.00100\n",
      "Epoch [15/100],\tTraining Loss: 1.3068, Training accuracy: 56.63,\tTest Loss: 1.4070, Test Accuracy: 52.44,\tLearning rate: 0.00100\n",
      "Epoch [16/100],\tTraining Loss: 1.2951, Training accuracy: 56.70,\tTest Loss: 1.4001, Test Accuracy: 52.21,\tLearning rate: 0.00100\n",
      "Epoch [17/100],\tTraining Loss: 1.2852, Training accuracy: 57.13,\tTest Loss: 1.3956, Test Accuracy: 52.10,\tLearning rate: 0.00100\n",
      "Epoch [18/100],\tTraining Loss: 1.2762, Training accuracy: 57.33,\tTest Loss: 1.3910, Test Accuracy: 52.52,\tLearning rate: 0.00100\n",
      "Epoch [19/100],\tTraining Loss: 1.2688, Training accuracy: 57.33,\tTest Loss: 1.3856, Test Accuracy: 52.81,\tLearning rate: 0.00100\n",
      "Epoch [20/100],\tTraining Loss: 1.2573, Training accuracy: 57.71,\tTest Loss: 1.3840, Test Accuracy: 53.09,\tLearning rate: 0.00010\n",
      "Epoch [21/100],\tTraining Loss: 1.2458, Training accuracy: 57.93,\tTest Loss: 1.3827, Test Accuracy: 53.51,\tLearning rate: 0.00010\n",
      "Epoch [22/100],\tTraining Loss: 1.2449, Training accuracy: 58.15,\tTest Loss: 1.3821, Test Accuracy: 53.26,\tLearning rate: 0.00010\n",
      "Epoch [23/100],\tTraining Loss: 1.2453, Training accuracy: 58.08,\tTest Loss: 1.3816, Test Accuracy: 53.12,\tLearning rate: 0.00010\n",
      "Epoch [24/100],\tTraining Loss: 1.2428, Training accuracy: 58.16,\tTest Loss: 1.3810, Test Accuracy: 53.40,\tLearning rate: 0.00010\n",
      "Epoch [25/100],\tTraining Loss: 1.2426, Training accuracy: 58.05,\tTest Loss: 1.3805, Test Accuracy: 53.26,\tLearning rate: 0.00010\n",
      "Epoch [26/100],\tTraining Loss: 1.2404, Training accuracy: 58.11,\tTest Loss: 1.3800, Test Accuracy: 53.26,\tLearning rate: 0.00010\n",
      "Epoch [27/100],\tTraining Loss: 1.2396, Training accuracy: 58.05,\tTest Loss: 1.3798, Test Accuracy: 53.26,\tLearning rate: 0.00010\n",
      "Epoch [28/100],\tTraining Loss: 1.2395, Training accuracy: 57.99,\tTest Loss: 1.3797, Test Accuracy: 53.40,\tLearning rate: 0.00010\n",
      "Epoch [29/100],\tTraining Loss: 1.2381, Training accuracy: 58.09,\tTest Loss: 1.3791, Test Accuracy: 53.26,\tLearning rate: 0.00010\n",
      "Epoch [30/100],\tTraining Loss: 1.2376, Training accuracy: 58.16,\tTest Loss: 1.3787, Test Accuracy: 53.26,\tLearning rate: 0.00010\n",
      "Epoch [31/100],\tTraining Loss: 1.2358, Training accuracy: 58.05,\tTest Loss: 1.3784, Test Accuracy: 53.40,\tLearning rate: 0.00010\n",
      "Epoch [32/100],\tTraining Loss: 1.2365, Training accuracy: 58.19,\tTest Loss: 1.3782, Test Accuracy: 53.40,\tLearning rate: 0.00010\n",
      "Epoch [33/100],\tTraining Loss: 1.2340, Training accuracy: 58.17,\tTest Loss: 1.3776, Test Accuracy: 53.26,\tLearning rate: 0.00010\n",
      "Epoch [34/100],\tTraining Loss: 1.2380, Training accuracy: 58.17,\tTest Loss: 1.3773, Test Accuracy: 53.26,\tLearning rate: 0.00010\n",
      "Epoch [35/100],\tTraining Loss: 1.2330, Training accuracy: 58.14,\tTest Loss: 1.3770, Test Accuracy: 53.54,\tLearning rate: 0.00010\n",
      "Epoch [36/100],\tTraining Loss: 1.2326, Training accuracy: 58.36,\tTest Loss: 1.3766, Test Accuracy: 53.26,\tLearning rate: 0.00010\n",
      "Epoch [37/100],\tTraining Loss: 1.2331, Training accuracy: 58.20,\tTest Loss: 1.3765, Test Accuracy: 53.68,\tLearning rate: 0.00010\n",
      "Epoch [38/100],\tTraining Loss: 1.2322, Training accuracy: 58.34,\tTest Loss: 1.3762, Test Accuracy: 53.54,\tLearning rate: 0.00010\n",
      "Epoch [39/100],\tTraining Loss: 1.2314, Training accuracy: 58.28,\tTest Loss: 1.3757, Test Accuracy: 53.54,\tLearning rate: 0.00010\n",
      "Epoch [40/100],\tTraining Loss: 1.2306, Training accuracy: 58.35,\tTest Loss: 1.3754, Test Accuracy: 53.54,\tLearning rate: 0.00001\n",
      "Epoch [41/100],\tTraining Loss: 1.2287, Training accuracy: 58.45,\tTest Loss: 1.3754, Test Accuracy: 53.54,\tLearning rate: 0.00001\n",
      "Epoch [42/100],\tTraining Loss: 1.2281, Training accuracy: 58.48,\tTest Loss: 1.3753, Test Accuracy: 53.68,\tLearning rate: 0.00001\n",
      "Epoch [43/100],\tTraining Loss: 1.2279, Training accuracy: 58.48,\tTest Loss: 1.3753, Test Accuracy: 53.54,\tLearning rate: 0.00001\n",
      "Epoch [44/100],\tTraining Loss: 1.2299, Training accuracy: 58.48,\tTest Loss: 1.3753, Test Accuracy: 53.68,\tLearning rate: 0.00001\n",
      "Epoch [45/100],\tTraining Loss: 1.2293, Training accuracy: 58.48,\tTest Loss: 1.3753, Test Accuracy: 53.68,\tLearning rate: 0.00001\n",
      "Epoch [46/100],\tTraining Loss: 1.2290, Training accuracy: 58.48,\tTest Loss: 1.3752, Test Accuracy: 53.68,\tLearning rate: 0.00001\n",
      "Epoch [47/100],\tTraining Loss: 1.2275, Training accuracy: 58.48,\tTest Loss: 1.3752, Test Accuracy: 53.68,\tLearning rate: 0.00001\n",
      "Epoch [48/100],\tTraining Loss: 1.2304, Training accuracy: 58.55,\tTest Loss: 1.3752, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [49/100],\tTraining Loss: 1.2267, Training accuracy: 58.48,\tTest Loss: 1.3752, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [50/100],\tTraining Loss: 1.2278, Training accuracy: 58.55,\tTest Loss: 1.3751, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [51/100],\tTraining Loss: 1.2282, Training accuracy: 58.55,\tTest Loss: 1.3751, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [52/100],\tTraining Loss: 1.2278, Training accuracy: 58.48,\tTest Loss: 1.3751, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [53/100],\tTraining Loss: 1.2263, Training accuracy: 58.55,\tTest Loss: 1.3751, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [54/100],\tTraining Loss: 1.2280, Training accuracy: 58.48,\tTest Loss: 1.3751, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [55/100],\tTraining Loss: 1.2264, Training accuracy: 58.59,\tTest Loss: 1.3750, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [56/100],\tTraining Loss: 1.2286, Training accuracy: 58.55,\tTest Loss: 1.3750, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [57/100],\tTraining Loss: 1.2267, Training accuracy: 58.62,\tTest Loss: 1.3749, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [58/100],\tTraining Loss: 1.2280, Training accuracy: 58.48,\tTest Loss: 1.3749, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [59/100],\tTraining Loss: 1.2273, Training accuracy: 58.52,\tTest Loss: 1.3749, Test Accuracy: 53.82,\tLearning rate: 0.00001\n",
      "Epoch [60/100],\tTraining Loss: 1.2294, Training accuracy: 58.55,\tTest Loss: 1.3749, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [61/100],\tTraining Loss: 1.2261, Training accuracy: 58.48,\tTest Loss: 1.3749, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [62/100],\tTraining Loss: 1.2260, Training accuracy: 58.55,\tTest Loss: 1.3749, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [63/100],\tTraining Loss: 1.2267, Training accuracy: 58.52,\tTest Loss: 1.3749, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [64/100],\tTraining Loss: 1.2291, Training accuracy: 58.52,\tTest Loss: 1.3749, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [65/100],\tTraining Loss: 1.2259, Training accuracy: 58.52,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [66/100],\tTraining Loss: 1.2257, Training accuracy: 58.52,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [67/100],\tTraining Loss: 1.2260, Training accuracy: 58.52,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [68/100],\tTraining Loss: 1.2279, Training accuracy: 58.52,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [69/100],\tTraining Loss: 1.2262, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [70/100],\tTraining Loss: 1.2256, Training accuracy: 58.52,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [71/100],\tTraining Loss: 1.2266, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [72/100],\tTraining Loss: 1.2264, Training accuracy: 58.52,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [73/100],\tTraining Loss: 1.2262, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [74/100],\tTraining Loss: 1.2267, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [75/100],\tTraining Loss: 1.2281, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [76/100],\tTraining Loss: 1.2254, Training accuracy: 58.52,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [77/100],\tTraining Loss: 1.2277, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [78/100],\tTraining Loss: 1.2272, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [79/100],\tTraining Loss: 1.2251, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [80/100],\tTraining Loss: 1.2278, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [81/100],\tTraining Loss: 1.2251, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [82/100],\tTraining Loss: 1.2254, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [83/100],\tTraining Loss: 1.2266, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [84/100],\tTraining Loss: 1.2285, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [85/100],\tTraining Loss: 1.2281, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [86/100],\tTraining Loss: 1.2249, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [87/100],\tTraining Loss: 1.2266, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [88/100],\tTraining Loss: 1.2271, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [89/100],\tTraining Loss: 1.2259, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [90/100],\tTraining Loss: 1.2246, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [91/100],\tTraining Loss: 1.2270, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [92/100],\tTraining Loss: 1.2271, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [93/100],\tTraining Loss: 1.2255, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [94/100],\tTraining Loss: 1.2267, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [95/100],\tTraining Loss: 1.2281, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [96/100],\tTraining Loss: 1.2267, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [97/100],\tTraining Loss: 1.2268, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [98/100],\tTraining Loss: 1.2278, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [99/100],\tTraining Loss: 1.2260, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n",
      "Epoch [100/100],\tTraining Loss: 1.2270, Training accuracy: 58.55,\tTest Loss: 1.3748, Test Accuracy: 53.82,\tLearning rate: 0.00000\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy only on dynamic proteins: 0.3122448979591837\n"
     ]
    }
   ],
   "source": [
    "# Setting seed\n",
    "torch.manual_seed(30026)\n",
    "if(torch.cuda.is_available()):\n",
    "    torch.cuda.manual_seed(30026)\n",
    "    torch.cuda.manual_seed_all(30026)\n",
    "\n",
    "# Run training\n",
    "train_loss, train_accuracy, test_loss, test_accuracy, model = dynamic_training(num_features, num_classes, num_timesteps, embedding_dim, dynamic_params, best_static_model, dataloader_train, dataloader_test, verbose, device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Important remarks can be done on this model: since it is not accounting for temporal relationships between data, it is not performing as well as the other models (as expected). However, it is still learning relatively well. This can be caused by the fact that ESM-embeddings are the most usefull data we have in our dataset, since they are obtained from elaborated architecture pre-trained on milions of proteins."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## K-Fold Cross-Validation on LSTM Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since LSTM is the model which performs best, we run cross validation for its hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Setting seed\n",
    "torch.manual_seed(30026)\n",
    "if(torch.cuda.is_available()):\n",
    "    torch.cuda.manual_seed(30026)\n",
    "    torch.cuda.manual_seed_all(30026)\n",
    "\n",
    "# Choosing hyperparameters\n",
    "parameter_grid = {\n",
    "    \"dropout\": [0.2],\n",
    "    \"learning_rate\": [0.001, 0.01],\n",
    "    \"weight_decay\": [0.01],\n",
    "    \"scheduler\": ['StepLR'],\n",
    "    \"lambda_penalty\": [0, 1e-5, 1e-2],\n",
    "    \"static_learnable\": [True, False],\n",
    "    \"hidden_size\": [32, 64]\n",
    "}\n",
    "\n",
    "# Load best parameters for static model\n",
    "with open('hyperparameters/best_results_static.json', 'r') as file:\n",
    "    best_params_static = json.load(file)\n",
    "\n",
    "\n",
    "best_params, results = k_fold_cross_validation_dynamic(\n",
    "    StaticModelMultibranch,\n",
    "    LSTMDynamicModel,\n",
    "    data,\n",
    "    CrossEntropyWithTemporalSmoothness,\n",
    "    parameter_grid,\n",
    "    best_params_static,\n",
    "    dynamic_params[\"num_epochs\"],\n",
    "    dynamic_params[\"patience\"],\n",
    "    params[\"batch_size\"],\n",
    "    device,\n",
    "    k_folds=3,\n",
    "    seed=seed,\n",
    "    cross_validation_on_loss=True,\n",
    "    verbose=True\n",
    ")\n",
    "\n",
    "results_file = 'hyperparameters/cross_val_results_dynamic.json'\n",
    "with open(results_file, 'w') as f:\n",
    "    json.dump(results, f, indent=4)\n",
    "\n",
    "best_file = 'hyperparameters/best_results_dynamic.json'\n",
    "with open(best_file, 'w') as f:\n",
    "    json.dump(best_params, f, indent=4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Towards Intepretability: Modulable LSTM Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we have selected our best model, we try to interpret the results. Apart from correctly predicting protein locations, the main question of our problem is whether all different data are equally important for this task.\n",
    "\n",
    "Moving in this direction, we make the LSTM Module modulable: it can be trained on the whole dataset or only on part of it (in particular it can decide wheter to take __protein interactions matrix__, __protein concentration levels__ and __static_data__, with all the possible combinations available). As all other hyperparameters choice, refer to the `hyperparameters_cross_val.json` file.\n",
    "\n",
    "You can modify the three parameters there and afterwards run the following cells"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define hyperparameters\n",
    "with open('hyperparameters/hyperparameters_cross_val.json', 'r') as file:\n",
    "    params = json.load(file)\n",
    "\n",
    "dynamic_params = params[\"dynamic_model\"]\n",
    "\n",
    "# Set embedding dimensions based on json parameter\n",
    "if static_params[\"long_embeddings\"]:\n",
    "    embedding_dim = 640\n",
    "else:\n",
    "    embedding_dim = 320"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Setting seed\n",
    "torch.manual_seed(30026)\n",
    "if(torch.cuda.is_available()):\n",
    "    torch.cuda.manual_seed(30026)\n",
    "    torch.cuda.manual_seed_all(30026)\n",
    "\n",
    "# Run training\n",
    "train_loss, train_accuracy, test_loss, test_accuracy, model = dynamic_training(num_features, num_classes, num_timesteps, embedding_dim, dynamic_params, best_static_model, dataloader_train, dataloader_test, verbose, device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Results of these multiple runs are given in detail on the report"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}