train_test.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training and testing on EEG summary-statistics dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib widget\n",
    "# widget requires ipympl\n",
    "\n",
    "# Standard library\n",
    "import importlib\n",
    "import dataclasses\n",
    "from dataclasses import dataclass\n",
    "import time\n",
    "from os import path, makedirs\n",
    "\n",
    "# Nonstandard packages\n",
    "from matplotlib import pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import ipywidgets as widgets\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from torch.utils.data import Dataset, DataLoader, SubsetRandomSampler\n",
    "from torch.optim import Adam, SGD, Rprop\n",
    "\n",
    "from sklearn.model_selection import KFold\n",
    "\n",
    "# Custom files\n",
    "import static_nets\n",
    "import constr_casc\n",
    "import performance_metrics as pm\n",
    "\n",
    "plt.ioff(); # Avoids plot-duplicates in Jupyter when cell execution finishes\n",
    "\n",
    "metaseed = 187598710957 # Seed for generating seeds"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loss function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "bceLogitLossMean = nn.BCEWithLogitsLoss(reduction='mean')\n",
    "# Applies sigmoid internally. More stable than Sigmoid + BCELoss.\n",
    "\n",
    "def loss_accuracy(pred_logits, y):\n",
    "  loss = bceLogitLossMean(pred_logits, y)\n",
    "  pred_class = F.sigmoid(pred_logits).round()\n",
    "  accuracy = torch.mean(pred_class == y, dtype=torch.float32)\n",
    "  return loss.item(), accuracy.item()\n",
    "\n",
    "# Further metrics in performance_metrics.py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load pre-processed data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "class TableDataset(Dataset):\n",
    "  def __init__(self, data, name, outdir='./results/'):\n",
    "    self.data = data\n",
    "    self.n_features = data.shape[1] - 1 # Last column is target\n",
    "    self.n_subjects = data.shape[0] // 6 # 6 samples per subject\n",
    "\n",
    "    # Keep data in order with n_subjects splits, to achieve\n",
    "    # leave-one-subject-out cross validation\n",
    "    self.kf = KFold(n_splits=self.n_subjects, shuffle=False)\n",
    "\n",
    "    self.perfs = {} # KFoldPerformance for each model\n",
    "    self.eval_epochs = {} # Epoch to evaluate each model at\n",
    "\n",
    "    # Try load existing results (avoids having to re-run all models).\n",
    "    self.name = name\n",
    "    self.outdir = outdir\n",
    "    self._outfile = path.join(outdir, f'perf_{name}.csv')\n",
    "    try:\n",
    "      self.perf_table = pd.read_csv(self._outfile, index_col='model')\n",
    "    except FileNotFoundError:\n",
    "      self.perf_table = pd.DataFrame(\n",
    "      index = pd.CategoricalIndex(data=[], categories=[], name='model'),\n",
    "      columns=[\n",
    "        'n_runs', 'n_folds', 'epochs', 'time', 'size', 'train-acc', 'acc',\n",
    "        'prec','sens', 'spec', 'f1'\n",
    "        ])\n",
    "  \n",
    "  def __len__(self):\n",
    "    return self.data.size(0)\n",
    "  \n",
    "  def __getitem__(self, idx):\n",
    "    return (self.data[idx, :-1], self.data[idx, -1:])\n",
    "    # idx is a list of indices, or a single index.\n",
    "    # ':' after -1 preserves tensor dimension\n",
    "  \n",
    "  def split(self): #KFold (train,test) indices\n",
    "    return self.kf.split(self.data)\n",
    "\n",
    "  def save(self):\n",
    "    makedirs(self.outdir, exist_ok=True)\n",
    "    self.perf_table.to_csv(self._outfile, index=True)\n",
    "\n",
    "  def recompute_model_perfs(self, m_ids=None):\n",
    "    # Compute/store perfomance summary for each model.\n",
    "    # Assumes each m_id has a KFoldPerformance in self.perfs.\n",
    "    # Epoch defaults to last epoch if not in m_epoch_dict, or > last epoch.\n",
    "    if m_ids is None:\n",
    "      m_ids = self.perfs.keys()\n",
    "\n",
    "    for m_id in m_ids:\n",
    "      if m_id not in self.perfs:\n",
    "        print(f\"Warning: {self.name} dataset asked to recompute performance for model {m_id}, but no KFoldPerformance is stored for this model\")\n",
    "        continue\n",
    "      perf = self.perfs[m_id]\n",
    "      tm = perf.train_metrics\n",
    "\n",
    "      #Assumes tm.epoch is ascending. Argmax returns first index.\n",
    "      max_epoch = tm.epoch[-1].item()\n",
    "      eval_epoch = min(self.eval_epochs.get(m_id, max_epoch), max_epoch)\n",
    "      record_idx = torch.argmax((tm.epoch >= eval_epoch).to(dtype=torch.int64))\n",
    "\n",
    "      testMet = pm.PredictionMetrics.from_confusion(perf.confusions[record_idx,:,:])\n",
    "\n",
    "      self.perf_table.loc[m_id] = {\n",
    "        'n_runs': perf.n_runs,\n",
    "        'n_folds': perf.n_folds,\n",
    "        'epochs': eval_epoch,\n",
    "        'time': tm.time[record_idx].item(),\n",
    "        'size': (tm.hidden_neurons[record_idx].item() if 'hidden_neurons' in tm.names else None), # Only cascade networks record size, otherwise blank\n",
    "        'train-acc': tm.train_accuracy[record_idx].item(),\n",
    "        'acc': testMet.accuracy,\n",
    "        'prec': testMet.precision,\n",
    "        'sens': testMet.sensitivity,\n",
    "        'spec': testMet.specificity,\n",
    "        'f1': testMet.f1\n",
    "      }\n",
    "\n",
    "    self.save()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 14 brain regions * 15 stats -> 210 features (or subset thereof)\n",
    "Xy_all = torch.load('./data/features_processed_all.pt') #float32; see data_prep.ipynb\n",
    "eeg_all = TableDataset(Xy_all, 'all')\n",
    "\n",
    "Xy_no_outliers = torch.load('./data/features_processed_no_outliers.pt')\n",
    "eeg_no_outliers = TableDataset(Xy_no_outliers, 'no_outliers')\n",
    "\n",
    "Xy_f7 = torch.load('./data/features_processed_f7.pt')\n",
    "eeg_f7 = TableDataset(Xy_f7, 'f7')\n",
    "\n",
    "#Xy_f3 = torch.load('./data/features_processed_f3.pt')\n",
    "#eeg_f3 = TableDataset(Xy_f3, 'f3')\n",
    "\n",
    "#Xy_rahman = torch.load('./data/features_processed_rahman.pt')\n",
    "#eeg_rahman = TableDataset(Xy_rahman, 'rahman')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Manually-chosen epoch at which to evaluate each model,\n",
    "# based on output graphs below (after running for each dataset)\n",
    "# This is so we can pick the best performance for each model,\n",
    "# (in case it declines) and compare the time taken to achieve.\n",
    "\n",
    "eeg_f7.eval_epochs.update({\n",
    "  'linGDlr0.1': 500,\n",
    "  #'linSGDlr0.01bs12': 390,\n",
    "  #'linAdam': 500,\n",
    "  'linAdamLr0.001bs12': 670,\n",
    "  #'linAdamlr0.01': 400,\n",
    "  'linRprop': 75,\n",
    "  #'h30GD': 800,\n",
    "  'h30GDlr0.1': 1000,\n",
    "  #'h30GDlr0.001bs12': 800,\n",
    "  'h30AdamBs12': 500,\n",
    "  #'h30AdamLr0.01': 100,\n",
    "  'hid30Rprop': 110\n",
    "})\n",
    "\n",
    "eeg_no_outliers.eval_epochs.update({\n",
    "  'linGDlr0.1': 150,\n",
    "  #'linSGDlr0.01bs12': 100,\n",
    "  #'linAdam': 499,\n",
    "  'linAdamLr0.001bs12': 650,\n",
    "  #'linAdamlr0.01': 250,\n",
    "  'linRprop': 50,\n",
    "  #'h30GD': 400,\n",
    "  'h30GDlr0.1': 300,\n",
    "  #'h30GDlr0.001bs12': 400\n",
    "  'h30AdamBs12': 1400,\n",
    "  'hid30Rprop': 50\n",
    "})\n",
    "# best models: linGDlr0.1, linRprop (best, and super fast), all h30 variants but particularly h30GDlr0.1\n",
    "# small batch improved linAdam considerably, at no time cost.\n",
    "\n",
    "eeg_all.eval_epochs.update({\n",
    "  'linGDlr0.1': 140,\n",
    "  #'linSGDlr0.01bs12': #80,\n",
    "  ##'linAdam': 500,\n",
    "  'linAdamLr0.001bs12': 90,\n",
    "  ##'linAdamlr0.01': 400,\n",
    "  'linRprop': 100,\n",
    "  #'h30GD': -1,\n",
    "  'h30GDlr0.1': 330,\n",
    "  ##'h30GDlr0.001bs12': 800,\n",
    "  'h30AdamBs12': 130,\n",
    "  ##'h30AdamLr0.01': 50,\n",
    "  'hid30Rprop': 50,\n",
    "})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train standard feedforward networks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Model description\n",
    "@dataclass\n",
    "class StaticModel:\n",
    "  desc: str\n",
    "\n",
    "  initNet: nn.Module # will be passed (n_features, **net_kw)\n",
    "  net_kw: dict\n",
    "  initOpt: torch.optim.Optimizer # passed (net.parameters(), **opt_kw)\n",
    "  opt_kw: dict\n",
    "  #initScheduler: torch.optim.lr_scheduler._LRScheduler = None\n",
    "  #scheduler_kwargs: dict = None\n",
    "\n",
    "  batch_size: int = 1000 # >>> n_samples, i.e. no batching by default\n",
    "  n_epochs: int = 1001 # Required epochs varies with e.g. network size\n",
    "  \n",
    "  performance: pm.KFoldPerformance = None\n",
    "\n",
    "  def __str__(self):\n",
    "    return f\"{self.desc} ({', '.join([f'{k}={v}' for k,v in self.opt_kw.items()])})\"\n",
    "\n",
    "  def clone(self, **kwargs):\n",
    "    return dataclasses.replace(self,**kwargs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define a set of static-architecture models to test\n",
    "# (network architecture, optimiser, hyperparameters)\n",
    "importlib.reload(static_nets)\n",
    "\n",
    "bc = static_nets.BinaryClassifier # defaults to a no-hidden-layer network.\n",
    "rprop_kw = {'etas':(0.5, 1.2), 'step_sizes':(1e-06, 50)}\n",
    "\n",
    "# Logistic regression base models\n",
    "linGD = StaticModel(\"Linear Full GD\", bc, {}, SGD, {\"lr\": 0.01}, n_epochs=1600)\n",
    "linAdam = StaticModel(\"Linear Adam\", bc, {}, Adam, {\"lr\": 0.001}, n_epochs=1600)\n",
    "linRprop = StaticModel(\"Linear Rprop\", bc, {}, Rprop, {'lr':0.01, **rprop_kw}, n_epochs=500)\n",
    "# f7 only needs 800 epochs for logistic models and 1000 for hid30; trying more for eeg_all\n",
    "\n",
    "# Single hidden layer with 30 units, as in Brain Melody Informatics paper.\n",
    "hid30 = lambda n_feat, **kwargs: bc(n_feat, hidden_sizes=[30], **kwargs)\n",
    "hid30GD = StaticModel(\"HL30 ReLU GD\", hid30, {}, SGD, {\"lr\": 0.01}, n_epochs=2000)\n",
    "\n",
    "# Comment-out variants from the list to avoid running them.\n",
    "\n",
    "# Hyperparameter variants to run\n",
    "staticModels = { #id: StaticModel\n",
    "  # Logistic regression (i.e. no hidden layers)\n",
    "  'linGDlr0.1': linGD.clone(opt_kw={\"lr\": 0.1}),\n",
    "  'linAdamLr0.001bs12': linAdam.clone(batch_size=12, desc=\"Linear Adam (batch=12)\"),\n",
    "  'linRprop': linRprop,\n",
    "  # H30 (like Brain Melody Informatics paper)\n",
    "  'h30GDlr0.1': hid30GD.clone(opt_kw={\"lr\": 0.1}),\n",
    "  'h30AdamBs12': hid30GD.clone(initOpt=Adam, opt_kw={\"lr\": 0.001}, batch_size=12, desc=\"HL30 Adam (batch=12)\"),\n",
    "  'hid30Rprop': hid30GD.clone(initOpt=Rprop, opt_kw={'lr':0.01, **rprop_kw}, desc=\"HL30 ReLU Rprop\", n_epochs=500),\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "466abf01436f49059fcd8d9155cedcac",
       "version_major": 2,
       "version_minor": 0
      },
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAyAAAAEsCAYAAAA7Ldc6AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAABOkElEQVR4nO3de1xU1fo/8M+AMIiACggIoihe8QIlgnclISLD1EzScxS1tFOSJl0UTfBS0tE0SjFKReqEQZpyOmEkEkQqRV6wvKCieBcUTVFQGJn1+8Mf820aQMBhD7A/79eLV+41a+/9PEPO8pm919oKIYQAERERERGRBIwMHQAREREREckHCxAiIiIiIpIMCxAiIiIiIpIMCxAiIiIiIpIMCxAiIiIiIpIMCxAiokZg2rRpcHFxqde+S5YsgUKh0G9AREREDYQFCBFRDRQKRa1+MjIyDB2qQUybNk3rfbCysoK7uztWr16NsrIyQ4dHRESNkILPASEiqt6XX36ptf3FF18gNTUV//nPf7Ta/fz8YG9vX+/zqFQqqNVqKJXKOu97//593L9/H2ZmZvU+f31NmzYNCQkJ2LhxIwDg5s2b+Oabb5CRkYGgoCAkJCRIHhMRETVuLECIiOogJCQE0dHReNhHZ2lpKczNzSWKynCmTZuGbdu24c6dO5o2tVoNb29v7N+/H5cuXYKjo6POfkII3Lt3Dy1btpQkTrn8PoiImgLegkVE9IhGjhyJPn364MCBAxg+fDjMzc2xcOFCAMB///tfjB49Go6OjlAqlXB1dcXy5ctRUVGhdYy/zwE5e/YsFAoFPvjgA3z22WdwdXWFUqnEgAED8Ntvv2ntW9UcEIVCgZCQECQlJaFPnz5QKpXo3bs3UlJSdOLPyMiAp6cnzMzM4Orqik8//fSR5pUYGRlh5MiRmjwAwMXFBc888wx++OEHeHp6omXLlvj0008BAGfOnMHzzz8Pa2trmJubY+DAgUhOTtY57rlz5zBmzBi0atUKdnZ2mDdvHn744QedW+Bq+n2UlZUhIiICXbt2hVKphLOzM95++22d28VSU1MxdOhQtGnTBhYWFujRo4fmGJXWrl2L3r17w9zcHG3btoWnpye2bNlSr/eMiEhOWhg6ACKi5uD69esICAjACy+8gH/+85+a27Hi4uJgYWGB0NBQWFhY4Mcff0R4eDiKi4uxatWqhx53y5YtuH37Nl5++WUoFAqsXLkS48ePx5kzZ2BiYlLjvnv27MH27dvx6quvwtLSEh9//DGee+45nD9/HjY2NgCAQ4cO4amnnkL79u2xdOlSVFRUYNmyZWjXrt0jvR+nT58GAM15AODEiROYNGkSXn75ZcycORM9evRAYWEhBg8ejNLSUsyZMwc2Njb4/PPPMWbMGGzbtg3jxo0DAJSUlOCJJ57AlStXMHfuXDg4OGDLli1IT0+v8vxV/T7UajXGjBmDPXv2YNasWejVqxf++OMPfPjhhzh58iSSkpIAAEePHsUzzzyDfv36YdmyZVAqlcjLy8PevXs1x9+wYQPmzJmDCRMmYO7cubh37x5+//13/Prrr5g8efIjvXdERM2eICKiWps9e7b4+0fniBEjBAARExOj07+0tFSn7eWXXxbm5ubi3r17mrbg4GDRqVMnzXZ+fr4AIGxsbMSNGzc07f/9738FAPG///1P0xYREaETEwBhamoq8vLyNG2HDx8WAMTatWs1bYGBgcLc3FxcunRJ03bq1CnRokULnWNWJTg4WLRq1Upcu3ZNXLt2TeTl5YkVK1YIhUIh+vXrp+nXqVMnAUCkpKRo7f/6668LAOLnn3/WtN2+fVt07txZuLi4iIqKCiGEEKtXrxYARFJSkqbf3bt3Rc+ePQUAkZ6ermmv7vfxn//8RxgZGWmdSwghYmJiBACxd+9eIYQQH374oQAgrl27Vm3ezz77rOjdu/dD3x8iItLFW7CIiPRAqVRi+vTpOu1/neNw+/ZtFBUVYdiwYSgtLUVubu5DjxsUFIS2bdtqtocNGwbgwW1LD+Pr6wtXV1fNdr9+/WBlZaXZt6KiArt378bYsWO15ml07doVAQEBDz1+pZKSErRr1w7t2rVD165dsXDhQgwaNAg7duzQ6te5c2f4+/trte3cuRNeXl4YOnSops3CwgKzZs3C2bNncezYMQBASkoKnJycMGbMGE0/MzMzzJw5s8qYqvp9bN26Fb169ULPnj1RVFSk+XniiScAQHM1pU2bNgAe3D6nVqurPH6bNm1w8eJFndvhiIjo4ViAEBHpgZOTE0xNTXXajx49inHjxqF169awsrJCu3bt8M9//hMAcOvWrYcet2PHjlrblcXIn3/+Wed9K/ev3Pfq1au4e/cuunbtqtOvqrbqmJmZITU1FampqcjMzMSFCxewd+9edOnSRatf586ddfY9d+4cevToodPeq1cvzeuV/3V1ddWZl1JdnFX9Pk6dOoWjR49qiqXKn+7duwN48H4AD4q+IUOG4KWXXoK9vT1eeOEFfP3111rFyPz582FhYQEvLy9069YNs2fP1rpFi4iIqsc5IEREelDVak43b97EiBEjYGVlhWXLlsHV1RVmZmY4ePAg5s+fX+23639lbGxcZbuoxQKGj7JvXRgbG8PX1/eh/aRa8aq6c6nVavTt2xdr1qypch9nZ2fNvpmZmUhPT0dycjJSUlKQmJiIJ554Art27YKxsTF69eqFEydO4LvvvkNKSgq++eYbrF+/HuHh4Vi6dGmD5kZE1NSxACEiaiAZGRm4fv06tm/fjuHDh2va8/PzDRjV/7Gzs4OZmRny8vJ0XquqrSF06tQJJ06c0GmvvD2tU6dOmv8eO3YMQgitqyB1idPV1RWHDx/GqFGjHrrCl5GREUaNGoVRo0ZhzZo1WLFiBRYtWoT09HRNsdWqVSsEBQUhKCgI5eXlGD9+PN577z2EhYUZ5JksRERNBW/BIiJqIJVXIP56xaG8vBzr1683VEhaKq9cJCUl4fLly5r2vLw8fP/995LE8PTTTyM7OxtZWVmatpKSEnz22WdwcXGBm5sbAMDf3x+XLl3Ct99+q+l37949bNiwodbnmjhxIi5dulTlPnfv3kVJSQkA4MaNGzqve3h4AIBmud7r169rvW5qago3NzcIIaBSqWodExGRHPEKCBFRAxk8eDDatm2L4OBgzJkzBwqFAv/5z3/0fgvUo1iyZAl27dqFIUOG4JVXXkFFRQXWrVuHPn36ICcnp8HPv2DBAnz11VcICAjAnDlzYG1tjc8//xz5+fn45ptvYGT04Huyl19+GevWrcOkSZMwd+5ctG/fHvHx8ZorDbV5ZsmUKVPw9ddf41//+hfS09MxZMgQVFRUIDc3F19//bXmGSXLli1DZmYmRo8ejU6dOuHq1atYv349OnTooJks/+STT8LBwQFDhgyBvb09jh8/jnXr1mH06NGwtLRsuDeMiKgZYAFCRNRAbGxs8N133+GNN97AO++8g7Zt2+Kf//wnRo0apbMalKH0798f33//Pd58800sXrwYzs7OWLZsGY4fP16rVboelb29Pfbt24f58+dj7dq1uHfvHvr164f//e9/GD16tKZf5TNUXnvtNXz00UewsLDA1KlTMXjwYDz33HO1uuXJyMgISUlJ+PDDD/HFF19gx44dMDc3R5cuXTB37lzNZPQxY8bg7NmziI2NRVFREWxtbTFixAgsXboUrVu3BvCgIIqPj8eaNWtw584ddOjQAXPmzME777zTMG8UEVEzohCN6as4IiJqFMaOHYujR4/i1KlThg6lRlFRUZg3bx4uXrwIJycnQ4dDRES1wDkgREQyd/fuXa3tU6dOYefOnRg5cqRhAqrG3+O8d+8ePv30U3Tr1o3FBxFRE8JbsIiIZK5Lly6YNm0aunTpgnPnzuGTTz6Bqakp3n77bUOHpmX8+PHo2LEjPDw8cOvWLXz55ZfIzc1FfHy8oUMjIqI6YAFCRCRzTz31FL766isUFBRAqVRi0KBBWLFiBbp162bo0LT4+/tj48aNiI+PR0VFBdzc3JCQkICgoCBDh0ZERHXAOSBERERERCQZzgEhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhIiIiIiLJsAAhakBxcXFQKBTYv3+/oUMhIqI6Wr9+PRQKBby9vQ0dClGzwgKEiIiIqArx8fFwcXFBdnY28vLyDB0OUbPBAoSIiIjob/Lz87Fv3z6sWbMG7dq1Q3x8vKFDqlJJSYmhQyCqMxYgRAZ26NAhBAQEwMrKChYWFhg1ahR++eUXrT4qlQpLly5Ft27dYGZmBhsbGwwdOhSpqamaPgUFBZg+fTo6dOgApVKJ9u3b49lnn8XZs2clzoiIqOmLj49H27ZtMXr0aEyYMKHKAuTmzZuYN28eXFxcoFQq0aFDB0ydOhVFRUWaPvfu3cOSJUvQvXt3mJmZoX379hg/fjxOnz4NAMjIyIBCoUBGRobWsc+ePQuFQoG4uDhN27Rp02BhYYHTp0/j6aefhqWlJf7xj38AAH7++Wc8//zz6NixI5RKJZydnTFv3jzcvXtXJ+7c3FxMnDgR7dq1Q8uWLdGjRw8sWrQIAJCeng6FQoEdO3bo7LdlyxYoFApkZWXV+f0k+qsWhg6ASM6OHj2KYcOGwcrKCm+//TZMTEzw6aefYuTIkfjpp5809x0vWbIEkZGReOmll+Dl5YXi4mLs378fBw8ehJ+fHwDgueeew9GjR/Haa6/BxcUFV69eRWpqKs6fPw8XFxcDZklE1PTEx8dj/PjxMDU1xaRJk/DJJ5/gt99+w4ABAwAAd+7cwbBhw3D8+HHMmDEDjz/+OIqKivDtt9/i4sWLsLW1RUVFBZ555hmkpaXhhRdewNy5c3H79m2kpqbiyJEjcHV1rXNc9+/fh7+/P4YOHYoPPvgA5ubmAICtW7eitLQUr7zyCmxsbJCdnY21a9fi4sWL2Lp1q2b/33//HcOGDYOJiQlmzZoFFxcXnD59Gv/73//w3nvvYeTIkXB2dkZ8fDzGjRun8564urpi0KBBj/DOEgEQRNRgNm/eLACI3377rcrXx44dK0xNTcXp06c1bZcvXxaWlpZi+PDhmjZ3d3cxevToas/z559/CgBi1apV+gueiEim9u/fLwCI1NRUIYQQarVadOjQQcydO1fTJzw8XAAQ27dv19lfrVYLIYSIjY0VAMSaNWuq7ZOeni4AiPT0dK3X8/PzBQCxefNmTVtwcLAAIBYsWKBzvNLSUp22yMhIoVAoxLlz5zRtw4cPF5aWllptf41HCCHCwsKEUqkUN2/e1LRdvXpVtGjRQkREROich6iueAsWkYFUVFRg165dGDt2LLp06aJpb9++PSZPnow9e/aguLgYANCmTRscPXoUp06dqvJYLVu2hKmpKTIyMvDnn39KEj8RUXMVHx8Pe3t7+Pj4AAAUCgWCgoKQkJCAiooKAMA333wDd3d3nasElf0r+9ja2uK1116rtk99vPLKKzptLVu21Py5pKQERUVFGDx4MIQQOHToEADg2rVryMzMxIwZM9CxY8dq45k6dSrKysqwbds2TVtiYiLu37+Pf/7zn/WOm6gSCxAiA7l27RpKS0vRo0cPndd69eoFtVqNCxcuAACWLVuGmzdvonv37ujbty/eeust/P7775r+SqUS//73v/H999/D3t4ew4cPx8qVK1FQUCBZPkREzUFFRQUSEhLg4+OD/Px85OXlIS8vD97e3igsLERaWhoA4PTp0+jTp0+Nxzp9+jR69OiBFi30d8d7ixYt0KFDB5328+fPY9q0abC2toaFhQXatWuHESNGAABu3boFADhz5gwAPDTunj17YsCAAVrzXuLj4zFw4EB07dpVX6mQjLEAIWoChg8fjtOnTyM2NhZ9+vTBxo0b8fjjj2Pjxo2aPq+//jpOnjyJyMhImJmZYfHixejVq5fmmy8iInq4H3/8EVeuXEFCQgK6deum+Zk4cSIA6H01rOquhFReafk7pVIJIyMjnb5+fn5ITk7G/PnzkZSUhNTUVM0EdrVaXee4pk6dip9++gkXL17E6dOn8csvv/DqB+kNJ6ETGUi7du1gbm6OEydO6LyWm5sLIyMjODs7a9qsra0xffp0TJ8+HXfu3MHw4cOxZMkSvPTSS5o+rq6ueOONN/DGG2/g1KlT8PDwwOrVq/Hll19KkhMRUVMXHx8POzs7REdH67y2fft27NixAzExMXB1dcWRI0dqPJarqyt+/fVXqFQqmJiYVNmnbdu2AB6sqPVX586dq3XMf/zxB06ePInPP/8cU6dO1bT/daVEAJrbfR8WNwC88MILCA0NxVdffYW7d+/CxMQEQUFBtY6JqCa8AkJkIMbGxnjyySfx3//+V2up3MLCQmzZsgVDhw6FlZUVAOD69eta+1pYWKBr164oKysDAJSWluLevXtafVxdXWFpaanpQ0RENbt79y62b9+OZ555BhMmTND5CQkJwe3bt/Htt9/iueeew+HDh6tcrlYIAeDB6oRFRUVYt25dtX06deoEY2NjZGZmar2+fv36WsdtbGysdczKP3/00Uda/dq1a4fhw4cjNjYW58+frzKeSra2tggICMCXX36J+Ph4PPXUU7C1ta11TEQ14RUQIgnExsYiJSVFp33JkiVITU3F0KFD8eqrr6JFixb49NNPUVZWhpUrV2r6ubm5YeTIkejfvz+sra2xf/9+bNu2DSEhIQCAkydPYtSoUZg4cSLc3NzQokUL7NixA4WFhXjhhRcky5OIqCn79ttvcfv2bYwZM6bK1wcOHKh5KOGWLVuwbds2PP/885gxYwb69++PGzdu4Ntvv0VMTAzc3d0xdepUfPHFFwgNDUV2djaGDRuGkpIS7N69G6+++iqeffZZtG7dGs8//zzWrl0LhUIBV1dXfPfdd7h69Wqt4+7ZsydcXV3x5ptv4tKlS7CyssI333xT5aIkH3/8MYYOHYrHH38cs2bNQufOnXH27FkkJycjJydHq+/UqVMxYcIEAMDy5ctr/0YSPYwhl+Aiau4ql+Gt7ufChQvi4MGDwt/fX1hYWAhzc3Ph4+Mj9u3bp3Wcd999V3h5eYk2bdqIli1bip49e4r33ntPlJeXCyGEKCoqErNnzxY9e/YUrVq1Eq1btxbe3t7i66+/NkTaRERNUmBgoDAzMxMlJSXV9pk2bZowMTERRUVF4vr16yIkJEQ4OTkJU1NT0aFDBxEcHCyKioo0/UtLS8WiRYtE586dhYmJiXBwcBATJkzQWn792rVr4rnnnhPm5uaibdu24uWXXxZHjhypchneVq1aVRnXsWPHhK+vr7CwsBC2trZi5syZ4vDhwzrHEEKII0eOiHHjxok2bdoIMzMz0aNHD7F48WKdY5aVlYm2bduK1q1bi7t379byXSR6OIUQf7vmRkRERESyd//+fTg6OiIwMBCbNm0ydDjUjHAOCBERERHpSEpKwrVr17QmthPpg2wLkO+++w49evRAt27dtJYyJSIiIpKzX3/9FRs2bEBoaCgee+wxzfNEiPRFlrdg3b9/H25ubkhPT0fr1q3Rv39/7Nu3DzY2NoYOjYiIiMigpk2bhi+//BIeHh6Ii4t76IMLiepKlldAsrOz0bt3bzg5OcHCwgIBAQHYtWuXocMiIiIiMri4uDjcv38f+/fvZ/FBDaJJFiCZmZkIDAyEo6MjFAoFkpKSdPpER0fDxcUFZmZm8Pb2RnZ2tua1y5cvw8nJSbPt5OSES5cuSRE6EREREZGsNckCpKSkBO7u7lU+pRQAEhMTERoaioiICBw8eBDu7u7w9/ev05raRERERESkf03yQYQBAQEICAio9vU1a9Zg5syZmD59OgAgJiYGycnJiI2NxYIFC+Do6Kh1xePSpUvw8vKq9nhlZWVaT5NWq9W4ceMGbGxsoFAo9JAREZHhCCFw+/ZtODo6wsioSX4vZRBqtRqXL1+GpaUlxwIiavKkHAuaZAFSk/Lychw4cABhYWGaNiMjI/j6+iIrKwsA4OXlhSNHjuDSpUto3bo1vv/+eyxevLjaY0ZGRmLp0qUNHjsRkSFduHABHTp0MHQYTcbly5fh7Oxs6DCIiPRKirGg2RUgRUVFqKiogL29vVa7vb09cnNzAQAtWrTA6tWr4ePjA7VajbfffrvGFbDCwsIQGhqq2b516xY6duyIkydPwtraumESaYRUKhXS09Ph4+MDExMTQ4cjGTnmLcecAfnmfePGDXTv3h2WlpaGDqVJqXy/8vPzZTMWqFQq7Nq1C08++aSs/o4wb+YtBzdu3EDnzp0lGQuaXQFSW2PGjMGYMWNq1VepVEKpVOq0W1tby2rpXpVKBXNzc9jY2MjqL6Qc85ZjzoB8867E24jqpvL9srS0hJWVlYGjkUbl3xErKytZ/R1h3sxbDlQqFQBpxoJmd7Ovra0tjI2NUVhYqNVeWFgIBwcHA0VFRERERERAMyxATE1N0b9/f6SlpWna1Go10tLSMGjQIANGRkRERERETfIWrDt37iAvL0+znZ+fj5ycHFhbW6Njx44IDQ1FcHAwPD094eXlhaioKJSUlGhWxSIiIiIiIsNokgXI/v374ePjo9munCAeHByMuLg4BAUF4dq1awgPD0dBQQE8PDyQkpKiMzGdiIiIiIik1SQLkJEjR0IIUWOfkJAQhISESBQRERERERHVRrObA0JERERERI0XCxAiIiIiIpIMCxAiIiIiIpIMCxAiImrSMjMzERgYCEdHRygUCiQlJT10n4yMDDz++ONQKpXo2rUr4uLiGjxOIiJ6gAUIERE1aSUlJXB3d0d0dHSt+ufn52P06NHw8fFBTk4OXn/9dbz00kv44YcfGjhSIiICmugqWERERJUCAgIQEBBQ6/4xMTHo3LkzVq9eDQDo1asX9uzZgw8//BD+/v4NFSYREf1/vAJCRESykpWVBV9fX602f39/ZGVlGSgiIiJ54RUQIiKSlYKCAp0H09rb26O4uBh3795Fy5Ytq9yvrKwMZWVlmu3i4mIAgEqlgkqlariAG5HKPOWSbyXmzbzlQMp8WYAQERHVQmRkJJYuXarTnp6eDnNzcwNEZDipqamGDsEgmLe8yC3v0tJSyc7FAoSIiGTFwcEBhYWFWm2FhYWwsrKq9uoHAISFhSE0NFSzXVxcDGdnZ/j4+MDGxqbB4m1MVCoVUlNT4efnBxMTE0OHIxnmzbzl4Pr165KdS7YFyLhx45CRkYFRo0Zh27Zthg6HiIgkMmjQIOzcuVOrLTU1FYMGDapxP6VSCaVSqdNuYmIiq3+kAPLMGWDeciO3vKXMVbaT0OfOnYsvvvjC0GEQEdEjunPnDnJycpCTkwPgwTK7OTk5OH/+PIAHVy6mTp2q6f+vf/0LZ86cwdtvv43c3FysX78eX3/9NebNm2eI8ImIZEe2BcjIkSNhaWlp6DCIiOgR7d+/H4899hgee+wxAEBoaCgee+wxhIeHAwCuXLmiKUYAoHPnzkhOTkZqairc3d2xevVqbNy4kUvwEhFJpFEWILV5qm10dDRcXFxgZmYGb29vZGdnSx8oEREZ3MiRIyGE0PmpfLp5XFwcMjIydPY5dOgQysrKcPr0aUybNk3yuImI5KpRFiAPe6ptYmIiQkNDERERgYMHD8Ld3R3+/v64evWqpo+Hhwf69Omj83P58mWp0iAiIiIior9plJPQH/ZU2zVr1mDmzJmYPn06gAdPtU1OTkZsbCwWLFgAAJp7gfWBa78/IPd1seWUtxxzBpg3ERGRFBplAVKT8vJyHDhwAGFhYZo2IyMj+Pr6NthTbLn2uza5rYtdSY55yzFnQH55S7n2OxERUZMrQIqKilBRUVHlU2xzc3NrfRxfX18cPnwYJSUl6NChA7Zu3VrtEoxc+/0Bua6LLce85ZgzIN+8pVz7nYiIqMkVIPqye/fuWvfl2u/amLd8yDFnQH55yylXIiIyvEY5Cb0mtra2MDY2rvIptg4ODgaKioiIiIiIaqPJFSCmpqbo378/0tLSNG1qtRppaWkPfYotEREREREZVqO8BevOnTvIy8vTbFc+1dba2hodO3ZEaGgogoOD4enpCS8vL0RFRaGkpESzKhYRERERETVOjbIA2b9/P3x8fDTblRPAg4ODERcXh6CgIFy7dg3h4eEoKCiAh4cHUlJSdCamExERERFR49IoC5DKp9rWJCQkBCEhIRJFRERERERE+tDk5oAQEREREVHTxQKEiIiIiIgkwwKEiIiIiIgkwwKEiIiIiIgkwwKEiIiIiIgkwwKEiIiIiIgkwwKEiIiIiIgkwwKEiIiIiIgkwwKEiIiIiIgkwwKEiIiIiIgkI8sC5MKFCxg5ciTc3NzQr18/bN261dAhERERERHJgiwLkBYtWiAqKgrHjh3Drl278Prrr6OkpMTQYRER0SOIjo6Gi4sLzMzM4O3tjezs7Br7R0VFoUePHmjZsiWcnZ0xb9483Lt3T6JoiYjkS5YFSPv27eHh4QEAcHBwgK2tLW7cuGHYoIiIqN4SExMRGhqKiIgIHDx4EO7u7vD398fVq1er7L9lyxYsWLAAEREROH78ODZt2oTExEQsXLhQ4siJiOSnURYgmZmZCAwMhKOjIxQKBZKSknT61PWbruocOHAAFRUVcHZ2fsSoiYjIUNasWYOZM2di+vTpcHNzQ0xMDMzNzREbG1tl/3379mHIkCGYPHkyXFxc8OSTT2LSpEn1HkuIiKj2Whg6gKqUlJTA3d0dM2bMwPjx43Ver/ymKyYmBt7e3oiKioK/vz9OnDgBOzs7AICHhwfu37+vs++uXbvg6OgIALhx4wamTp2KDRs2NGxCRETUYMrLy3HgwAGEhYVp2oyMjODr64usrKwq9xk8eDC+/PJLZGdnw8vLC2fOnMHOnTsxZcqUas9TVlaGsrIyzXZxcTEAQKVSQaVS6Smbxq0yT7nkW4l5M285kDLfRlmABAQEICAgoNrX//pNFwDExMQgOTkZsbGxWLBgAQAgJyenxnOUlZVh7NixWLBgAQYPHqy32ImISFpFRUWoqKiAvb29Vru9vT1yc3Or3Gfy5MkoKirC0KFDIYTA/fv38a9//avGW7AiIyOxdOlSnfb09HSYm5s/WhJNTGpqqqFDMAjmLS9yy7u0tFSyczXKAqQm9fmm6++EEJg2bRqeeOKJGr/tqsRvvR6Q+zcCcspbjjkDzFtOMjIysGLFCqxfvx7e3t7Iy8vD3LlzsXz5cixevLjKfcLCwhAaGqrZLi4uhrOzM3x8fGBjYyNV6AalUqmQmpoKPz8/mJiYGDocyTBv5i0H169fl+xcTa4Aqc83XX+3d+9eJCYmol+/fpr5Jf/5z3/Qt2/fKvvzWy9tcvtGoJIc85ZjzoD88pbyW6+GYGtrC2NjYxQWFmq1FxYWwsHBocp9Fi9ejClTpuCll14CAPTt2xclJSWYNWsWFi1aBCMj3SmSSqUSSqVSp93ExERW/0gB5JkzwLzlRm55S5lrkytA9GHo0KFQq9W17s9vvR6Q6zcCcsxbjjkD8s1bym+9GoKpqSn69++PtLQ0jB07FgCgVquRlpaGkJCQKvcpLS3VKTKMjY0BPLhKTkREDafJFSD1+abrUfFbL23MWz7kmDMgv7ybQ66hoaEIDg6Gp6cnvLy8EBUVhZKSEs1cwalTp8LJyQmRkZEAgMDAQKxZswaPPfaY5hasxYsXIzAwUFOIEBFRw2hyBUh9vukiIqLmLSgoCNeuXUN4eDgKCgrg4eGBlJQUze2658+f17ri8c4770ChUOCdd97BpUuX0K5dOwQGBuK9994zVApERLLRKAuQO3fuIC8vT7Odn5+PnJwcWFtbo2PHjg/9pouIiOQnJCSk2i+iMjIytLZbtGiBiIgIRERESBAZERH9VaMsQPbv3w8fHx/NduX8i+DgYMTFxT30my4iIiIiImqcGmUBMnLkyIdOAqzpmy4iIiIiImqcdNcZJCIiIiIiaiAsQIiISHIuLi5YtmwZzp8/b+hQiIhIYixAiIhIcq+//jq2b9+OLl26wM/PDwkJCSgrKzN0WEREJAEWIEREJLnXX38dOTk5yM7ORq9evfDaa6+hffv2CAkJwcGDBw0dHhERNSAWIEREZDCPP/44Pv74Y1y+fBkRERHYuHEjBgwYAA8PD8TGxvKp5EREzVCjXAWLiIjkQaVSYceOHdi8eTNSU1MxcOBAvPjii7h48SIWLlyI3bt3Y8uWLYYOk4iI9IgFCBERSe7gwYPYvHkzvvrqKxgZGWHq1Kn48MMP0bNnT02fcePGYcCAAQaMkoiIGgILECIiktyAAQPg5+eHTz75BGPHjoWJiYlOn86dO+OFF14wQHRERNSQWIAQEZHkzpw5g06dOtXYp1WrVti8ebNEERERkVQ4CZ2IiCR39epV/Prrrzrtv/76K/bv32+AiIiISCqyLEBu3rwJT09PeHh4oE+fPtiwYYOhQyIikpXZs2fjwoULOu2XLl3C7NmzDRARERFJRZa3YFlaWiIzMxPm5uYoKSlBnz59MH78eNjY2Bg6NCIiWTh27Bgef/xxnfbHHnsMx44dM0BEREQkFVleATE2Noa5uTkAoKysDEIIrjVPRCQhpVKJwsJCnfYrV66gRQtZfjdGRCQbjbIAyczMRGBgIBwdHaFQKJCUlKTTJzo6Gi4uLjAzM4O3tzeys7PrdI6bN2/C3d0dHTp0wFtvvQVbW1s9RU9ERA/z5JNPIiwsDLdu3dK03bx5EwsXLoSfn58BIyMioobWKAuQkpISuLu7Izo6usrXExMTERoaioiICBw8eBDu7u7w9/fH1atXNX0q53f8/efy5csAgDZt2uDw4cPIz8/Hli1bqvwmjoiIGsYHH3yACxcuoFOnTvDx8YGPjw86d+6MgoICrF692tDhERFRA2qU17kDAgIQEBBQ7etr1qzBzJkzMX36dABATEwMkpOTERsbiwULFgAAcnJyanUue3t7uLu74+eff8aECRMeOXYiIno4Jycn/P7774iPj8fhw4fRsmVLTJ8+HZMmTarymSBERNR8NMoCpCbl5eU4cOAAwsLCNG1GRkbw9fVFVlZWrY5RWFgIc3NzWFpa4tatW8jMzMQrr7xSbf+ysjKUlZVptouLiwEAKpUKKpWqnpk0PZW5yilnQJ55yzFngHlLrVWrVpg1a5ZBzk1ERIbT5AqQoqIiVFRUwN7eXqvd3t4eubm5tTrGuXPnMGvWLM3k89deew19+/attn9kZCSWLl2q056enq6ZzC4nqamphg7BIOSYtxxzBuSXd2lpqcHOfezYMZw/fx7l5eVa7WPGjDFQRERE1ND0VoBcuHABCoUCHTp0AABkZ2djy5YtcHNza3TfcHl5edX6Fi0ACAsLQ2hoqGa7uLgYzs7O8PHxkdXSvSqVCqmpqfDz85PVLRJyzFuOOQPyzfv69euSn/PMmTMYN24c/vjjDygUCs1KhAqFAgBQUVFR52NGR0dj1apVKCgogLu7O9auXQsvL69q+9+8eROLFi3C9u3bcePGDXTq1AlRUVF4+umn65cUERHVit4KkMmTJ2PWrFmYMmUKCgoK4Ofnh969eyM+Ph4FBQUIDw/Xy3lsbW1hbGysM2m8sLAQDg4OejnH3ymVSiiVSp12ExMTWf0jpRLzlg855gzIL29D5Dp37lx07twZaWlp6Ny5M7Kzs3H9+nW88cYb+OCDD+p8vMrFSWJiYuDt7Y2oqCj4+/vjxIkTsLOz0+lfXl4OPz8/2NnZYdu2bXBycsK5c+fQpk0bPWRHREQ10dsqWEeOHNF80/T111+jT58+2LdvH+Lj4xEXF6ev08DU1BT9+/dHWlqapk2tViMtLQ2DBg3S23mIiKjhZGVlYdmyZbC1tYWRkRGMjIwwdOhQREZGYs6cOXU+3l8XJ3Fzc0NMTAzMzc0RGxtbZf/Y2FjcuHEDSUlJGDJkCFxcXDBixAi4u7s/ampERPQQersColKpNFcJdu/erbl/t2fPnrhy5UqdjnXnzh3k5eVptvPz85GTkwNra2t07NgRoaGhCA4OhqenJ7y8vBAVFYWSkhLNqlhERNS4VVRUwNLSEsCDK9uXL19Gjx490KlTJ5w4caJOx6rP4iTffvstBg0ahNmzZ+O///0v2rVrh8mTJ2P+/PkwNjauch8uSMKFGpi3PMg9bynorQDp3bs3YmJiMHr0aKSmpmL58uUAgMuXL9d5nsT+/fvh4+Oj2a6cfxEcHIy4uDgEBQXh2rVrCA8PR0FBATw8PJCSkqIzMZ2IiBqnPn364PDhw+jcuTO8vb2xcuVKmJqa4rPPPkOXLl3qdKz6LE5y5swZ/Pjjj/jHP/6BnTt3Ii8vD6+++ipUKhUiIiKq3IcLkvwfuS3UUIl5y4vc8pZyQRK9FSD//ve/MW7cOKxatQrBwcGay9jffvttjZMAqzJy5EjNhMTqhISEICQkpN7xEhGR4bzzzjsoKSkBACxbtgzPPPMMhg0bBhsbGyQmJjb4+dVqNezs7PDZZ5/B2NgY/fv3x6VLl7Bq1apqCxAuSCLfhRqYN/OWAykXJNFbATJy5EgUFRWhuLgYbdu21bTPmjVLdt8MERFRzfz9/TV/7tq1K3Jzc3Hjxg20bdtWsxJWbdVncZL27dvDxMRE63arXr16oaCgAOXl5TA1NdXZhwuS/B855gwwb7mRW95S5qq3Seh3795FWVmZpvg4d+4coqKiql2BhIiI5EmlUqFFixY4cuSIVru1tXWdiw+gfouTDBkyBHl5eVCr1Zq2kydPon379lUWH0REpD96K0CeffZZfPHFFwAerK3u7e2N1atXY+zYsfjkk0/0dRoiImriTExM0LFjx3o966M6oaGh2LBhAz7//HMcP34cr7zyitbiJFOnTtWapP7KK6/gxo0bmDt3Lk6ePInk5GSsWLECs2fP1ltMRERUNb0VIAcPHsSwYcMAANu2bYO9vT3OnTuHL774Ah9//LG+TkNERM3AokWLsHDhQty4cUMvxwsKCsIHH3yA8PBweHh4ICcnR2txkvPnz2utyOjs7IwffvgBv/32G/r164c5c+Zg7ty5WLBggV7iISKi6ultDkhpaalmScVdu3Zh/PjxMDIywsCBA3Hu3Dl9nYaIiJqBdevWIS8vD46OjujUqRNatWql9frBgwfrfMyaFifJyMjQaRs0aBB++eWXOp+HiIgejd4KkK5duyIpKQnjxo3DDz/8gHnz5gEArl69CisrK32dhoiImoGxY8caOgQiIjIQvRUg4eHhmDx5MubNm4cnnnhCM/Fv165deOyxx/R1GiIiagaqW+qWiIiaP70VIBMmTMDQoUNx5coVzTNAAGDUqFEYN26cvk5DRERERERNmN4KEABwcHCAg4MDLl68CADo0KFDnR9CSEREzZ+RkVGNS+7qc4UsIiJqXPRWgKjVarz77rtYvXo17ty5AwCwtLTEG2+8gUWLFsHISG8LbhERURO3Y8cOrW2VSoVDhw7h888/x9KlSw0UFRERSUFvBciiRYuwadMmvP/++xgyZAgAYM+ePViyZAnu3buH9957T1+nIiKiJu7ZZ5/VaZswYQJ69+6NxMREvPjiiwaIioiIpKC3yxKff/45Nm7ciFdeeQX9+vVDv3798Oqrr2LDhg2Ii4vT12n0qrS0FJ06dcKbb75p6FCIiAjAwIEDtZ5oTkREzY/eCpAbN26gZ8+eOu09e/bU24Om9O29997DwIEDDR0GEREBuHv3Lj7++GM4OTkZOhQiImpAersFy93dHevWrdN56vm6devQr18/fZ1Gb06dOoXc3FwEBgbiyJEjhg6HiEhW2rZtqzUJXQiB27dvw9zcHF9++aUBIyMiooamtwJk5cqVGD16NHbv3q15BkhWVhYuXLiAnTt31ulYmZmZWLVqFQ4cOIArV65gx44dOg+tio6OxqpVq1BQUAB3d3esXbu2Tituvfnmm1i1ahX27dtXp9iIiOjRffjhh1oFiJGREdq1awdvb2+0bdvWgJEREVFD01sBMmLECJw8eRLR0dHIzc0FAIwfPx6zZs3Cu+++i2HDhtX6WCUlJXB3d8eMGTMwfvx4ndcTExMRGhqKmJgYeHt7IyoqCv7+/jhx4gTs7OwAAB4eHrh//77Ovrt27cJvv/2G7t27o3v37ixAiIgMYNq0aYYOgYiIDESvzwFxdHTUWe3q8OHD2LRpEz777LNaHycgIAABAQHVvr5mzRrMnDkT06dPBwDExMQgOTkZsbGxWLBgAQAgJyen2v1/+eUXJCQkYOvWrbhz5w5UKhWsrKwQHh5eZf+ysjKUlZVptouLiwE8WDZSpVLVOq+mrjJXOeUMyDNvOeYMMG8pbd68GRYWFnj++ee12rdu3YrS0lIEBwdLHhMREUlDrwWIFMrLy3HgwAGEhYVp2oyMjODr64usrKxaHSMyMhKRkZEAgLi4OBw5cqTa4qOyf1Xr0qenp8Pc3LyOGTR9qamphg7BIOSYtxxzBuSXd2lpqeTnjIyMxKeffqrTbmdnh1mzZrEAISJqxppcAVJUVISKigrY29trtdvb22tu/dK3sLAwhIaGaraLi4vh7OwMHx8f2NjYNMg5GyOVSoXU1FT4+fnBxMTE0OFIRo55yzFnQL55X79+XfJznj9/Hp07d9Zp79SpE86fPy95PEREJJ0mV4DoW23uQ1YqlVAqlTrtJiYmsvpHSiXmLR9yzBmQX96GyNXOzg6///47XFxctNoPHz4sqy92iIjk6JELkKomif/VzZs3H/UUWmxtbWFsbIzCwkKt9sLCQjg4OOj1XERE1DAmTZqEOXPmwNLSEsOHDwcA/PTTT5g7dy5eeOEFA0dHREQN6ZELkNatWz/09alTpz7qaTRMTU3Rv39/pKWlaZbmVavVSEtLQ0hIiN7OQ0REDWf58uU4e/YsRo0ahRYtHgxFarUaU6dOxYoVKwwcHRERNaRHLkA2b96sjzi03LlzB3l5eZrt/Px85OTkwNraGh07dkRoaCiCg4Ph6ekJLy8vREVFoaSkRLMqFhERNW6mpqZITEzEu+++i5ycHLRs2RJ9+/ZFp06dDB0aERE1sEY5B2T//v3w8fHRbFdOAA8ODkZcXByCgoJw7do1hIeHo6CgAB4eHkhJSdGZmE5ERI1bt27d0K1bN0OHQUREEmqUBcjIkSMhhKixT0hICG+5IiJqop577jl4eXlh/vz5Wu0rV67Eb7/9hq1btxooMiIiamhGhg6AiIjkJzMzE08//bROe0BAADIzM+t1zOjoaLi4uMDMzAze3t7Izs6u1X4JCQlQKBSaeYVERNSwWIAQEZHk7ty5A1NTU512ExMTFBcX1/l4iYmJCA0NRUREBA4ePAh3d3f4+/vj6tWrNe539uxZvPnmmxg2bFidz0lERPXDAoSIiCTXt29fJCYm6rQnJCTAzc2tzsdbs2YNZs6cienTp8PNzQ0xMTEwNzdHbGxstftUVFTgH//4B5YuXYouXbrU+ZxERFQ/jXIOCBERNW+LFy/G+PHjcfr0aTzxxBMAgLS0NGzZsgXbtm2r07HKy8tx4MABhIWFadqMjIzg6+uLrKysavdbtmwZ7Ozs8OKLL+Lnn39+6HnKyspQVlam2a68UqNSqaBSqeoUc1NVmadc8q3EvJm3HEiZLwsQIiKSXGBgIJKSkrBixQps27YNLVu2hLu7O3788UdYW1vX6VhFRUWoqKjQWQnR3t4eubm5Ve6zZ88ebNq0CTk5ObU+T2RkJJYuXarTnp6eDnNz8zrF3NSlpqYaOgSDYN7yIre8S0tLJTsXCxAiIjKI0aNHY/To0QAeXE346quv8Oabb+LAgQOoqKhosPPevn0bU6ZMwYYNG2Bra1vr/cLCwjTLwgMPYnZ2doaPjw9sbGwaItRGR6VSITU1FX5+fjAxMTF0OJJh3sxbDq5fvy7ZuViAEBGRwWRmZmLTpk345ptv4OjoiPHjxyM6OrpOx7C1tYWxsTEKCwu12gsLC+Hg4KDT//Tp0zh79iwCAwM1bWq1GgDQokULnDhxAq6urjr7KZVKKJVKnXYTExNZ/SMFkGfOAPOWG7nlLWWuLECIiEhSBQUFiIuLw6ZNm1BcXIyJEyeirKwMSUlJ9ZqAbmpqiv79+yMtLU2zlK5arUZaWlqVz4vq2bMn/vjjD622d955B7dv38ZHH30EZ2fneuVFRES1wwKEiIgkExgYiMzMTIwePRpRUVF46qmnYGxsjJiYmEc6bmhoKIKDg+Hp6QkvLy9ERUWhpKQE06dPBwBMnToVTk5OiIyMhJmZGfr06aO1f5s2bQBAp52IiPSPBQgREUnm+++/x5w5c/DKK6+gW7duejtuUFAQrl27hvDwcBQUFMDDwwMpKSmaiennz5+HkRFXniciagxkW4C4uLjAysoKRkZGaNu2LdLT0w0dEhFRs1e5+lT//v3Rq1cvTJkyBS+88IJejh0SElLlLVcAkJGRUeO+cXFxeomBiIgeTtZfB+3btw85OTksPoiIJDJw4EBs2LABV65cwcsvv4yEhAQ4OjpCrVYjNTUVt2/fNnSIRETUwGRdgBARkWG0atUKM2bMwJ49e/DHH3/gjTfewPvvvw87OzuMGTPG0OEREVEDapQFSGZmJgIDA+Ho6AiFQoGkpCSdPtHR0XBxcYGZmRm8vb2RnZ1dp3MoFAqMGDECAwYMQHx8vJ4iJyKiuurRowdWrlyJixcv4quvvjJ0OERE1MAa5RyQkpISuLu7Y8aMGRg/frzO64mJiQgNDUVMTAy8vb0RFRUFf39/nDhxAnZ2dgAADw8P3L9/X2ffXbt2wdHREXv27IGTkxOuXLkCX19f9O3bF/369Wvw3IiIqGrGxsYYO3asZildIiJqnhplARIQEICAgIBqX1+zZg1mzpypWV4xJiYGycnJiI2NxYIFCwAAOTk5NZ7DyckJANC+fXs8/fTTOHjwYLUFSFlZGcrKyjTbxcXFAB48KVOlUtU6r6auMlc55QzIM2855gwwbyIiIik0ygKkJuXl5Thw4ADCwsI0bUZGRvD19UVWVlatjlFSUgK1Wg1LS0vcuXMHP/74IyZOnFht/8jISCxdulSnPT09Hebm5nVPoolLTU01dAgGIce85ZgzIL+8S0tLDR0CERHJSJMrQIqKilBRUaFZ272Svb09cnNza3WMwsJCjBs3DgBQUVGBmTNnYsCAAdX2DwsLQ2hoqGa7uLgYzs7O8PHxgY2NTT2yaJpUKhVSU1Ph5+cHExMTQ4cjGTnmLcecAfnmff36dUOHQEREMtLkChB96NKlCw4fPlzr/kqlEkqlUqfdxMREVv9IqcS85UOOOQPyy1tOuRIRkeE1ylWwamJrawtjY2MUFhZqtRcWFsLBwcFAURERERERUW00uQLE1NQU/fv3R1pamqZNrVYjLS0NgwYNMmBkRERERET0MI3yFqw7d+4gLy9Ps52fn4+cnBxYW1ujY8eOCA0NRXBwMDw9PeHl5YWoqCiUlJRoVsUiIiIiIqLGqVEWIPv374ePj49mu3ICeHBwMOLi4hAUFIRr164hPDwcBQUF8PDwQEpKis7EdCIiIiIialwaZQEycuRICCFq7BMSEoKQkBCJIiIiIiIiIn1ocnNAiIiIiIio6WIBQkREREREkmEBQkREREREkmEBQkREREREkmEBQkREREREkmEBQkREREREkmEBQkREREREkmEBQkREzUJ0dDRcXFxgZmYGb29vZGdnV9t3w4YNGDZsGNq2bYu2bdvC19e3xv5ERKQ/LECIiKjJS0xMRGhoKCIiInDw4EG4u7vD398fV69erbJ/RkYGJk2ahPT0dGRlZcHZ2RlPPvkkLl26JHHkRETywwKEiIiavDVr1mDmzJmYPn063NzcEBMTA3Nzc8TGxlbZPz4+Hq+++io8PDzQs2dPbNy4EWq1GmlpaRJHTkQkP7ItQPLz8+Hj4wM3Nzf07dsXJSUlhg6JiIjqoby8HAcOHICvr6+mzcjICL6+vsjKyqrVMUpLS6FSqWBtbd1QYRIR0f/XwtABGMq0adPw7rvvYtiwYbhx4waUSqWhQyIionooKipCRUUF7O3ttdrt7e2Rm5tbq2PMnz8fjo6OWkXM35WVlaGsrEyzXVxcDABQqVRQqVT1iLzpqcxTLvlWYt7MWw6kzFeWBcjRo0dhYmKCYcOGAQC/8SIikrH3338fCQkJyMjIgJmZWbX9IiMjsXTpUp329PR0mJubN2SIjU5qaqqhQzAI5i0vcsu7tLRUsnM1ygIkMzMTq1atwoEDB3DlyhXs2LEDY8eO1eoTHR2NVatWoaCgAO7u7li7di28vLxqdfxTp07BwsICgYGBuHTpEiZMmICFCxc2QCZERNTQbG1tYWxsjMLCQq32wsJCODg41LjvBx98gPfffx+7d+9Gv379auwbFhaG0NBQzXZxcTGcnZ3h4+MDGxub+ifQhKhUKqSmpsLPzw8mJiaGDkcyzJt5y8H169clO1ejLEBKSkrg7u6OGTNmYPz48TqvV652EhMTA29vb0RFRcHf3x8nTpyAnZ0dAMDDwwP379/X2XfXrl24f/8+fv75Z+Tk5MDOzg5PPfUUBgwYAD8/vwbPjYiI9MvU1BT9+/dHWlqa5suqygnlISEh1e63cuVKvPfee/jhhx/g6en50PMolcoqb9c1MTGR1T9SAHnmDDBvuZFb3lLm2igLkICAAAQEBFT7+l9XOwGAmJgYJCcnIzY2FgsWLAAA5OTkVLu/k5MTPD094ezsDAB4+umnkZOTU20Bwvt+H5D7PZFyyluOOQPMuykLDQ1FcHAwPD094eXlhaioKJSUlGjGialTp8LJyQmRkZEAgH//+98IDw/Hli1b4OLigoKCAgCAhYUFLCwsDJYHEZEcNMoCpCaVq52EhYVp2uq62smAAQNw9epV/Pnnn2jdujUyMzPx8ssvV9uf9/1qk9s9kZXkmLcccwbkl7eU9/02lKCgIFy7dg3h4eEoKCiAh4cHUlJSNBPTz58/DyOj/1v48ZNPPkF5eTkmTJigdZyIiAgsWbJEytCJiGSnyRUg+ljtpEWLFlixYgWGDx8OIQSefPJJPPPMM9X2532/D8j1nkg55i3HnAH55i3lfb8NKSQkpNpbrjIyMrS2z5492/ABERFRlZpcAaIvD7vN669436825i0fcswZkF/ecsqViIgMr8k9iPBRVjshIiIiIiLDanIFyF9XO6lUudrJoEGDDBgZERERERE9TKO8BevOnTvIy8vTbOfn5yMnJwfW1tbo2LHjQ1c7ISIiIiKixqlRFiD79++Hj4+PZrtyAnhwcDDi4uIeutoJERERERE1To2yABk5ciSEEDX2qWm1EyIiIiIiapya3BwQIiIiIiJquliAEBERERGRZFiAEBERERGRZFiAEBERERGRZFiAEBERERGRZFiAEBERERGRZFiAEBERERGRZFiAEBERERGRZFiAEBERERGRZFiAEBERERGRZFiAEBERERGRZFiAEBERERGRZFiAEBERERGRZFiAEBERERGRZGRZgERHR8PFxQVmZmbw9vZGdna2oUMiIqJHVNfP9q1bt6Jnz54wMzND3759sXPnTokiJSKSN9kVIImJiQgNDUVERAQOHjwId3d3+Pv74+rVq4YOjYiI6qmun+379u3DpEmT8OKLL+LQoUMYO3Ysxo4diyNHjkgcORGR/MiuAFmzZg1mzpyJ6dOnw83NDTExMTA3N0dsbKyhQyMionqq62f7Rx99hKeeegpvvfUWevXqheXLl+Pxxx/HunXrJI6ciEh+ZFWAlJeX48CBA/D19dW0GRkZwdfXF1lZWQaMjIiI6qs+n+1ZWVla/QHA39+fYwERkQRaGDoAKRUVFaGiogL29vZa7fb29sjNza12v7KyMpSVlWm2b926BQC4ceNGwwTaSKlUKpSWluL69eswMTExdDiSkWPecswZkG/elZ9lQggDR1I/9flsLygoqLJ/QUFBtefhWCDfvyPMm3nLgZRjgawKkPqKjIzE0qVLddq7d+9ugGiIiBrG9evX0bp1a0OH0WhxLCAiOZBiLJBVAWJrawtjY2MUFhZqtRcWFsLBwaHa/cLCwhAaGqrZvnnzJjp16oTz58/LarAuLi6Gs7MzLly4ACsrK0OHIxk55i3HnAH55n3r1i107NgR1tbWhg6lXurz2e7g4MCxoB7k+neEeTNvOZByLJBVAWJqaor+/fsjLS0NY8eOBQCo1WqkpaUhJCSk2v2USiWUSqVOe+vWrWX1P2YlKysr5i0TcswZkG/eRkZNc1pgfT7bBw0ahLS0NLz++uuattTUVAwaNKja83As+D9y/TvCvOVFrnlLMRbIqgABgNDQUAQHB8PT0xNeXl6IiopCSUkJpk+fbujQiIionh722T516lQ4OTkhMjISADB37lyMGDECq1evxujRo5GQkID9+/fjs88+M2QaRESyILsCJCgoCNeuXUN4eDgKCgrg4eGBlJQUncmIRETUdDzss/38+fNa3+oNHjwYW7ZswTvvvIOFCxeiW7duSEpKQp8+fQyVAhGRbMiuAAGAkJCQGm+5ehilUomIiIgqL8U3Z8xbPnnLMWeAeTf1vGv6bM/IyNBpe/755/H888/X+3zN5X2rCznmDDBv5i0PUuatEE113UUiIiIiImpymuaMQyIiIiIiapJYgBARERERkWRYgBARERERkWRYgFQjOjoaLi4uMDMzg7e3N7Kzs2vsv3XrVvTs2RNmZmbo27cvdu7cKVGk+lWXvDds2IBhw4ahbdu2aNu2LXx9fR/6PjVWdf19V0pISIBCodA8e6ApqWvON2/exOzZs9G+fXsolUp07969Sf5/Xte8o6Ki0KNHD7Rs2RLOzs6YN28e7t27J1G0+pGZmYnAwEA4OjpCoVAgKSnpoftkZGTg8ccfh1KpRNeuXREXF9fgcTY2HAc4DtRGUx4HAI4FHAtq1mBjgSAdCQkJwtTUVMTGxoqjR4+KmTNnijZt2ojCwsIq++/du1cYGxuLlStXimPHjol33nlHmJiYiD/++EPiyB9NXfOePHmyiI6OFocOHRLHjx8X06ZNE61btxYXL16UOPJHU9e8K+Xn5wsnJycxbNgw8eyzz0oTrJ7UNeeysjLh6ekpnn76abFnzx6Rn58vMjIyRE5OjsSRP5q65h0fHy+USqWIj48X+fn54ocffhDt27cX8+bNkzjyR7Nz506xaNEisX37dgFA7Nixo8b+Z86cEebm5iI0NFQcO3ZMrF27VhgbG4uUlBRpAm4EOA5wHGju44AQHAs4FuyosX9DjgUsQKrg5eUlZs+erdmuqKgQjo6OIjIyssr+EydOFKNHj9Zq8/b2Fi+//HKDxqlvdc377+7fvy8sLS3F559/3lAhNoj65H3//n0xePBgsXHjRhEcHNzkBp665vzJJ5+ILl26iPLycqlCbBB1zXv27NniiSee0GoLDQ0VQ4YMadA4G1JtBp23335b9O7dW6stKChI+Pv7N2BkjQvHgQc4DjTfcUAIjgWVOBZUrSHHAt6C9Tfl5eU4cOAAfH19NW1GRkbw9fVFVlZWlftkZWVp9QcAf3//avs3RvXJ++9KS0uhUqlgbW3dUGHqXX3zXrZsGezs7PDiiy9KEaZe1Sfnb7/9FoMGDcLs2bNhb2+PPn36YMWKFaioqJAq7EdWn7wHDx6MAwcOaC7NnzlzBjt37sTTTz8tScyG0hw+0x4FxwGOA819HAA4FnAseLiG/FyT5YMIa1JUVISKigqdJ6Pb29sjNze3yn0KCgqq7F9QUNBgcepbffL+u/nz58PR0VHnf9bGrD5579mzB5s2bUJOTo4EEepffXI+c+YMfvzxR/zjH//Azp07kZeXh1dffRUqlQoRERFShP3I6pP35MmTUVRUhKFDh0IIgfv37+Nf//oXFi5cKEXIBlPdZ1pxcTHu3r2Lli1bGigyaXAc4DgANO9xAOBYwLHg4RpyLOAVENKL999/HwkJCdixYwfMzMwMHU6DuX37NqZMmYINGzbA1tbW0OFIRq1Ww87ODp999hn69++PoKAgLFq0CDExMYYOrUFlZGRgxYoVWL9+PQ4ePIjt27cjOTkZy5cvN3RoRI0Ox4Hmj2MBxwJ94RWQv7G1tYWxsTEKCwu12gsLC+Hg4FDlPg4ODnXq3xjVJ+9KH3zwAd5//33s3r0b/fr1a8gw9a6ueZ8+fRpnz55FYGCgpk2tVgMAWrRogRMnTsDV1bVhg35E9fldt2/fHiYmJjA2Nta09erVCwUFBSgvL4epqWmDxqwP9cl78eLFmDJlCl566SUAQN++fVFSUoJZs2Zh0aJFMDJqnt/hVPeZZmVl1eyvfgAcBzgOPNCcxwGAYwHHgodryLGgeb5jj8DU1BT9+/dHWlqapk2tViMtLQ2DBg2qcp9BgwZp9QeA1NTUavs3RvXJGwBWrlyJ5cuXIyUlBZ6enlKEqld1zbtnz574448/kJOTo/kZM2YMfHx8kJOTA2dnZynDr5f6/K6HDBmCvLw8zSALACdPnkT79u2bxIAD1C/v0tJSnYGlcuB9MIeveWoOn2mPguMAx4HmPg4AHAs4Fjxcg36uPfI09mYoISFBKJVKERcXJ44dOyZmzZol2rRpIwoKCoQQQkyZMkUsWLBA03/v3r2iRYsW4oMPPhDHjx8XERERTXb5xbrk/f777wtTU1Oxbds2ceXKFc3P7du3DZVCvdQ1779riquf1DXn8+fPC0tLSxESEiJOnDghvvvuO2FnZyfeffddQ6VQL3XNOyIiQlhaWoqvvvpKnDlzRuzatUu4urqKiRMnGiqFerl9+7Y4dOiQOHTokAAg1qxZIw4dOiTOnTsnhBBiwYIFYsqUKZr+lUsvvvXWW+L48eMiOjpalsvwchzgONCcxwEhOBZwLDDcWMACpBpr164VHTt2FKampsLLy0v88ssvmtdGjBghgoODtfp//fXXonv37sLU1FT07t1bJCcnSxyxftQl706dOgkAOj8RERHSB/6I6vr7/qumOvDUNed9+/YJb29voVQqRZcuXcR7770n7t+/L3HUj64ueatUKrFkyRLh6uoqzMzMhLOzs3j11VfFn3/+KX3gjyA9Pb3Kv6uVuQYHB4sRI0bo7OPh4SFMTU1Fly5dxObNmyWP29A4DnAcqNRcxwEhOBZwLDDMWKAQohlfOyIiIiIiokaFc0CIiIiIiEgyLECIiIiIiEgyLECIiIiIiEgyLECIiIiIiEgyLECIiIiIiEgyLECIiIiIiEgyLECIiIiIiEgyLECIiIiIiEgyLECImpmMjAwoFArcvHnT0KEQEZGBcCygxowFCBERERERSYYFCBERERERSYYFCJGeqdVqREZGonPnzmjZsiXc3d2xbds2AP93STw5ORn9+vWDmZkZBg4ciCNHjmgd45tvvkHv3r2hVCrh4uKC1atXa71eVlaG+fPnw9nZGUqlEl27dsWmTZu0+hw4cACenp4wNzfH4MGDceLEiYZNnIiINDgWENVAEJFevfvuu6Jnz54iJSVFnD59WmzevFkolUqRkZEh0tPTBQDRq1cvsWvXLvH777+LZ555Rri4uIjy8nIhhBD79+8XRkZGYtmyZeLEiRNi8+bNomXLlmLz5s2ac0ycOFE4OzuL7du3i9OnT4vdu3eLhIQEIYTQnMPb21tkZGSIo0ePimHDhonBgwcb4u0gIpIljgVE1WMBQqRH9+7dE+bm5mLfvn1a7S+++KKYNGmSZkCoHCCEEOL69euiZcuWIjExUQghxOTJk4Wfn5/W/m+99ZZwc3MTQghx4sQJAUCkpqZWGUPlOXbv3q1pS05OFgDE3bt39ZInERFVj2MBUc14CxaRHuXl5aG0tBR+fn6wsLDQ/HzxxRc4ffq0pt+gQYM0f7a2tkaPHj1w/PhxAMDx48cxZMgQreMOGTIEp06dQkVFBXJycmBsbIwRI0bUGEu/fv00f27fvj0A4OrVq4+cIxER1YxjAVHNWhg6AKLm5M6dOwCA5ORkODk5ab2mVCq1Bp76atmyZa36mZiYaP6sUCgAPLgnmYiIGhbHAqKa8QoIkR65ublBqVTi/Pnz6Nq1q9aPs7Ozpt8vv/yi+fOff/6JkydPolevXgCAXr16Ye/evVrH3bt3L7p37w5jY2P07dsXarUaP/30kzRJERFRnXAsIKoZr4AQ6ZGlpSXefPNNzJs3D2q1GkOHDsWtW7ewd+9eWFlZoVOnTgCAZcuWwcbGBvb29li0aBFsbW0xduxYAMAbb7yBAQMGYPny5QgKCkJWVhbWrVuH9evXAwBcXFwQHByMGTNm4OOPP4a7uzvOnTuHq1evYuLEiYZKnYiI/j+OBUQPYehJKETNjVqtFlFRUaJHjx7CxMREtGvXTvj7+4uffvpJMynwf//7n+jdu7cwNTUVXl5e4vDhw1rH2LZtm3BzcxMmJiaiY8eOYtWqVVqv3717V8ybN0+0b99emJqaiq5du4rY2FghxP9NPPzzzz81/Q8dOiQAiPz8/IZOn4iIBMcCopoohBDCkAUQkZxkZGTAx8cHf/75J9q0aWPocIiIyAA4FpDccQ4IERERERFJhgUIERERERFJhrdgERERERGRZHgFhIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJMMChIiIiIiIJPP/AML2nC/1xUxJAAAAAElFTkSuQmCC",
      "text/html": [
       "\n",
       "            <div style=\"display: inline-block;\">\n",
       "                <div class=\"jupyter-widgets widget-label\" style=\"text-align: center;\">\n",
       "                    Figure\n",
       "                </div>\n",
       "                <img src='data:image/png;base64,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' width=800.0/>\n",
       "            </div>\n",
       "        "
      ],
      "text/plain": [
       "Canvas(header_visible=False, toolbar=Toolbar(toolitems=[('Home', 'Reset original view', 'home', 'home'), ('Bac…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Realtime training progress (for static-architecture models)\n",
    "importlib.reload(pm)\n",
    "if 'static_rtf' in globals():\n",
    "  plt.close(static_rtf.fig) # Release old figure memory.\n",
    "static_rtf = pm.RealtimeTrainingFig('epoch')\n",
    "display(static_rtf.fig.canvas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pause between generating figure and updating it in training loop;\n",
    "# only necessary when 'run all' is used, otherwise figure is not displayed until training loop finishes.\n",
    "time.sleep(0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "importlib.reload(pm)\n",
    "\n",
    "record_every = 10 #epochs\n",
    "dataset = eeg_all # CHANGE TO RUN DIFFERENT DATASETS\n",
    "\n",
    "# Even when experimenting, we want to use as much training data as possible\n",
    "# (i.e. leave only one subject out) as this significantly affects the results. \n",
    "# However, for speed we can limit the number of folds. Use dataset.n_subjects for final results.\n",
    "n_folds = dataset.n_subjects\n",
    "n_runs = 1 # CHANGE TO 5 for paper results\n",
    "\n",
    "metric_names = ['epoch', 'time', 'train_loss', 'train_accuracy', 'test_loss', 'test_accuracy'] # For tracking training\n",
    "\n",
    "for M_id, M in staticModels.items():\n",
    "  # Re-initialise RNG for this model, and generate seeds for each fold/run.\n",
    "  torch.manual_seed(metaseed)\n",
    "  M_seeds = torch.randint(2**30, 2**32, (n_runs, n_folds))\n",
    "\n",
    "  plot_every = max(100, M.n_epochs//10)\n",
    "  n_records = int(M.n_epochs/record_every) + 1 # Per run and fold\n",
    "\n",
    "  # Metrics for every record of every fold of every run\n",
    "  all_confusions = torch.zeros(n_runs, n_folds, n_records, 2, 2) # Test-set confusion matrix\n",
    "  # Training and test loss/accuracy\n",
    "  all_metrics = torch.zeros(n_runs, n_folds, n_records, len(metric_names))\n",
    "\n",
    "  for run_idx in range(n_runs):\n",
    "    # Clear plot\n",
    "    static_rtf.reset(M.n_epochs)\n",
    "\n",
    "    # Leave-one-subject-out cross validation (via kf)\n",
    "    for fold_idx, (train_idx, test_idx) in enumerate(dataset.split()):\n",
    "      if fold_idx >= n_folds:\n",
    "        break\n",
    "\n",
    "      # Initialise RNG and network for this fold.\n",
    "      torch.manual_seed(M_seeds[run_idx, fold_idx])\n",
    "      net = M.initNet(dataset.n_features, **M.net_kw)\n",
    "      optimiser = M.initOpt(net.parameters(), **M.opt_kw)\n",
    "\n",
    "      # Batching\n",
    "      train_loader = DataLoader(dataset, batch_size=M.batch_size,\n",
    "        drop_last = False, sampler=SubsetRandomSampler(train_idx))\n",
    "\n",
    "      # Full train/test sets for periodically reporting performance\n",
    "      trainX, trainY = dataset[train_idx]\n",
    "      testX, testY = dataset[test_idx]\n",
    "\n",
    "      # New line on plot\n",
    "      static_rtf.begin_fold(f\"(Run {run_idx+1}/{n_runs}, Fold {fold_idx+1}/{n_folds}) {str(M)}\")\n",
    "\n",
    "      fold_metrics = pm.Metrics(all_metrics[run_idx, fold_idx, :, :], metric_names)\n",
    "\n",
    "      t0 = time.time() # Start time\n",
    "      for epoch_idx in range(M.n_epochs):\n",
    "        for batch_idx, (x, y) in enumerate(train_loader):\n",
    "          loss = bceLogitLossMean(net(x), y)\n",
    "          optimiser.zero_grad()\n",
    "          loss.backward()\n",
    "          optimiser.step()\n",
    "        \n",
    "        # Occasionally record loss and accuracy for training/test sets.\n",
    "        # Much simpler than computing for every epoch, to avoid interweaving with batching.\n",
    "        # Assume this consumes negligible time, provided record_every is large.\n",
    "        if (epoch_idx % record_every == 0) or (epoch_idx == M.n_epochs-1):\n",
    "          with torch.inference_mode(): # Disables autograd temporarily.\n",
    "            train_loss, train_acc = loss_accuracy(net(trainX), trainY)\n",
    "            \n",
    "            test_logits = net(testX)\n",
    "            test_loss, test_acc = loss_accuracy(test_logits, testY)\n",
    "            test_confusion = pm.confusion_matrix(F.sigmoid(test_logits), testY)\n",
    "            all_confusions[run_idx, fold_idx, fold_metrics.n_rec, :, :] = test_confusion\n",
    "\n",
    "            fold_metrics.update(epoch_idx, time.time() - t0, train_loss, train_acc, test_loss, test_acc)\n",
    "        \n",
    "        # Occasionally update real-time training plot. Blitting (i.e. only \n",
    "        # redrawing the data lines) makes this fairly efficient.\n",
    "        if (epoch_idx % plot_every == 0) or (epoch_idx == M.n_epochs-1):\n",
    "          static_rtf.update(fold_metrics)\n",
    "          #print(f\"Epoch {epoch_idx}:\\n  Train: {train_acc:.3f} accuracy, {train_loss:.3e} loss\\n  Test: {test_acc:.3f} accuracy, {test_loss:.3e} loss\")\n",
    "  \n",
    "  # Compute statistics over all folds/runs\n",
    "  total_confusion = all_confusions.sum(dim=(0,1)) #(n_records, num_metrics)\n",
    "  avg_metrics = pm.Metrics(all_metrics.mean(dim=(0,1)), metric_names, n_records)\n",
    "  # Note data for each fold/run is recorded at same epochs\n",
    "\n",
    "  # Record performance for this model\n",
    "  M_perf = pm.KFoldPerformance(total_confusion, avg_metrics, n_folds, n_runs)\n",
    "  dataset.perfs[M_id] = M_perf\n",
    "  \n",
    "  # Plot this model\n",
    "  fold_fig = M_perf.plot(f\"{n_runs}-run {n_folds}-fold avg: {str(M)}\")\n",
    "  display(fold_fig)\n",
    "  plt.close(fold_fig) # Avoid large memory use from many open figures\n",
    "\n",
    "dataset.recompute_model_perfs(staticModels.keys()) # just those run above"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\n# For eeg_all dataset with linGD, linAdam and h30Adam: tradeoff between acc/prec/etc. is non-unanimous and made by taking minimum of loss function. Uncomment and run after running these models.\\n\\nlinGDminL, linGDbestIdx = torch.min(eeg_all.perfs['linGDlr0.1'].train_metrics.test_loss, dim=0)\\n\\nlinAdamMinL, linAdamBestIdx = torch.min(eeg_all.perfs['linAdamLr0.001bs12'].train_metrics.test_loss, dim=0)\\n\\nlinAdamBestEpoch = eeg_all.perfs['linAdamLr0.001bs12'].train_metrics.epoch[linAdamBestIdx].item()\\n\\nlinGDbestEpoch = eeg_all.perfs['linGDlr0.1'].train_metrics.epoch[linGDbestIdx].item()\\n\\nprint(linGDminL, linGDbestEpoch)\\nprint(linAdamMinL, linAdamBestEpoch)\\n\""
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "# For eeg_all dataset with linGD, linAdam and h30Adam: tradeoff between acc/prec/etc. is non-unanimous and made by taking minimum of loss function. Uncomment and run after running these models.\n",
    "\n",
    "linGDminL, linGDbestIdx = torch.min(eeg_all.perfs['linGDlr0.1'].train_metrics.test_loss, dim=0)\n",
    "\n",
    "linAdamMinL, linAdamBestIdx = torch.min(eeg_all.perfs['linAdamLr0.001bs12'].train_metrics.test_loss, dim=0)\n",
    "\n",
    "linAdamBestEpoch = eeg_all.perfs['linAdamLr0.001bs12'].train_metrics.epoch[linAdamBestIdx].item()\n",
    "\n",
    "linGDbestEpoch = eeg_all.perfs['linGDlr0.1'].train_metrics.epoch[linGDbestIdx].item()\n",
    "\n",
    "print(linGDminL, linGDbestEpoch)\n",
    "print(linAdamMinL, linAdamBestEpoch)\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# eeg_all.recompute_model_perfs(staticModels.keys()) # if needs recomputing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performance: all\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>n_runs</th>\n",
       "      <th>n_folds</th>\n",
       "      <th>epochs</th>\n",
       "      <th>time</th>\n",
       "      <th>size</th>\n",
       "      <th>train-acc</th>\n",
       "      <th>acc</th>\n",
       "      <th>prec</th>\n",
       "      <th>sens</th>\n",
       "      <th>spec</th>\n",
       "      <th>f1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>h30GD</th>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>799.0</td>\n",
       "      <td>4.515297</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.951389</td>\n",
       "      <td>0.933333</td>\n",
       "      <td>0.972222</td>\n",
       "      <td>0.930556</td>\n",
       "      <td>0.952381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linGDlr0.1</th>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>140.0</td>\n",
       "      <td>0.123304</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.965278</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.930556</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.964029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linAdamLr0.001bs12</th>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>90.0</td>\n",
       "      <td>0.204294</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.958333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.916667</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.956522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linRprop</th>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>100.0</td>\n",
       "      <td>0.097409</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.958333</td>\n",
       "      <td>0.971429</td>\n",
       "      <td>0.944444</td>\n",
       "      <td>0.972222</td>\n",
       "      <td>0.957746</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h30GDlr0.1</th>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>330.0</td>\n",
       "      <td>0.300006</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.958333</td>\n",
       "      <td>0.945946</td>\n",
       "      <td>0.972222</td>\n",
       "      <td>0.944444</td>\n",
       "      <td>0.958904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h30AdamBs12</th>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>130.0</td>\n",
       "      <td>0.440034</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.958333</td>\n",
       "      <td>0.945946</td>\n",
       "      <td>0.972222</td>\n",
       "      <td>0.944444</td>\n",
       "      <td>0.958904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hid30Rprop</th>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>50.0</td>\n",
       "      <td>0.066969</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.944444</td>\n",
       "      <td>0.957143</td>\n",
       "      <td>0.930556</td>\n",
       "      <td>0.958333</td>\n",
       "      <td>0.943662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>csp</th>\n",
       "      <td>5</td>\n",
       "      <td>24</td>\n",
       "      <td>60.0</td>\n",
       "      <td>0.165057</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.977778</td>\n",
       "      <td>0.967391</td>\n",
       "      <td>0.988889</td>\n",
       "      <td>0.966667</td>\n",
       "      <td>0.978022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cspAdam</th>\n",
       "      <td>5</td>\n",
       "      <td>24</td>\n",
       "      <td>175.0</td>\n",
       "      <td>0.279234</td>\n",
       "      <td>0.425</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.930556</td>\n",
       "      <td>0.981366</td>\n",
       "      <td>0.877778</td>\n",
       "      <td>0.983333</td>\n",
       "      <td>0.926686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cspGD</th>\n",
       "      <td>5</td>\n",
       "      <td>24</td>\n",
       "      <td>750.0</td>\n",
       "      <td>0.889227</td>\n",
       "      <td>2.000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.944444</td>\n",
       "      <td>0.990798</td>\n",
       "      <td>0.897222</td>\n",
       "      <td>0.991667</td>\n",
       "      <td>0.941691</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    n_runs  n_folds  epochs      time   size  train-acc  \\\n",
       "model                                                                     \n",
       "h30GD                    1       24   799.0  4.515297    NaN        1.0   \n",
       "linGDlr0.1               1       24   140.0  0.123304    NaN        1.0   \n",
       "linAdamLr0.001bs12       1       24    90.0  0.204294    NaN        1.0   \n",
       "linRprop                 1       24   100.0  0.097409    NaN        1.0   \n",
       "h30GDlr0.1               1       24   330.0  0.300006    NaN        1.0   \n",
       "h30AdamBs12              1       24   130.0  0.440034    NaN        1.0   \n",
       "hid30Rprop               1       24    50.0  0.066969    NaN        1.0   \n",
       "csp                      5       24    60.0  0.165057  0.000        1.0   \n",
       "cspAdam                  5       24   175.0  0.279234  0.425        1.0   \n",
       "cspGD                    5       24   750.0  0.889227  2.000        1.0   \n",
       "\n",
       "                         acc      prec      sens      spec        f1  \n",
       "model                                                                 \n",
       "h30GD               0.951389  0.933333  0.972222  0.930556  0.952381  \n",
       "linGDlr0.1          0.965278  1.000000  0.930556  1.000000  0.964029  \n",
       "linAdamLr0.001bs12  0.958333  1.000000  0.916667  1.000000  0.956522  \n",
       "linRprop            0.958333  0.971429  0.944444  0.972222  0.957746  \n",
       "h30GDlr0.1          0.958333  0.945946  0.972222  0.944444  0.958904  \n",
       "h30AdamBs12         0.958333  0.945946  0.972222  0.944444  0.958904  \n",
       "hid30Rprop          0.944444  0.957143  0.930556  0.958333  0.943662  \n",
       "csp                 0.977778  0.967391  0.988889  0.966667  0.978022  \n",
       "cspAdam             0.930556  0.981366  0.877778  0.983333  0.926686  \n",
       "cspGD               0.944444  0.990798  0.897222  0.991667  0.941691  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performance: no_outliers\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>n_runs</th>\n",
       "      <th>n_folds</th>\n",
       "      <th>epochs</th>\n",
       "      <th>time</th>\n",
       "      <th>size</th>\n",
       "      <th>train-acc</th>\n",
       "      <th>acc</th>\n",
       "      <th>prec</th>\n",
       "      <th>sens</th>\n",
       "      <th>spec</th>\n",
       "      <th>f1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>linGDlr0.1</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>150.0</td>\n",
       "      <td>0.667928</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.978788</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.957576</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.978328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linSGDlr0.01bs12</th>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.583971</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.977273</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.954545</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.976744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linAdam</th>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>499.0</td>\n",
       "      <td>2.736360</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.924242</td>\n",
       "      <td>0.982759</td>\n",
       "      <td>0.863636</td>\n",
       "      <td>0.984848</td>\n",
       "      <td>0.919355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linAdamLr0.001bs12</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>650.0</td>\n",
       "      <td>7.567780</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.978788</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.957576</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.978328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linAdamlr0.01</th>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>250.0</td>\n",
       "      <td>1.454465</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.946970</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.893939</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.944000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linRprop</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>50.0</td>\n",
       "      <td>0.270667</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.978788</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.957576</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.978328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h30GD</th>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>400.0</td>\n",
       "      <td>1.798959</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.999639</td>\n",
       "      <td>0.977273</td>\n",
       "      <td>0.970149</td>\n",
       "      <td>0.984848</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.977444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h30GDlr0.1</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>300.0</td>\n",
       "      <td>1.653244</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.978788</td>\n",
       "      <td>0.978788</td>\n",
       "      <td>0.978788</td>\n",
       "      <td>0.978788</td>\n",
       "      <td>0.978788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h30GDlr0.001bs12</th>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>400.0</td>\n",
       "      <td>6.576123</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.977273</td>\n",
       "      <td>0.970149</td>\n",
       "      <td>0.984848</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.977444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>csp</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>70.0</td>\n",
       "      <td>0.194177</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.995455</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.990909</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.995434</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cspAdam</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>1.803169</td>\n",
       "      <td>0.945455</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.957576</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.915152</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.955696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cspGD</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>500.0</td>\n",
       "      <td>0.575566</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.975758</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.951515</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.975155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h30AdamBs12</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>1400.0</td>\n",
       "      <td>50.911968</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.984848</td>\n",
       "      <td>0.984848</td>\n",
       "      <td>0.984848</td>\n",
       "      <td>0.984848</td>\n",
       "      <td>0.984848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hid30Rprop</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>50.0</td>\n",
       "      <td>0.305404</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.978788</td>\n",
       "      <td>0.996855</td>\n",
       "      <td>0.960606</td>\n",
       "      <td>0.996970</td>\n",
       "      <td>0.978395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cspGDtst</th>\n",
       "      <td>3</td>\n",
       "      <td>22</td>\n",
       "      <td>2999.0</td>\n",
       "      <td>8.934856</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.979798</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.959596</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.979381</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    n_runs  n_folds  epochs       time       size  train-acc  \\\n",
       "model                                                                          \n",
       "linGDlr0.1               5       22   150.0   0.667928        NaN   1.000000   \n",
       "linSGDlr0.01bs12         1       22   100.0   1.583971        NaN   1.000000   \n",
       "linAdam                  1       22   499.0   2.736360        NaN   1.000000   \n",
       "linAdamLr0.001bs12       5       22   650.0   7.567780        NaN   1.000000   \n",
       "linAdamlr0.01            1       22   250.0   1.454465        NaN   1.000000   \n",
       "linRprop                 5       22    50.0   0.270667        NaN   1.000000   \n",
       "h30GD                    1       22   400.0   1.798959        NaN   0.999639   \n",
       "h30GDlr0.1               5       22   300.0   1.653244        NaN   1.000000   \n",
       "h30GDlr0.001bs12         1       22   400.0   6.576123        NaN   1.000000   \n",
       "csp                      5       22    70.0   0.194177   0.000000   1.000000   \n",
       "cspAdam                  5       22  1000.0   1.803169   0.945455   1.000000   \n",
       "cspGD                    5       22   500.0   0.575566   0.000000   1.000000   \n",
       "h30AdamBs12              5       22  1400.0  50.911968        NaN   1.000000   \n",
       "hid30Rprop               5       22    50.0   0.305404        NaN   1.000000   \n",
       "cspGDtst                 3       22  2999.0   8.934856  10.000000   1.000000   \n",
       "\n",
       "                         acc      prec      sens      spec        f1  \n",
       "model                                                                 \n",
       "linGDlr0.1          0.978788  1.000000  0.957576  1.000000  0.978328  \n",
       "linSGDlr0.01bs12    0.977273  1.000000  0.954545  1.000000  0.976744  \n",
       "linAdam             0.924242  0.982759  0.863636  0.984848  0.919355  \n",
       "linAdamLr0.001bs12  0.978788  1.000000  0.957576  1.000000  0.978328  \n",
       "linAdamlr0.01       0.946970  1.000000  0.893939  1.000000  0.944000  \n",
       "linRprop            0.978788  1.000000  0.957576  1.000000  0.978328  \n",
       "h30GD               0.977273  0.970149  0.984848  0.969697  0.977444  \n",
       "h30GDlr0.1          0.978788  0.978788  0.978788  0.978788  0.978788  \n",
       "h30GDlr0.001bs12    0.977273  0.970149  0.984848  0.969697  0.977444  \n",
       "csp                 0.995455  1.000000  0.990909  1.000000  0.995434  \n",
       "cspAdam             0.957576  1.000000  0.915152  1.000000  0.955696  \n",
       "cspGD               0.975758  1.000000  0.951515  1.000000  0.975155  \n",
       "h30AdamBs12         0.984848  0.984848  0.984848  0.984848  0.984848  \n",
       "hid30Rprop          0.978788  0.996855  0.960606  0.996970  0.978395  \n",
       "cspGDtst            0.979798  1.000000  0.959596  1.000000  0.979381  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performance: f7\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>n_runs</th>\n",
       "      <th>n_folds</th>\n",
       "      <th>epochs</th>\n",
       "      <th>time</th>\n",
       "      <th>size</th>\n",
       "      <th>train-acc</th>\n",
       "      <th>acc</th>\n",
       "      <th>prec</th>\n",
       "      <th>sens</th>\n",
       "      <th>spec</th>\n",
       "      <th>f1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>linGDlr0.1</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>500.0</td>\n",
       "      <td>2.557012</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998485</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.996970</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998483</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linSGDlr0.01bs12</th>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>390.0</td>\n",
       "      <td>4.060837</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.999278</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linAdam</th>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>499.0</td>\n",
       "      <td>2.196307</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.785714</td>\n",
       "      <td>0.712121</td>\n",
       "      <td>0.718750</td>\n",
       "      <td>0.696970</td>\n",
       "      <td>0.727273</td>\n",
       "      <td>0.707692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linAdamLr0.001bs12</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>670.0</td>\n",
       "      <td>9.790519</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.992424</td>\n",
       "      <td>0.987988</td>\n",
       "      <td>0.996970</td>\n",
       "      <td>0.987879</td>\n",
       "      <td>0.992459</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linAdamlr0.01</th>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>400.0</td>\n",
       "      <td>1.808603</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.969697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>linRprop</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>75.0</td>\n",
       "      <td>0.523490</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h30GD</th>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>799.0</td>\n",
       "      <td>3.597783</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.988095</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.984375</td>\n",
       "      <td>0.954545</td>\n",
       "      <td>0.984848</td>\n",
       "      <td>0.969231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h30GDlr0.1</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>999.0</td>\n",
       "      <td>4.708344</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.984848</td>\n",
       "      <td>0.987805</td>\n",
       "      <td>0.981818</td>\n",
       "      <td>0.987879</td>\n",
       "      <td>0.984802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h30GDlr0.001bs12</th>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>799.0</td>\n",
       "      <td>10.685843</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.990981</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.984375</td>\n",
       "      <td>0.954545</td>\n",
       "      <td>0.984848</td>\n",
       "      <td>0.969231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h30AdamBs12</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>500.0</td>\n",
       "      <td>7.636065</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.986364</td>\n",
       "      <td>0.979104</td>\n",
       "      <td>0.993939</td>\n",
       "      <td>0.978788</td>\n",
       "      <td>0.986466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h30AdamLr0.01</th>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>100.0</td>\n",
       "      <td>0.490162</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.969697</td>\n",
       "      <td>0.969697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hid30Rprop</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>110.0</td>\n",
       "      <td>0.678064</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.989394</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.978788</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.989280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>csp</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>100.0</td>\n",
       "      <td>0.189121</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cspAdam</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>300.0</td>\n",
       "      <td>0.312546</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.983333</td>\n",
       "      <td>0.984802</td>\n",
       "      <td>0.981818</td>\n",
       "      <td>0.984848</td>\n",
       "      <td>0.983308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cspGD</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>750.0</td>\n",
       "      <td>0.882205</td>\n",
       "      <td>3.727273</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.977273</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.954545</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.976744</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    n_runs  n_folds  epochs       time      size  train-acc  \\\n",
       "model                                                                         \n",
       "linGDlr0.1               5       22   500.0   2.557012       NaN   1.000000   \n",
       "linSGDlr0.01bs12         1       22   390.0   4.060837       NaN   0.999278   \n",
       "linAdam                  1       22   499.0   2.196307       NaN   0.785714   \n",
       "linAdamLr0.001bs12       5       22   670.0   9.790519       NaN   1.000000   \n",
       "linAdamlr0.01            1       22   400.0   1.808603       NaN   1.000000   \n",
       "linRprop                 5       22    75.0   0.523490       NaN   1.000000   \n",
       "h30GD                    1       22   799.0   3.597783       NaN   0.988095   \n",
       "h30GDlr0.1               5       22   999.0   4.708344       NaN   1.000000   \n",
       "h30GDlr0.001bs12         1       22   799.0  10.685843       NaN   0.990981   \n",
       "h30AdamBs12              5       22   500.0   7.636065       NaN   1.000000   \n",
       "h30AdamLr0.01            1       22   100.0   0.490162       NaN   1.000000   \n",
       "hid30Rprop               5       22   110.0   0.678064       NaN   1.000000   \n",
       "csp                      5       22   100.0   0.189121  0.000000   1.000000   \n",
       "cspAdam                  5       22   300.0   0.312546  0.000000   1.000000   \n",
       "cspGD                    5       22   750.0   0.882205  3.727273   1.000000   \n",
       "\n",
       "                         acc      prec      sens      spec        f1  \n",
       "model                                                                 \n",
       "linGDlr0.1          0.998485  1.000000  0.996970  1.000000  0.998483  \n",
       "linSGDlr0.01bs12    1.000000  1.000000  1.000000  1.000000  1.000000  \n",
       "linAdam             0.712121  0.718750  0.696970  0.727273  0.707692  \n",
       "linAdamLr0.001bs12  0.992424  0.987988  0.996970  0.987879  0.992459  \n",
       "linAdamlr0.01       0.969697  0.969697  0.969697  0.969697  0.969697  \n",
       "linRprop            1.000000  1.000000  1.000000  1.000000  1.000000  \n",
       "h30GD               0.969697  0.984375  0.954545  0.984848  0.969231  \n",
       "h30GDlr0.1          0.984848  0.987805  0.981818  0.987879  0.984802  \n",
       "h30GDlr0.001bs12    0.969697  0.984375  0.954545  0.984848  0.969231  \n",
       "h30AdamBs12         0.986364  0.979104  0.993939  0.978788  0.986466  \n",
       "h30AdamLr0.01       0.969697  0.969697  0.969697  0.969697  0.969697  \n",
       "hid30Rprop          0.989394  1.000000  0.978788  1.000000  0.989280  \n",
       "csp                 1.000000  1.000000  1.000000  1.000000  1.000000  \n",
       "cspAdam             0.983333  0.984802  0.981818  0.984848  0.983308  \n",
       "cspGD               0.977273  1.000000  0.954545  1.000000  0.976744  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show all existing results for all datasets.\n",
    "# use 'dst' to avoid overwriting dataset variable\n",
    "for dst in [eeg_all, eeg_no_outliers, eeg_f7]: # also eeg_f3\n",
    "  print(f\"Performance: {dst.name}\")\n",
    "  display(dst.perf_table)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train Constructive Cascade Networks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "@dataclass\n",
    "class ConstrCascModel:\n",
    "  \"\"\"All ConstrCasc models use same architecture, so just specify training mechanism\"\"\"\n",
    "  desc: str\n",
    "\n",
    "  initNet: nn.Module # will be passed (trainX, **net_kw)\n",
    "  net_kw: dict\n",
    "  initOpt: torch.optim.Optimizer # passed (net.get_parameters(), **opt_kw)\n",
    "  opt_kw: dict\n",
    "  \n",
    "  generate_neuron: any\n",
    "  train_period: any\n",
    "  # and kwargs for each?\n",
    "\n",
    "  k: float = 0.0 # Weight-decay constant\n",
    "\n",
    "  n_epochs: int = 1500\n",
    "\n",
    "  performance: pm.KFoldPerformance = None\n",
    "\n",
    "  def __str__(self):\n",
    "    return f\"{self.desc} ({', '.join([f'{k}={v}' for k,v in self.opt_kw.items()])})\"\n",
    "\n",
    "  def clone(self, **kwargs):\n",
    "    return dataclasses.replace(self,**kwargs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Manually-identified best epochs to evaluate at, for Casper models.\n",
    "eeg_no_outliers.eval_epochs.update({\n",
    "  'csp': 70,\n",
    "  'cspAdam': 1000,\n",
    "  'cspGD': 500\n",
    "})\n",
    "\n",
    "eeg_all.eval_epochs.update({\n",
    "  'csp': 60,\n",
    "  'cspAdam': 175,\n",
    "  'cspGD': 750\n",
    "})\n",
    "\n",
    "eeg_f7.eval_epochs.update({\n",
    "  'csp': 100,\n",
    "  'cspAdam': 300,\n",
    "  'cspGD': 750\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Weight variability for candidates\n",
    "# ** WHEN to add a new neuron\n",
    "\n",
    "importlib.reload(constr_casc)\n",
    "Casper = constr_casc.Casper\n",
    "#Cascor = constr_casc.Cascor\n",
    "\n",
    "# For casper models, optimiser lr is specified by L1,L2,L3 (not opt_kw)\n",
    "\n",
    "# Base models\n",
    "casper = ConstrCascModel(\"Casper Original\",\n",
    "    Casper, {'L1': 0.2, 'L2': 0.005, 'L3': 0.001},\n",
    "    Rprop, {'etas':(0.5, 1.2), 'step_sizes':(1e-06, 50)},\n",
    "    lambda net: net.gen_neuron(), lambda net: 15 * (1.5**net.H), k=0)\n",
    "\n",
    "constrCascModels = {\n",
    "  #'cspWD': casper.clone(desc=\"Casper WD\", k=0.005),\n",
    "  'csp': casper,\n",
    "  #'cspAdamW': casper.clone(initOpt=torch.optim.AdamW, desc=\"Casper AdamW\",opt_kw={'weight_decay': 0.005}, n_epochs = 3000),\n",
    "  'cspAdam': casper.clone(initOpt=Adam, desc=\"Casper Adam\",opt_kw={},\n",
    "    net_kw={'L1': 0.01, 'L2': 0.003, 'L3': 0.0001}, n_epochs = 3000),\n",
    "  'cspGD': casper.clone(initOpt=SGD, opt_kw={}, desc=\"Casper GD\",\n",
    "    net_kw={'L1': 0.1, 'L2': 0.01, 'L3': 0.001}, n_epochs = 3000),\n",
    "}\n",
    "\n",
    "improve_thresh = 0.02 # 2% improvement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "19eab81f6a7c44c6b0fc9143a5965b26",
       "version_major": 2,
       "version_minor": 0
      },
      "image/png": "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",
      "text/html": [
       "\n",
       "            <div style=\"display: inline-block;\">\n",
       "                <div class=\"jupyter-widgets widget-label\" style=\"text-align: center;\">\n",
       "                    Figure\n",
       "                </div>\n",
       "                <img src='data:image/png;base64,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' width=800.0/>\n",
       "            </div>\n",
       "        "
      ],
      "text/plain": [
       "Canvas(header_visible=False, toolbar=Toolbar(toolitems=[('Home', 'Reset original view', 'home', 'home'), ('Bac…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Set up real-time training plot\n",
    "importlib.reload(pm)\n",
    "if 'cc_rtf' in globals():\n",
    "  plt.close(cc_rtf.fig) # Release old figure memory.\n",
    "cc_rtf = pm.RealtimeTrainingFig('epoch')\n",
    "display(cc_rtf.fig.canvas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "time.sleep(0.5) # In case 'run all' is used; need a delay between generating and updating figure."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "record_every = 20 # epochs\n",
    "dataset = eeg_f7\n",
    "n_folds = dataset.n_subjects #for final results\n",
    "n_runs = 5\n",
    "\n",
    "metric_names = ['epoch', 'time', 'train_loss', 'train_accuracy', 'test_loss', 'test_accuracy', 'hidden_neurons']\n",
    "\n",
    "for M_id, M in constrCascModels.items():\n",
    "  # Re-initialise RNG for this model, and generate seeds for each fold/run.\n",
    "  torch.manual_seed(metaseed)\n",
    "  M_seeds = torch.randint(2**30, 2**32, (n_runs, n_folds))\n",
    "\n",
    "  plot_every = max(100, M.n_epochs//10)\n",
    "  n_records = int(M.n_epochs/record_every) + 1 # Per run and fold\n",
    "\n",
    "  # Metrics for every record of every fold of every run\n",
    "  all_confusions = torch.zeros(n_runs, n_folds, n_records, 2, 2)\n",
    "  all_metrics = torch.zeros(n_runs, n_folds, n_records, len(metric_names))\n",
    "\n",
    "  for run_idx in range(n_runs):\n",
    "    cc_rtf.reset(M.n_epochs)\n",
    "    # Leave-one-subject-out cross validation (via kf)\n",
    "    for fold_idx, (train_idx, test_idx) in enumerate(dataset.split()):\n",
    "      if fold_idx >= n_folds:\n",
    "        break\n",
    "      # No need for dataloader here, since no batching.\n",
    "      trainX, trainY = dataset[train_idx]\n",
    "      testX, testY = dataset[test_idx]\n",
    "\n",
    "      # Initialise network for this fold.\n",
    "      torch.manual_seed(M_seeds[run_idx, fold_idx])\n",
    "      stage_rng = torch.Generator() # So that each 'stage' has same neurons added, even if they last different numbers of periods for different models\n",
    "      net = M.initNet(trainX)\n",
    "      param_groups = net.get_parameters()\n",
    "      optimiser = M.initOpt(param_groups, **M.opt_kw)\n",
    "      # Optimiser will be overwritten each time a new neuron is added.\n",
    "      # If M.opt_kw specifies lr, this will be overwritten by individual\n",
    "      # parameter groups in net.get_parameters().\n",
    "      params = [p for group in param_groups for p in group['params']]\n",
    "\n",
    "      # Plotting\n",
    "      cc_rtf.begin_fold(f\"(Run {run_idx+1}/{n_runs}, Fold {fold_idx+1}/{n_folds}) {str(M)}\")\n",
    "      fold_metrics = pm.Metrics(all_metrics[run_idx, fold_idx, :, :], metric_names)\n",
    "\n",
    "      # Initial error and final epoch of current stage (err not actually rms)\n",
    "      rms_err_prev = bceLogitLossMean(net.compute_output(), trainY).item()\n",
    "      period_epoch_max = M.train_period(net)\n",
    "      hepoch = 0 # Epochs since last hidden neuron\n",
    "\n",
    "      # Training loop\n",
    "      t0 = time.time() # Start time\n",
    "      for epoch_idx in range(M.n_epochs):\n",
    "        # Train using optimiser.\n",
    "        # For Casper, this trains the whole network.\n",
    "        # For Cascor, this trains the output layer only.\n",
    "        pred_logits = net.compute_output()\n",
    "        train_loss = bceLogitLossMean(pred_logits, trainY)\n",
    "        optimiser.zero_grad()\n",
    "        train_loss.backward() # TODO: inputs=... to avoid computing unnecessary gradients.\n",
    "\n",
    "        # Apply weight decay\n",
    "        if M.k != 0.0:\n",
    "          for p in params:\n",
    "            p.grad -= M.k * torch.sign(p) * (p**2) * (0.5**(0.01*hepoch))\n",
    "\n",
    "        # Update weights\n",
    "        optimiser.step()\n",
    "\n",
    "        # Occasionally record loss and accuracy for training/test sets.\n",
    "        if (epoch_idx % record_every == 0) or (epoch_idx == M.n_epochs-1):\n",
    "          with torch.inference_mode(): # Disables autograd temporarily.\n",
    "            train_pred = F.sigmoid(pred_logits).round()\n",
    "            train_acc = torch.mean(train_pred == trainY, dtype=torch.float32).item()\n",
    "            test_logits = net(testX) # .compute_output() is only for training data\n",
    "            test_loss, test_acc = loss_accuracy(test_logits, testY)\n",
    "            test_confusion = pm.confusion_matrix(F.sigmoid(test_logits), testY)\n",
    "\n",
    "            all_confusions[run_idx, fold_idx, fold_metrics.n_rec,:,:] = test_confusion\n",
    "            fold_metrics.update(epoch_idx, time.time()-t0, train_loss, train_acc, test_loss, test_acc, net.H)\n",
    "        \n",
    "        # Update plot occasionally\n",
    "        if (epoch_idx % plot_every == 0) or (epoch_idx == M.n_epochs-1):\n",
    "          cc_rtf.update(fold_metrics)\n",
    "        \n",
    "        # Check if it's time to end\n",
    "        hepoch += 1\n",
    "        #if epoch_idx >= M.n_epochs-1:\n",
    "        #  break\n",
    "        if epoch_idx >= period_epoch_max: # or check if need new neuron?\n",
    "          # Add new neuron if needed (i.e. start new stage).\n",
    "          # Skip if train_loss < 1e-7, to avoid adding copious neurons\n",
    "          rms_err = train_loss\n",
    "          if (rms_err > 1e-7) and (rms_err_prev - rms_err) / rms_err_prev < improve_thresh:\n",
    "            # Nice edge case: rms_err_prev=0 -> NaN !< 0 -> False\n",
    "            # Set seed for this stage, using stage_rng\n",
    "            torch.manual_seed(\n",
    "              torch.randint(2**30, 2**32, (1,), generator=stage_rng).item()\n",
    "            )\n",
    "            net.add_neuron(M.generate_neuron(net)) # For Cascor, this would take time (which should count as epochs, but don't right now)\n",
    "            optimiser = M.initOpt(net.get_parameters(), **M.opt_kw)\n",
    "            hepoch = 0\n",
    "          # Either way, update err_prev and period_epoch_max\n",
    "          rms_err_prev = rms_err\n",
    "          period_epoch_max = epoch_idx + M.train_period(net)\n",
    "  \n",
    "  # Compute statistics over all folds/runs\n",
    "  total_confusion = all_confusions.sum(dim=(0,1)) #(n_records, num_metrics)\n",
    "  avg_metrics = pm.Metrics(all_metrics.mean(dim=(0,1)), metric_names, n_records)\n",
    "\n",
    "  # Record performance for this model\n",
    "  M_perf = pm.KFoldPerformance(total_confusion, avg_metrics, n_folds, n_runs)\n",
    "  dataset.perfs[M_id] = M_perf\n",
    "  \n",
    "  # Plot this model\n",
    "  fold_fig = M_perf.plot(f\"{n_runs}-run {n_folds}-fold avg: {str(M)}\")\n",
    "  display(fold_fig)\n",
    "  plt.close(fold_fig) # Avoid large memory use from many open figures\n",
    "\n",
    "dataset.recompute_model_perfs(constrCascModels.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\ncspGDMin, cspGDBestIdx = torch.min(eeg_all.perfs['cspGD'].train_metrics.test_loss, dim=0)\\n\\ncspGDBestEpoch = eeg_all.perfs['cspGD'].train_metrics.epoch[cspGDBestIdx].item()\\n\\nprint(cspGDMin, cspGDBestEpoch)\\n\""
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "cspGDMin, cspGDBestIdx = torch.min(eeg_all.perfs['cspGD'].train_metrics.test_loss, dim=0)\n",
    "\n",
    "cspGDBestEpoch = eeg_all.perfs['cspGD'].train_metrics.epoch[cspGDBestIdx].item()\n",
    "\n",
    "print(cspGDMin, cspGDBestEpoch)\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot size vs epoch for each constrCasc model\n",
    "dst = dataset\n",
    "for M_id, M in constrCascModels.items():\n",
    "  M_perf = dst.perfs[M_id]\n",
    "  fig, ax = plt.subplots()\n",
    "  ax.plot(M_perf.train_metrics.epoch, M_perf.train_metrics.hidden_neurons)\n",
    "  ax.set_title(f\"{M_id} avg hidden neurons vs epoch\")\n",
    "  ax.set_xlabel(\"Epoch\")\n",
    "  ax.set_ylabel(\"Hidden neurons\")\n",
    "  display(fig)\n",
    "  plt.close(fig)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}