0preprocess_ABCD_get_metrics.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from sympy.solvers import solve\n",
    "from sympy import Symbol\n",
    "import scipy.stats as sstats\n",
    "\n",
    "from os import path\n",
    "\n",
    "from scripts.utils import SimulateData\n",
    "from stopsignalmetrics import SSRTmodel, StopData\n",
    "\n",
    "from scipy.stats import exponnorm, norm, norminvgauss\n",
    "from scipy import stats\n",
    "\n",
    "from batch_files.generate_remaining_sim_cmds import replace"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "OVERWRITE_MUS = False"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# __Preprocessing__"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Prepare data for use by stopsignalmetrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data = pd.read_csv('abcd_data/abcd_w_finger_press.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8464"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(abcd_data.NARGUID.unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "for rt_col in ['go_rt_adjusted', 'stop_rt_adjusted']:\n",
    "    abcd_data.loc[abcd_data['finger_press'].isnull(), rt_col] = np.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data['block'] = np.NaN\n",
    "abcd_data.loc[abcd_data['TrialNum'] < 180, 'block'] = '1'\n",
    "abcd_data.loc[abcd_data['TrialNum'] >= 180, 'block'] = '2'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data['choice_accuracy'] = np.where(\n",
    "    abcd_data['finger_press'].notnull(),\n",
    "    np.where(\n",
    "        abcd_data['finger_press']==abcd_data['correct_response'],\n",
    "        1,\n",
    "        0),\n",
    "    np.nan\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8464"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(abcd_data.NARGUID.unique())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.5 replace IDs with ` (backticks) in them\n",
    "#### shell scripts do not interpret them well"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def strip_backticks(string):\n",
    "    return string.replace('`', '')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "original_unique_IDs = abcd_data.NARGUID.unique()\n",
    "fix_id_map = {i:strip_backticks(i) for i in abcd_data.NARGUID.unique() if '`' in i}\n",
    "abcd_data.NARGUID = abcd_data.NARGUID.replace(fix_id_map)\n",
    "assert len(original_unique_IDs)==len(abcd_data.NARGUID.unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data.to_csv('abcd_data/minimal_abcd_with_issue_3.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Drop Issue 3 people"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data_w_issue_3 = pd.read_csv('abcd_data/minimal_abcd_with_issue_3.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1.,  0., nan])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "abcd_data_w_issue_3['correct_go_response'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "n affected: 226\n",
      "p affect: 0.03125432167058498\n",
      "n remaining: 8238\n"
     ]
    }
   ],
   "source": [
    "issue_3_people = abcd_data_w_issue_3.loc[(abcd_data_w_issue_3['stop_rt_adjusted'] < 50) & (abcd_data_w_issue_3['stop_rt_adjusted'] > 0) & (abcd_data_w_issue_3['SSDDur'] ==50), 'NARGUID'].unique()\n",
    "\n",
    "print('n affected:', len(issue_3_people))\n",
    "print('p affect:', len(issue_3_people)/ 7231)\n",
    "\n",
    "abcd_data = abcd_data_w_issue_3[~abcd_data_w_issue_3.NARGUID.isin(issue_3_people)].copy()\n",
    "print('n remaining:', abcd_data.NARGUID.nunique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data.to_csv('abcd_data/minimal_abcd_clean.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# __Metrics for Simulation__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data = pd.read_csv('abcd_data/minimal_abcd_clean.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. get a distribution of mean go RTs and SSRTs to sample from"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "variable_dict = {\n",
    "   \"columns\": {\n",
    "      \"ID\": \"NARGUID\", #subject identifier\n",
    "      \"condition\": \"trial_type\", #col with trial types \n",
    "      \"correct_response\": \"correct_response\", #col with correct reponse codes\n",
    "      \"response\": \"finger_press\", #col with actual response codes \n",
    "      \"SSD\": \"SSDDur\", #col with stop signal delay \n",
    "      \"block\": \"block\", #col with which block a trial is accuring during\n",
    "      \"goRT\": \"go_rt_adjusted\", # col with go reaction time recording \n",
    "      \"stopRT\": \"stop_rt_adjusted\", #col with stop failure reaction time recording\n",
    "      \"choice_accuracy\": \"choice_accuracy\" #col with whether a response was correct\n",
    "   },\n",
    "   \"key_codes\": {\n",
    "      \"go\": \"GoTrial\", # cell values for go trials  \n",
    "      \"stop\": \"StopTrial\",  #cell values for stop trials \n",
    "      \"correct\": 1.0,\n",
    "       \"incorrect\": 0.0,\n",
    "       \"noResponse\": np.nan\n",
    "   }\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dropping 31 subs for having P(respond|signal) == 1 or 0\n"
     ]
    }
   ],
   "source": [
    "abcd_ssrt = StopData(var_dict=variable_dict, compute_acc_col=False)\n",
    "\n",
    "abcd_proc = abcd_ssrt.fit_transform(abcd_data) \n",
    "\n",
    "ssrt_model = SSRTmodel(model='replacement')\n",
    "\n",
    "ssrt_metrics = ssrt_model.fit_transform(abcd_proc, level='group')\n",
    "\n",
    "problem_subs = ssrt_metrics[ssrt_metrics.SSRT.isnull()].index\n",
    "\n",
    "print(f'dropping {len(problem_subs)} subs for having P(respond|signal) == 1 or 0')\n",
    "\n",
    "ssrt_metrics = ssrt_metrics[ssrt_metrics.SSRT.notnull()].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "31"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(problem_subs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['0E4CT74P', '0LA6MBBY', '1PN1MPWF', '2L2B6VXP', '3FJ6DHVU', '4DVGGJE9',\n",
       "       '4HWVY2FD', '5DJB8FFX', '7YA52BU1', '7ZRUXUX9', '8VA4L6RD', 'AB3XR5P1',\n",
       "       'AF517NF3', 'B3438R23', 'C2ZW147D', 'E9KNUZDH', 'EC1NEE93', 'F86XUHXD',\n",
       "       'K9MXKUUB', 'KDZR86PA', 'KP8W0VHU', 'L3258K1X', 'L5W5CDJU', 'L9NJJJBZ',\n",
       "       'LW0J7PJB', 'MR60CAWU', 'T5EXYEGX', 'TY02C986', 'U1THG28C', 'YK2J75TH',\n",
       "       'vcahyykd'],\n",
       "      dtype='object', name='ID')"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "problem_subs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "ssrt_metrics.to_csv('abcd_data/abcd_ssrt_metrics.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# def get_mus(sub_row):\n",
    "#     sub_row = sub_row.copy()\n",
    "#     mu_dict = {}\n",
    "#     # init go and stop vars\n",
    "#     g = Symbol('g')\n",
    "#     s = Symbol('s')\n",
    "    \n",
    "#     # formulas (must be solved as \"expression = 0\")\n",
    "#     # subject_{go/ss}rt = threshold/mu_go + nondecision_time\n",
    "#     # threshold = 100\n",
    "#     # nondecision_time = 50\n",
    "    \n",
    "#     go_sol = solve((sub_row['mean_go_RT'].values[0] - 50) * g - 100, g)\n",
    "# #     go_sol = solve(100/g + 50 - sub_row['mean_go_RT'].values[0], g)\n",
    "#     assert len(go_sol) == 1, f\"{len(go_sol)} solutions found based on {sub_row['mean_go_RT']}: {go_sol}\"\n",
    "#     mu_dict['go'] = float(go_sol[0])\n",
    "    \n",
    "#     stop_sol = solve(100/s + 50 - sub_row['SSRT'].values[0], s)\n",
    "#     assert len(stop_sol) == 1, f\"{len(stop_sol)} solutions found based on {sub_row['SSRT']}: {stop_sol}\"\n",
    "#     mu_dict['stop'] = float(stop_sol[0])\n",
    "    \n",
    "#     return mu_dict\n",
    "\n",
    "# mu_df = ssrt_metrics.groupby('ID').apply(get_mus)\n",
    "# mu_dict = mu_df.to_dict()\n",
    "# mu_dict['prob_subs'] = list(problem_subs)\n",
    "# mu_df = mu_df.apply(pd.Series)\n",
    "\n",
    "# json_mu_path = 'abcd_data/individual_mus.json'\n",
    "# with open(json_mu_path, 'w') as jp:\n",
    "#     json.dump(mu_dict, jp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(12,6))\n",
    "ax = sns.distplot(ssrt_metrics['SSRT'], kde=False, norm_hist=True, label='SSRT')\n",
    "ssrt_loc,ssrt_var = norm.fit(ssrt_metrics['SSRT'])\n",
    "xx = np.arange(-500, 3000, 1)\n",
    "ax.plot(xx, norm.pdf(xx, loc=ssrt_loc, scale=ssrt_var), 'k', lw=2)  \n",
    "_ = sns.distplot(ssrt_metrics['mean_go_RT'], kde=False,norm_hist=True, ax=ax, label='goRT')\n",
    "goRT_loc,goRT_var = norm.fit(ssrt_metrics['mean_go_RT'])\n",
    "ax.plot(xx, norm.pdf(xx, loc=goRT_loc, scale=goRT_var), 'k', lw=2) \n",
    "_ = plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# repeating but giving a reasonable scale for the SSRT distribution\n",
    "\n",
    "fig = plt.figure(figsize=(12,6))\n",
    "ax = sns.distplot(ssrt_metrics['SSRT'], kde=False, norm_hist=True, label='SSRT')\n",
    "xx = np.arange(-500, 3000, 1)\n",
    "ax.plot(xx, norm.pdf(xx, loc=ssrt_loc, scale=85), 'k', lw=2)  \n",
    "_ = sns.distplot(ssrt_metrics['mean_go_RT'], kde=False,norm_hist=True, ax=ax, label='goRT')\n",
    "ax.plot(xx, norm.pdf(xx, loc=goRT_loc, scale=goRT_var), 'k', lw=2) \n",
    "_ = plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAs8AAAF0CAYAAADcl9ScAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO3deZxcVZ338c8v3dkTQkgCAgESSECCQIAIKj6oBFlcCGgYAi6oKC7goDOMg/qIiDIzqCOjA6hRNnnYNIiGRVE2FZRAWEQDhERAaEAIJIRs3Ul3n+ePuhWappdau7q7Pu/Xq1+pOvfcW+dei87XX865N1JKSJIkSerdkFoPQJIkSRooDM+SJElSgQzPkiRJUoEMz5IkSVKBDM+SJElSgQzPkiRJUoEaaz2AYkycODFNmTKl1sOQJEnSIHbvvfe+kFKa1NW2ARWep0yZwuLFi2s9DEmSJA1iEfH37rY5bUOSJEkqkOFZkiRJKpDhWZIkSSrQgJrzLEmSpOrZtGkTTU1NNDc313oofWLEiBFMnjyZoUOHFryP4VmSJEkANDU1MXbsWKZMmUJE1Ho4VZVS4sUXX6SpqYmpU6cWvJ/TNiRJkgRAc3MzEyZMGPTBGSAimDBhQtFVdsOzJEmSNquH4JxXyrkaniVJktSvnH322eyxxx7stddezJw5k0WLFnH99dezzz77sPfeezNjxgx++MMfAnDmmWey/fbbM3PmTGbMmMGVV14JwMknn7y5beTIkcycOZOZM2eyYMGCssbmnGdJkiR16YpFT1b0eMcfsGOvff70pz9x/fXXc9999zF8+HBeeOEF1q1bx9FHH83dd9/N5MmTaWlp4Yknnti8z+c//3lOO+00li1bxn777cfcuXM5//zzAXjiiSd4z3vewwMPPFCRcyio8hwRh0fE0ohYHhGnd7F9eERcnW1fFBFTsvYJEXFbRKyNiPM67TMsIuZHxKMR8UhEvL8SJyRJkqSB69lnn2XixIkMHz4cgIkTJzJ27FhaW1uZMGECAMOHD2e33XZ7zb7Tp09n1KhRrFq1qmrj6zU8R0QDcD5wBDADOC4iZnTqdiKwKqU0DTgXOCdrbwa+ApzWxaG/DDyfUto1O+7vSjoDSZIkDRqHHnooTz31FLvuuiuf+cxn+N3vfsdWW23FkUceyU477cRxxx3H5ZdfTnt7+2v2ve+++5g+fTpbb7111cZXSOV5f2B5SumxlNJG4CpgTqc+c4BLs9cLgNkRESmldSmlO8iF6M4+BvwnQEqpPaX0QklnIEmSpEFjzJgx3HvvvcyfP59JkyZx7LHHcskll/DjH/+YW265hf33359vf/vbfOxjH9u8z7nnnstuu+3GAQccwJlnnlnV8RUSnrcHnurwvilr67JPSqkVWA1M6O6AEbFl9vLrEXFfRPwsIrbppu9JEbE4IhavWLGigOFKtfHoo49y/PHHV2xOlSRJ9aqhoYG3v/3tfO1rX+O8887jmmuuAWDPPffk85//PL/97W83t0FuzvPSpUu5+uqr+fCHP1zVh7wUEp67uodHKqFPR43AZODOlNK+wJ+Ab3fVMaU0P6U0K6U0a9KkSQUMV6qNL3zhC1x55ZV84AMfqPVQJEkasJYuXcqyZcs2v3/ggQfYZpttuP3221/VttNOO71m3/e9733MmjWLSy+99DXbKqWQu200ATt0eD8ZeKabPk0R0QiMA1b2cMwXgfXAtdn7n5GbNy0NWHfccQcADz30UI1HIknSwLV27Vo++9nP8tJLL9HY2Mi0adP47ne/yyc/+Uk++clPMnLkSEaPHs0ll1zS5f5nnHEGxx9/PJ/4xCcYMqTyd2WOlHoqEEMWhh8FZgNPA/cAx6eUlnToczKwZ0rpUxExD3hfSumfOmz/CDArpXRKh7argPkppVuz7e9OKR3T01hmzZqVFi9eXOQpStXX3t7OqFGjaGlpAWD16tVsscUWNR6VJEnFefjhh9l9991rPYw+1dU5R8S9KaVZXfXvtfKcUmqNiFOAm4AG4KKU0pKIOAtYnFJaCFwIXBYRy8lVnOd1+PAngC2AYRFxFHBoSukh4N+zff4HWAF8tOizlfqJlStXbg7OkLun5F577VXDEUmSpGoo6CEpKaUbgRs7tZ3R4XUz0GXVOKU0pZv2vwMHFTpQqT978cUXX/X+hRe8eYwkSYORj+eWKsDwLElSfTA8SxWwcuWr18caniVJGpwMz1IFdK48d34vSZIGB8OzVAGdw/KqVatqNBJJklRNhmepAvLhefLkyQCsWbOmlsORJKkufOQjH2Hq1KnMnDmTvffem1tuuQWAo48+mpkzZzJt2jTGjRvHzJkzmTlzJn/84x/L/syC7rYhqWerV68GcuG5qanJ8CxJGhwWX1zZ482q/J2Jv/WtbzF37lxuu+02TjrpJJYtW8a11+aew3f77bfz7W9/m+uvv75in2d4lipg7dq1AGy33XaAlWdJkkr19a9/ncsvv5wddtiBiRMnst9++3HIIYfwqU99ivXr17PLLrtw0UUXMX78+Fft9+Y3v5mnn3666uNz2oZUAevWrQNg2223BQzPkiSVYvHixVxzzTXcf//9/PznPyf/ZOkPf/jDnHPOOTz44IPsueeefO1rX3vNvr/+9a856qijqj5GK89SBeQrz6973esAw7MkSaW44447mDNnDiNHjgTgve99L+vWreOll17ibW97GwAnnHACxxzzyrP5/u3f/o0vfOELPP/889x1111VH6OVZ6kCrDxLklS+lFLR+3zrW99i+fLlfOMb3+CEE06owqhezfAsVUA+PFt5liSpdG9961u57rrraG5uZu3atdxwww2MHj2a8ePH84c//AGAyy67bHMVOm/IkCGceuqptLe3c9NNN1V1jE7bkCrAaRuSJJXvjW98I0ceeSR77703O+20E7NmzWLcuHFceumlmxcM7rzzzlx88WvvAhIR/N//+3/55je/yWGHHVa1MUYp5fFamTVrVspPHJf6k8mTJ/P000/z97//nalTp9Le3s6mTZtobPT/n0qSBo6HH36Y3XffvaZjWLt2LWPGjGH9+vUcdNBBzJ8/n3333bdqn9fVOUfEvSmlWV319292qQLy0zbGjh3L6NGjWbNmDevWrWPcuHE1HpkkSQPLSSedxEMPPURzczMnnHBCVYNzKQzPUgXkp22MHj2aUaNGsWbNGjZs2GB4liSpSFdccUWth9AjFwxKZdq4cSOtra00NjYybNiwzbfX2bBhQ41HJkmSKs3wLJUpP2VjzJgxAJvD8/r162s2JkmSSjWQ1sOVq5RzNTxLZeo4ZQOw8ixJGrBGjBjBiy++WBcBOqXEiy++yIgRI4razznPUpm6qzwbniVJA83kyZNpampixYoVtR5KnxgxYgSTJ08uah/Ds1SmzpXnUaNGAYZnSdLAM3ToUKZOnVrrYfRrTtuQypSvPHeetuGcZ0mSBh/Ds1Sm5uZm4JWKs9M2JEkavAzPUpny4Tm/4MDwLEnS4GV4lsqUD8/Dhw8HXqlAO21DkqTBx/AslcnKsyRJ9cPwLJXJ8CxJUv0wPEtlMjxLklQ/DM9SmTqHZ+c8S5I0eBmepTJZeZYkqX4YnqUyGZ4lSaofBYXniDg8IpZGxPKIOL2L7cMj4ups+6KImJK1T4iI2yJibUSc182xF0bEX8s5CamWDM+SJNWPXsNzRDQA5wNHADOA4yJiRqduJwKrUkrTgHOBc7L2ZuArwGndHPt9wNrShi71D92FZ+c8S5I0+BRSed4fWJ5SeiyltBG4CpjTqc8c4NLs9QJgdkRESmldSukOciH6VSJiDPAvwDdKHr3UD3S3YNDKsyRJg08h4Xl74KkO75uyti77pJRagdXAhF6O+3Xgv4Eey3MRcVJELI6IxStWrChguFLf6hye808abGlpqdmYJElSdRQSnqOLtlRCn1c6R8wEpqWUru3tw1NK81NKs1JKsyZNmtRbd6nPdRee8+2SJGnwKCQ8NwE7dHg/GXimuz4R0QiMA1b2cMw3A/tFxBPAHcCuEXF7YUOW+hcrz5Ik1Y9CwvM9wPSImBoRw4B5wMJOfRYCJ2Sv5wK3ppS6rTynlL6fUtoupTQFeCvwaErp7cUOXuoPOofn/J+GZ0mSBp/G3jqklFoj4hTgJqABuCiltCQizgIWp5QWAhcCl0XEcnIV53n5/bPq8hbAsIg4Cjg0pfRQ5U9Fqg0rz5Ik1Y9ewzNASulG4MZObWd0eN0MHNPNvlN6OfYTwBsKGYfUHxmeJUmqHz5hUCqT4VmSpPpheJbKZHiWJKl+GJ6lMvUUnntYNytJkgYgw7NUps7heciQITQ25pYTbNq0qWbjkiRJlWd4lsrUOTx3fO2DUiRJGlwMz1IZUkqbA3J+ukbH1857liRpcDE8S2XYuHEjAMOGDWPIkFf+czI8S5I0OBmepTJ0NWUDDM+SJA1WhmepDIZnSZLqi+FZKkNX8507vjc8S5I0uBiepTLkw7HhWZKk+mB4lsrQccFgR/lpHIZnSZIGF8OzVIZ8eLbyLElSfTA8S2XorvKcD88+JEWSpMHF8CyVIV9Z7i48W3mWJGlwMTxLZeit8mx4liRpcDE8S2UwPEuSVF8Mz1IZXDAoSVJ9MTxLZfBWdZIk1RfDs1QGFwxKklRfDM9SGZzzLElSfTE8S2UwPEuSVF8Mz1IZelsw6ENSJEkaXAzPUhmsPEuSVF8Mz1IZXDAoSVJ9MTxLZeit8pzfLkmSBgfDs1SG7sJz/r3hWZKkwcXwLJWhuwWDhmdJkgYnw7NUBivPkiTVF8OzVIbeFgwaniVJGlwKCs8RcXhELI2I5RFxehfbh0fE1dn2RRExJWufEBG3RcTaiDivQ/9REXFDRDwSEUsi4r8qdUJSX7LyLElSfek1PEdEA3A+cAQwAzguImZ06nYisCqlNA04Fzgna28GvgKc1sWhv51Sej2wD3BgRBxR2ilItWN4liSpvhRSed4fWJ5SeiyltBG4CpjTqc8c4NLs9QJgdkRESmldSukOciF6s5TS+pTSbdnrjcB9wOQyzkOqCRcMSpJUXwoJz9sDT3V435S1ddknpdQKrAYmFDKAiNgSeC9wSzfbT4qIxRGxeMWKFYUcUuozVp4lSaovhYTn6KItldDntQeOaASuBL6XUnqsqz4ppfkppVkppVmTJk3qdbBSX+puwaDhWZKkwamQ8NwE7NDh/WTgme76ZIF4HLCygGPPB5allP6ngL5Sv2PlWZKk+lJIeL4HmB4RUyNiGDAPWNipz0LghOz1XODWlFKPleeI+Aa5kP254oYs9R+GZ0mS6ktjbx1SSq0RcQpwE9AAXJRSWhIRZwGLU0oLgQuByyJiObmK87z8/hHxBLAFMCwijgIOBV4Gvgw8AtwXEQDnpZR+XMmTk6rNBYOSJNWXXsMzQErpRuDGTm1ndHjdDBzTzb5TujlsV/OkpQHFyrMkSfXFJwxKZehtwWB+uyRJGhwMz1IZrDxLklRfDM9SGboLzw0NDQwZMoT29nba2tpqMTRJklQFhmepDN0tGASrz5IkDUaGZ6kM3VWeO7YZniVJGjwMz1IZulsw2LHN8CxJ0uBheJZKlFLaHIyHDh36mu2GZ0mSBh/Ds1Si1tZWILc4sKGh4TXbDc+SJA0+hmepRD0tFgTDsyRJg5HhWSpRT4sFO7YbniVJGjwMz1KJelos2LHd8CxJ0uBheJZKZOVZkqT6Y3iWStRbeM7PhTY8S5I0eBiepRK5YFCSpPpjeJZK5LQNSZLqj+FZKpELBiVJqj+GZ6lEVp4lSao/hmepRIWG53yFWpIkDXyGZ6lELhiUJKn+GJ6lEjltQ5Kk+mN4lkrkgkFJkuqP4VkqkZVnSZLqj+FZKpHhWZKk+mN4lkrkgkFJkuqP4VkqkZVnSZLqj+FZKpELBiVJqj+GZ6lEVp4lSao/hmepRIZnSZLqj+FZKpELBiVJqj+GZ6lEVp4lSao/BYXniDg8IpZGxPKIOL2L7cMj4ups+6KImJK1T4iI2yJibUSc12mf/SLiL9k+34uIqMQJSX3FBYOSJNWfXsNzRDQA5wNHADOA4yJiRqduJwKrUkrTgHOBc7L2ZuArwGldHPr7wEnA9Ozn8FJOQKoVK8+SJNWfQirP+wPLU0qPpZQ2AlcBczr1mQNcmr1eAMyOiEgprUsp3UEuRG8WEdsCW6SU/pRSSsBPgKPKORGprxmeJUmqP4WE5+2Bpzq8b8rauuyTUmoFVgMTejlmUy/HBCAiToqIxRGxeMWKFQUMV+obLhiUJKn+FBKeu5qLnEroU1L/lNL8lNKslNKsSZMm9XBIqW/1VnnOh+r83GhJkjTwFRKem4AdOryfDDzTXZ+IaATGASt7OebkXo4p9WsuGJQkqf4UEp7vAaZHxNSIGAbMAxZ26rMQOCF7PRe4NZvL3KWU0rPAmoh4U3aXjQ8Dvyx69FINOedZkqT609hbh5RSa0ScAtwENAAXpZSWRMRZwOKU0kLgQuCyiFhOruI8L79/RDwBbAEMi4ijgENTSg8BnwYuAUYCv8p+pAHD8CxJUv3pNTwDpJRuBG7s1HZGh9fNwDHd7Dulm/bFwBsKHajU37hgUJKk+uMTBqUSWXmWJKn+GJ6lErlgUJKk+mN4lkpk5VmSpPpjeJZKZHiWJKn+GJ6lErlgUJKk+mN4lkrUW+W5sTF3M5vW1lba29v7bFySJKl6DM9SiXpbMBgRm7dt2rSpz8YlSZKqx/Aslai3ynPHbU7dkCRpcDA8SyUyPEuSVH8Mz1IJ2traaG9vZ8iQIZvnNnfF8CxJ0uBieJZKkA/DQ4cO7bGf4VmSpMHF8CyVIL9YsLvb1OUZniVJGlwMz1IJCpnv3HF7PmxLkqSBzfAslaDY8GzlWZKkwcHwLJWgt6cL5hmeJUkaXAzPUgmsPEuSVJ8Mz1IJenu6YJ7hWZKkwcXwLJWg0MpzflqH4VmSpMHB8CyVwGkbkiTVJ8OzVAIXDEqSVJ8Mz1IJvM+zJEn1yfAslcAFg5Ik1SfDs1QC5zxLklSfDM9SCQqd8+zdNiRJGlwMz1IJnPMsSVJ9MjxLJfA+z5Ik1SfDs1SCYhcMWnmWJGlwMDxLJXDBoCRJ9cnwLJWg2AWDVp4lSRocDM9SCaw8S5JUnwoKzxFxeEQsjYjlEXF6F9uHR8TV2fZFETGlw7YvZu1LI+KwDu2fj4glEfHXiLgyIkZU4oSkvuCCQUmS6lOv4TkiGoDzgSOAGcBxETGjU7cTgVUppWnAucA52b4zgHnAHsDhwAUR0RAR2wP/DMxKKb0BaMj6SQOCCwYlSapPhVSe9weWp5QeSyltBK4C5nTqMwe4NHu9AJgdEZG1X5VSakkpPQ4sz44H0AiMjIhGYBTwTHmnIvUdK8+SJNWnQsLz9sBTHd43ZW1d9kkptQKrgQnd7ZtSehr4NvAk8CywOqX0m64+PCJOiojFEbF4xYoVBQxXqr5CFwxaeZYkaXApJDxHF22pwD5dtkfEeHJV6anAdsDoiPhgVx+eUpqfUpqVUpo1adKkAoYrVZ8LBiVJqk+FhOcmYIcO7yfz2ikWm/tk0zDGASt72PcQ4PGU0oqU0ibg58BbSjkBqRaKnbZh5VmSpMGhkPB8DzA9IqZGxDByC/sWduqzEDghez0XuDWllLL2edndOKYC04G7yU3XeFNEjMrmRs8GHi7/dKS+UeyCQSvPkiQNDo29dUgptUbEKcBN5O6KcVFKaUlEnAUsTiktBC4ELouI5eQqzvOyfZdExE+Bh4BW4OSUUhuwKCIWAPdl7fcD8yt/elJ1uGBQkqT61Gt4Bkgp3Qjc2KntjA6vm4Fjutn3bODsLtq/Cny1mMFK/YULBiVJqk8+YVAqgZVnSZLqk+FZKkGxd9uw8ixJ0uBgeJZK4IJBSZLqk+FZKoG3qpMkqT4ZnqUSFLtg0MqzJEmDg+FZKoELBiVJqk+GZ6kELhiUJKk+GZ6lEhS6YLChoYGIoL29nba2tr4YmiRJqiLDs1SCQivPEeGiQUmSBhHDs1SCQhcMgosGJUkaTAzPUgkKrTyDiwYlSRpMDM9Skdrb22ltbQWgsbGx1/4uGpQkafAwPEtF6lh1johe+1t5liRp8DA8S0UqZspGx35WniVJGvgMz1KRilksCC4YlCRpMDE8S0UqtvLsreokSRo8DM9SkQp9QEqelWdJkgYPw7NUpFIrz4ZnSZIGPsOzVCQXDEqSVL8Mz1KRil0waOVZkqTBw/AsFcnKsyRJ9cvwLBXJBYOSJNUvw7NUJG9VJ0lS/TI8S0UqddqGlWdJkgY+w7NUJBcMSpJUvwzPUpFcMChJUv0yPEtFcsGgJEn1y/AsFckFg5Ik1S/Ds1QkFwxKklS/DM9SkfIV5GIXDFp5liRp4CsoPEfE4RGxNCKWR8TpXWwfHhFXZ9sXRcSUDtu+mLUvjYjDOrRvGRELIuKRiHg4It5ciROSqq3Y8GzlWZKkwaPX8BwRDcD5wBHADOC4iJjRqduJwKqU0jTgXOCcbN8ZwDxgD+Bw4ILseADfBX6dUno9sDfwcPmnI1Vfc3MzACNGjCiov7eqkyRp8Cik8rw/sDyl9FhKaSNwFTCnU585wKXZ6wXA7IiIrP2qlFJLSulxYDmwf0RsARwEXAiQUtqYUnqp/NORqq/UyrPTNiRJGvgKCc/bA091eN+UtXXZJ6XUCqwGJvSw787ACuDiiLg/In4cEaO7+vCIOCkiFkfE4hUrVhQwXKm6nLYhSVL9KiQ8RxdtqcA+3bU3AvsC308p7QOsA14zlxogpTQ/pTQrpTRr0qRJBQxXqq5Sp21YeZYkaeArJDw3ATt0eD8ZeKa7PhHRCIwDVvawbxPQlFJalLUvIBempX7PyrMkSfWrkPB8DzA9IqZGxDByCwAXduqzEDghez0XuDWllLL2edndOKYC04G7U0r/AJ6KiN2yfWYDD5V5LlKfKPVWdYZnSZIGvsbeOqSUWiPiFOAmoAG4KKW0JCLOAhanlBaSW/h3WUQsJ1dxnpftuyQifkouGLcCJ6eU2rJDfxa4PAvkjwEfrfC5SVVR7LQNFwxKkjR49BqeAVJKNwI3dmo7o8PrZuCYbvY9Gzi7i/YHgFnFDFbqD6w8S5JUv3zCoFSkfHi28ixJUv0xPEtFyk/bcMGgJEn1x/AsFanUaRtWniVJGvgMz1KRSp22YeVZkqSBz/AsFanYaRtWniVJGjwMz1KRnLYhSVL9MjxLRSp22ka+X75iLUmSBi7Ds1SkUu+2sWnTJtra2nrpLUmS+jPDs1SkYqdtRMTm6rNTNyRJGtgMz1IRUkpFh2fA8CxJ0iBheJaK0NraSnt7Ow0NDTQ2FvR0e8B5z5IkDRaGZ6kIpVSdwfAsSdJgYXiWilDsnTbyDM+SJA0OhmepCMXeaSPP8CxJ0uBgeJaKYOVZkqT6ZniWiuCcZ0mS6pvhWSqC0zYkSapvhmepCE7bkCSpvhV+o1pJm8Pz6o1wxaIne+1//AE7AoZnSZIGCyvPUhHy4bdx6LCi9stP8zA8S5I0sBmepSLkK8/DhjnnWZKkemR4loqQD8+Nw4qrPBueJUkaHAzPUhHy4XdokdM2DM+SJA0OhmepCPnK81CnbUiSVJcMz1IRnLYhSVJ9MzxLRXDahiRJ9c37PEsUds9mgLuXPwfA0OE+JEWSpHpk5VkqwsaWXPgdNkAfz/3ss89y2mmnse+++7LXXnvxyU9+kkceeaSmY5IkaSCx8iwV4ZXwPPAqz9dddx0f/OAHefnllze3/eUvf+Hiiy/me9/7Hp/61KdqNjZJkgaKgirPEXF4RCyNiOURcXoX24dHxNXZ9kURMaXDti9m7Usj4rBO+zVExP0RcX25JyL1hY3NG4CBN23jF7/4BUcffTQvv/wy7373u7n9B//OXRd/mY++961s2rSJT3/60/zv//5vTcYmSdJA0mt4jogG4HzgCGAGcFxEzOjU7URgVUppGnAucE627wxgHrAHcDhwQXa8vFOBh8s9CamvbMw/YXAAheeHHnqID37wg7S1tXH66adz3XXX8bb9duOAN+zCRWd8jB/84AcAnHrqqdxyyy19Pj5JkgaSQirP+wPLU0qPpZQ2AlcBczr1mQNcmr1eAMyOiMjar0optaSUHgeWZ8cjIiYD7wZ+XP5pSH2j3Gkb+Vvd9ZVNmzZx3HHHsW7dOo4//nj+4z/+g9x/mq/45Cc/yVe+8hVSSnzgAx9g1apVfTpGSZIGkkLC8/bAUx3eN2VtXfZJKbUCq4EJvez7P8AXgPaiRy3VyECb8/zd736XBx98kKlTpzJ//vzXBGcAFl/MV9+9E2+dOZ3nnnuOL33pS306RkmSBpJCwnMXf9uSCuzTZXtEvAd4PqV0b68fHnFSRCyOiMUrVqzofbRSFQ2k8PzMM8/w1a9+FYALLriA0aNHd9u3oWEIPzj9wzQ2NvLDH/6Q+++/v6+GKUnSgFJIeG4CdujwfjLwTHd9IqIRGAes7GHfA4EjI+IJctNADo6I/9fVh6eU5qeUZqWUZk2aNKmA4UrVsykLzwNhweB//dd/sX79eubMmcPhhx8Oiy9+5acLe+yyPaeccgopJc4444w+G6ckSQNJIeH5HmB6REyNiGHkFgAu7NRnIXBC9noucGtKKWXt87K7cUwFpgN3p5S+mFKanFKakh3v1pTSBytwPlJVDZTK89NPP838+fMBOOusswre74tf/CKjRo3i+uuv56677qrW8CRJGrB6Dc/ZHOZTgJvI3RnjpymlJRFxVkQcmXW7EJgQEcuBfwFOz/ZdAvwUeAj4NXBySqmt8qch9Y2BEp7POeccWlpaOOaQWey18d5uq82dbb311px66qkAnH322dUcoiRJA1JB93lOKd2YUto1pbRLSunsrO2MlNLC7HVzSumYlNK0lNL+KaXHOux7drbfbimlX3Vx7NtTSu+p1AlJ1TQQwvNLL73ERRddBMBXTjyyl96v9bnPfY7hw4dz/fXX8+ijj1Z6eJIkDWg+nlsqwqYyw/OGDRsqPqbOLr74YtatW8fs2bPZc9rkovffeuut+eAHc7Oovvvd71Z6eJIkDWiGZ6kILc2lhefhw4cTEWzcuJG2turNXGpra+C28FQAAB0MSURBVOO8//5PAP75iM7PMipAtqDwc4fsBMAll1zifZ8lSerA8CwVYePmu20ML2q/iGDUqFEArF+/vuLjyrvxxht57OkVTN1uIu9+694lH+cN0yYze/Zs1q9fzxVXXFHBEUqSNLAZnqUCtbe10da6iYhg6LDiwjPQJ+H5wgsvBODTc99BQ0N5/3l//OMff9UxJUmS4VkqWMfFgl0+qa8X1Q7PK1as4IYbbqChYQgfOuItZR/vqKOOYvz48dx///0+NEWSpIzhWSrQxhIfkJJX7fB8xRVX0NrayhFv3pPXTRxX9vFGjBjBhz70IcDqsyRJeYZnqUCl3qYur6rhefHFXHLBdwA44T0HVuyYJ75pAgCXX355nz4dUZKk/srwLBWo3PA8evRooDrh+YGlT/LAo08yfovRvPf/lL5QsLO9pu/Avq/fiZdeeokbb7yxYseVJGmgMjxLBerPledLb7gTgOMPO4Dhw4ZW9NjHH3YAAFdeeWVFjytJ0kBkeJYKtKmfznlua2vj6t/eDcCH313+QsHOjn3n/kQE1113HS+//HLFjy9J0kBieJYK9Erlufjb1MEr4XndunUVGxPAnXfeybMvrGbqdhN544ypFT02wORttuKgfXalpaWFX5z7LxU/viRJA4nhWSpQqU8XzKt45Tl7GuBPz/saAP90yBtLuoVeIY7Lpm5c8etFVTm+JEkDheFZKtCmfjjnua2tnQW3LAbgn965f8WO29ncg2fR2NDAzfc8xPPPP1+1z5Ekqb8zPEsFamneAMDwkaNK2r8a4fn39y/luZUvM22Hrdlntx0rdtzOJmw5hsPevAdtbe387Gc/q9rnSJLU3xmepQI1b8jNVR4+orTwXI1b1f305nuA6k7ZyJv3ztzUjQULFlT1cyRJ6s8Mz1KBWjbkKs8jRvWPynNraxvX3HovkAvP1fbeg/ZmaGMDv//973nuueeq/nmSJPVHhmepQC35yvPI0SXtX+nwfPu9j7Bi1Rp22+l17DV9h4ocsyfjxozi0DftQXt7O7/4xS+q/nmSJPVHhmepQM3rc+F5RD+Z89yXUzby5h48C3DqhiSpfjXWegDSQNG8IRd6i1kweMWiJze/vrcpF76XNr3wqnaA4w8obrHfpk2buOa2vpuykXfkQTNpbGzktttu44UXXmDixIl99tmSJPUHVp6lArVk4bnUyvOwESNzx8nu2lGS7N7Ot/3wdFauXsfuU7dlj122L/14Rdpq3Bhmz55NW1sbv/zlL/vscyVJ6i8Mz1KB8pXnEaNKm/OcD88bywnPmfy9nY+Z3XdTNvLmzp2bG4NTNyRJdcjwLBWopcxb1eUr1mVVnsndZePa2+8DYO7sWWUdqxRHTVlPQ8MQbv7tb1m1alWff74kSbVkeJYK1FzmreryTyYsNzz//v5HeeGltey64za8oQ+nbORN3HIsb993N1rb2li4cGGff74kSbVkeJYKVO6t6kaMHgNA8/q1ZY3jZ7fk7rIxd/asPp+ykZeveDt1Q5JUbwzPUoGa15e3YHBkNld6w7p1JY+hra2dn9+Wm7JxzOy+u8tGZ0e/fV+GDAl+85vfsHr16pqNQ5KkvmZ4lgrU0lz8reo6GpFVrJvXr6W9vb2kY9zxwDKeX/kyu0zemr13rf6DUbqzzYRxHLTPbmzcuJHrr7++ZuOQJKmvGZ6lAjVn0zZKvdvGkIaGzcE7f9u7Yi24NXeXjbkH71ezKRt5cw/eD3DqhiSpvhiepQK0tbayqaWFiNi88K8U+akb+acVFqO9vZ1r8uG5BnfZ6Ozod+xLRPCrX/2KNWvW1Ho4kiT1CcOzVICOUzbKqfiOHD0WgA3rig+bf/rTn3j2hdXstO0E9tt9SsljqJTtJo3nwAMPpKWlhRtuuKHWw5EkqU8YnqUCtORvU1fifOe8/JSPDSVUnvPTI+YeXLu7bHQ2d//cvOufzf92jUciSVLfKCg8R8ThEbE0IpZHxOldbB8eEVdn2xdFxJQO276YtS+NiMOyth0i4raIeDgilkTEqZU6Iakamjffpq5C4bmYyvPii2m/+0IWXHEp0D+mbOTlb5d3w51/5uWXX671cCRJqrpew3NENADnA0cAM4DjImJGp24nAqtSStOAc4Fzsn1nAPOAPYDDgQuy47UC/5pS2h14E3ByF8eU+o38HOVS7/Gcl5+2Ueyc57uXPE7T86uYvPV49t9jalljqKTttx7PQfvsSsvGVn75y1/WejiSJFVdIZXn/YHlKaXHUkobgauAOZ36zAEuzV4vAGZH7t+V5wBXpZRaUkqPA8uB/VNKz6aU7gNIKa0BHgb6/lFpUoE2rMs92GRkiXfayBtZ4rSNn92cezDK+w/ejyFD+tdsq3mH7g/AVVddVeORSJJUfYX8Lbw98FSH9028Nuhu7pNSagVWAxMK2Teb4rEPsKjwYUt9a302zWLUmC3KOs7I7CmDxUzbaGtr56rf3g3Ase/cv6zPr4b3H7wfDQ1D+M1vfsPKlStrPRxJkqqqkPDc1cqkVGCfHveNiDHANcDnUkpdTpiMiJMiYnFELF6xYkUBw5Uqb8PafHgeW9ZxRozKP6K78MrzHx54lGdWvMSU7Sbypj13Kevzq2HS+C2Y/cbdaW1t5ec//3mthyNJUlUVEp6bgI6PMpsMPNNdn4hoBMYBK3vaNyKGkgvOl6eUuv0bN6U0P6U0K6U0a9KkSQUMV6q89Wtz/99uZNnhObvPczYNpBBX/PouAI479IB+c5eNzua906kbkqT6UEh4vgeYHhFTI2IYuQWACzv1WQickL2eC9yaUkpZ+7zsbhxTgenA3dl86AuBh1NK36nEiUjVlA/Po8uetpG/z3Nh4Xnjxo0suPVeAI4/7ICyPruajn7HfgwdOpTbbruNf/zjH7UejiRJVdNreM7mMJ8C3ERuYd9PU0pLIuKsiDgy63YhMCEilgP/Apye7bsE+CnwEPBr4OSUUhtwIPAh4OCIeCD7eVeFz02qmPXZtI2RZYfn4hYM3nTTTax6eR17TpvMG6ZNLuuzq2nLsaM44ogjaG9v93HdkqRBrbGQTimlG4EbO7Wd0eF1M3BMN/ueDZzdqe0Oup4PLfVL6ys257nA+zwvvhiAK877AZCbstHfHXvssSxcuJArr7ySU045pdbDkSSpKvrXPa+kfio/baPcu22MHjPuVcfrydr1zSz8/QPAK7eD68+OnLya0SOH88c//pFly5bVejiSJFWF4VkqQKUqz6O3yIXntS+v7rXvglsWs755I2/ZaxpTt+//i2XHjBrBMdnTDy+55JLaDkaSpCoxPEsFqNSt6sZssSUA615+qde+Fy78AwAfO/KtZX1mX/rIe3Jj/clPfkJbW1uNRyNJUuUZnqUCVGraxphx4wFYu3pVj/0e/fs/uOOBZYweOZx/OuSNZX1mXzpo313ZeftJNDU1ccstt9R6OJIkVZzhWSrAK9M2ygvPI0aNpqGhkZbmDWxsae6238XX3QHAPx3yRsaOHlnWZ/aliOAj7zkQgIsvvrjGo5EkqfIMz1IvUkqbK8Vjxm1Z1rEiYnP1ed2aruc9t7a2cukNdwIDa8pG3gnvOZCI4Nprr2XVqp4r7JIkDTQF3apOGoiuWPRkRY6zYd0a2tpaGTFqDEOHDS/7eKO3GMfqlStYu/olxk/c5jXbf/WrX/HsC6vZdcdtOHDv6WV/Xl/b8XUTmD17NjfffDM/+clPOPXUU2s9JEmSKsbKs9SLNS/lqqdjtxxfkeONzc977mbR4Pnnnw/AiXMO6reP4+7Npz/9aQAuuOAC2tvbazwaSZIqx/As9WLNSysBGLvlVhU53ujsjhuvWTS4+GKWLvgPbrrpJkYOH8bH5/yfinxeLRx55JFMnjyZRx99lJtvvrnWw5EkqWIMz1IvXs6H53GVqjx3f7u68352KwAfPOJNbDVuTEU+rxYaH7iMTx+Ze7DL/379C7knJi52AaEkaeBzzrPUi2pVntesfiU8X7HoSbZ55GkuWpi7y8bb3vpGFj2+ssv9D5hamXFU2yeOOoiv/WghN9z5II81Pc/Ok7eu9ZAkSSqblWepF2tXV7byvPluG50qz9fdeg/rm1vYd49dmLbjthX5rFqaNH4L5h26Pyklzr3yt7UejiRJFWF4lnrxyoLBylR8x47LHeflVS9sbtu0sYUrrrsdgHnvPqgin9Mf/NsHDwfgx7/8Pc+92PsjySVJ6u8Mz1IvXn7pRaBy4Xn8pNz0hVUvPL+57fc3LGDFqpeZttO2vHXf3SvyOf3BG6ZNZs7b9qG5ZRP/Y/VZkjQIGJ6lXqxa8RwAW3ZxT+ZS5O/t/FIWnltbN3HdTy4A4CNHz2bIkMH1n+UXP/IuAM5fcKsPTZEkDXiD629pqQpWrvgHAFtNqkx43nJivvKcC+V33HgNK55tYqftJvGON+1Vkc/oTw54wy7MfuPurFnXzLe+9a1aD0eSpLIYnqVerMqH560rs4hviy0nMKShgTUvrWTt6lUs+NF3APjo+99JwyCrOued/Zn3AXDuuefS1NRU49FIklS6wfk3tVQhLc0bWPfyahqHDqvYnOchDQ2M22oSAL+89HxWrXiOqa/fk0MPnFmR4/dHB7xhF+bOnkVzczNf/epXaz0cSZJKZniWerDq+VzVefzErSv6qOzx2RSQWxZcAsBpxx446OY6d/Yfn3k/jY2NXHLJJdx///21Ho4kSSUZ3H9bS2XKz3ceP+l1FT3uhGwKSMvGTRw0aw/222NaRY/fH03fcRtOOeYdtLe3c9IH3kdbW1uthyRJUtEMz1IPnn/mSQAmvm77ih43pQTA0MYG/vVjR1f02P3ZWZ88islbj2fxw09w3nnn1Xo4kiQVzfAs9eAfTz4OwLY77VKxY65e+QJLFv8RgOlTtmObiVtW7Nj93djRI7ng3z8EwJe+9CUefvjhGo9IkqTiGJ6lHjz75GMAbLvj1Iocr729nR+c9S9sWLcGYNDeXaMn7z1oJh9615tZv349xx55KBvu+GGthyRJUsEaaz0AqT/7x1O5yvPrdqhMeP7lJefx4F2/Y/TYcaxbs5qm514s+hiLHl9ZcN8DplbmDiGVdv4XPshdf32Mvyxv4uRv/j8uPPCkii7IlCSpWuqv7CUVqLV1E881/R2oTHi+89fXsmD+fxMRnPnp9zFqxHBWrV7LytVryj72QDN29Eiu/o9PMXL4MC6+7g7+8z//s9ZDkiSpIIZnqRtNf1vKpo0tbDN5CiNHjynrWA/88TZ++I1/A+DUD7+X/zNrD3aduh0AjzxWnw8N2We3nbj8658gIvjyl7/Mj370o1oPSZKkXhmepW4sX/IAANP2KO/hJXffeiPf+cInaGvdxBHzTmTeuw8C4PVTJwPwyGNPlzfQAezod+zH//zLPABOOukkvvuvx9d4RJIk9czwLHXjb1l43nlGaeG5vb2day/6Ht/78mdoa93E8e95G2cc/frN23eftgMADy59vPzBDmD/PO+dfO+0XGj+3Heu5DOf+QwbN26s8agkSeqa4VnqQnt7Ow8u+h0Ar5/5xqL3f/6ZJ/nm509gwfz/BuBT847gsx96z6sWxc16w3QA7lvyNzY0t1Rg1APXZ489hEu+eiLDhjby/e9/n7e85S38+c9/rvWwJEl6De+2oQHnikVPVv0zlv/1Pl564XnGT3odO+26R8H7rVuzmpt+egnX/eQCNrY0M3qLcXzmzO/y/skvvKbvhC3HMmPaDjy0/Cnu+vOjvOOAPSt5CsDAujPHCe85kBlTt2Pu6Rdw7733st9++3HyySdz+umns+2229Z0bJIk5RVUeY6IwyNiaUQsj4jTu9g+PCKuzrYviogpHbZ9MWtfGhGHFXpMqZZ+s+BSAA487KiCbqH292UPcdn/nMWpRx3INT/6DhtbmnnzoUfyzStvZuZb3tHtfu98yz4A/PRXf6jMwAe4N+4xlb9ceRan/NNs2tvb+d73vsfOU3fi4x//OHfeeefmJzNKklQr0dtfRhHRADwKvBNoAu4BjkspPdShz2eAvVJKn4qIecDRKaVjI2IGcCWwP7AdcDOwa7Zbj8fsyqxZs9LixYuLP0sNKtWuPD+46Pecc+qHaBw6jG9ffSuTttvhVdvb29p4/ukn+fuyh3jo3j/yl7vv4LmmJzZv32PWWzj6Y6fynolP9fpZ69Y3c+Snv8G6Dc189ZTjOOKg/Sp9OhXXVxXqB5Y+yVk/Xsi1t9+3uW377bfn0EMP5eCDD2afffZht912o7HRf0CTJFVWRNybUprV5bYCwvObgTNTSodl778IkFL6zw59bsr6/CkiGoF/AJOA0zv2zffLduvxmF2pRXh+8MEHaWl5ZT5qSulV1a/O77vr01m+rfOfPbV19Tmdtbe39ziWrvYrdizVGO8dy17IdyLRab/O++Zf5z/31R/SzWd1Hstr+6xdvZK/PfRnFv/uJtpaW9nnrYewx6y3sG7NalateI6XXniOVS88x7N/f4yNLc2v2nfMFlvypne+l+P3G8+MaTu+5jx78oub7+K/5i9gSARzDnkTb9nn9UzaahyjR41gSAQRwZAhwZAIGGAPEtl3h/IfPb78qef46c33cM1t9/Lciy+/atvw4cPZcccd2W677dhuu+2YNGkSY8eOZYsttmCLLbZg9OjRDBs2jMbGRoYOHcrQoUNpbGxk2LBhDMme7pj/l4XIrnVPbZ3/FcIHu0hSdQ0bNoy99tqrzz+3p/BcSMlme6BjCa0JOKC7Piml1ohYDUzI2u/qtO/22evejtkv7LfffrS2ttZ6GKqB+++4mfvvuLnLbVttvS077LIb0/fcl8N33MTrd5lMY0NDSZ8zZ/YBrFi5mouuuZlrf/snrv3tn8oZdl1paWlh2bJlLFu2rNZDkSRVQWNjI5s2bar1MF6lkPDcVWmlc1mvuz7dtXc117rLEnhEnASclL1dGxFLuxlnNU0EXrviS5UwIK/tyuefZeXzz/LnP93OgloPpnsD8toOEF7b6vHaVo/Xtnq8tlXS2to6MSJqcW136m5DIeG5Ceg46XMy8Ew3fZqyaRvjgJW97NvbMQFIKc0H5hcwzqqJiMXdle5VHq9t9Xhtq8drWz1e2+rx2laP17Z6+uO1LeRuG/cA0yNiakQMA+YBCzv1WQickL2eC9yachNKFwLzsrtxTAWmA3cXeExJkiSpX+m18pzNYT4FuAloAC5KKS2JiLOAxSmlhcCFwGURsZxcxXletu+SiPgp8BDQCpycUmoD6OqYlT89SZIkqXIKusdTSulG4MZObWd0eN0MHNPNvmcDZxdyzH6sptNGBjmvbfV4bavHa1s9Xtvq8dpWj9e2evrdte31VnWSJEmScgp6wqAkSZIkw/NrRMSZEfF0RDyQ/byrwzYfNV5BXrfyRMQTEfGX7Hu6OGvbKiJ+GxHLsj/HZ+0REd/LrvWDEbFvbUff/0TERRHxfET8tUNb0dczIk7I+i+LiBO6+qx608219XdtmSJih4i4LSIejoglEXFq1u73tkw9XFu/t2WKiBERcXdE/Dm7tl/L2qdGxKLsO3h1dkMJInfTiauz67coIqZ0OFaX17zq8k+C82fzE/HOBE7ron0G8GdgODAV+Bu5xY4N2eudgWFZnxm1Po/+/uN1q8g1fAKY2Kntm8Dp2evTgXOy1+8CfkXu3utvAhbVevz97Qc4CNgX+Gup1xPYCngs+3N89np8rc+t1j/dXFt/15Z/XbcF9s1ejwUeza6f39vqXVu/t+Vf2wDGZK+HAouy7+NPgXlZ+w+AT2evPwP8IHs9D7i6p2veF+dg5blwc4CrUkotKaXHgeXA/tnP8pTSYymljcBVWV/1zOtWHXOAS7PXlwJHdWj/Scq5C9gyIratxQD7q5TS78ndLaijYq/nYcBvU0orU0qrgN8Ch1d/9P1bN9e2O/6uLVBK6dmU0n3Z6zXAw+Se4uv3tkw9XNvu+L0tUPb9W5u9HZr9JOBg2Pzcsc7f2/z3eQEwOyKC7q951Rmeu3ZK9k9aF+X/uYuuH1O+fQ/t6pnXrXwJ+E1E3Bu5J3ECbJNSehZyv/yBrbN2r3dpir2eXufi+Lu2QrJ/yt6HXBXP720Fdbq24Pe2bBHREBEPAM+T+z9rfwNeSim1Zl06XqfN1zDbvhqYQA2vbV2G54i4OSL+2sXPHOD7wC7ATOBZ4L/zu3VxqJ4eQa6eed3Kd2BKaV/gCODkiDioh75e78ry90H5/F1bIRExBrgG+FxK6eWeunbR5rXtQRfX1u9tBaSU2lJKM8k9YXp/YPeuumV/9rtrW9B9ngeblNIhhfSLiB8B12dvy37UuF6lkMe+qwcppWeyP5+PiGvJ/QJ6LiK2TSk9m/1z7PNZd693aYq9nk3A2zu1394H4xxwUkrP5V/7u7Z0ETGUXLi7PKX086zZ720FdHVt/d5WVkrppYi4ndyc5y0jojGrLne8Tvlr2xQRjcA4ctPAavb3Wl1WnnvSaR7o0UB+dbiPGq8sr1sZImJ0RIzNvwYOJfddXQjkV8qfAPwye70Q+HC22v5NwOr8P+uqR8Vez5uAQyNifPbPuYdmberE37Xly+Z9Xgg8nFL6TodNfm/L1N219XtbvoiYFBFbZq9HAoeQm1N+GzA369b5e5v/Ps8Fbk25FYPdXfPq64tViQPpB7gM+AvwYPY/zLYdtn2Z3LycpcARHdrfRW4l7t+AL9f6HAbKj9etrGu3M7lVxn8GluSvH7l5YLcAy7I/t8raAzg/u9Z/AWbV+hz62w9wJbl/ht1ErqJxYinXE/gYuYUry4GP1vq8+sNPN9fW37XlX9e3kvtn6geBB7Kfd/m9req19Xtb/rXdC7g/u4Z/Bc7I2ncmF36XAz8DhmftI7L3y7PtO/d2zav94xMGJUmSpAI5bUOSJEkqkOFZkiRJKpDhWZIkSSqQ4VmSJEkqkOFZkiRJKpDhWZIkSSqQ4VmSVLCI+EhErIiIByLikYj4fNb+5aztgYho6/D6n2s9ZkmqJO/zLEkqWER8hNzDNU6JiAnkHk6wT0rpqQ591qaUxtRqjJJUTVaeJakPRMSUrFL744j4a0RcHhGHRMSdEbEsIvbPHrt+UUTcExH3R8ScDvv+ISLuy37ekrW/PSJuj4gF2bEvzx4r3N0Y3pX1uyMivhcR12ftW0XELyLiwYi4KyL2KuScUkovknvq17a99ZWkwaKx1gOQpDoyDTgGOAm4Bzie3GOAjwS+BDwE3JpS+lhEbAncHRE3A88D70wpNUfEdHKPu56VHXMfYA/gGeBO4EDgjs4fHBEjgB8CB6WUHo+IKzts/hpwf0rpqIg4GPgJMLO3k4mIHck9OvfB4i6DJA1cVp4lqe88nlL6S0qpHVgC3JJyc+f+AkwBDgVOj4gHgNvJBdMdgaHAjyLiL8DPgBkdjnl3SqkpO+YD2XG68nrgsZTS49n7juH5rcBlACmlW4EJETGuh/M4NiKWAI8B300pNRdy8pI0GFh5lqS+09LhdXuH9+3kfh+3Ae9PKS3tuFNEnAk8B+xNrujRMax2PGYb3f9e73Y6RzfbeloQc3U25/nNwA0R8auU0j966C9Jg4aVZ0nqP24CPpuftxwR+2Tt44Bns+ryh4CGEo79CLBzREzJ3h/bYdvvgQ9kn/l24IWU0su9HTCl9CdyFetTSxiPJA1IhmdJ6j++Tm6KxoMR8dfsPcAFwAkRcRewK7Cu2AOnlDYAnwF+HRF3kKtkr842nwnMiogHgf8CTiji0OcAH42IscWOSZIGIm9VJ0l1IiLGpJTWZpXt84FlKaVzaz0uSRpIrDxLUv34RLYYcQm5qSA/rPF4JGnAsfIsSYNMRFwLTO3U/O8ppZuKOMZHee1c5jtTSieXOz5JGsgMz5IkSVKBnLYhSZIkFcjwLEmSJBXI8CxJkiQVyPAsSZIkFcjwLEmSJBXo/wMoaUw0r7FEXAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(12,6))\n",
    "ax = sns.distplot(ssrt_metrics['SSRT'], kde=False, norm_hist=True, label='SSRT')\n",
    "xx = np.arange(-500, 3000, 1)\n",
    "ax.plot(xx, norm.pdf(xx, loc=ssrt_loc, scale=25), 'k', lw=2)  \n",
    "_ = sns.distplot(ssrt_metrics['mean_go_RT'], kde=False,norm_hist=True, ax=ax, label='goRT')\n",
    "ax.plot(xx, norm.pdf(xx, loc=goRT_loc, scale=goRT_var), 'k', lw=2) \n",
    "_ = plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['SSRT', 'mean_SSD', 'p_respond', 'max_RT', 'mean_go_RT', 'sd_go_RT',\n",
       "       'mean_stopfail_RT', 'sd_stopfail_RT', 'omission_count', 'omission_rate',\n",
       "       'go_acc', 'stopfail_acc'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ssrt_metrics.keys()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. get SSD distributions per subject"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "MAX_SSD = 500 #for clipping distributions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data = pd.read_csv('abcd_data/minimal_abcd_clean.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "SSD_dist = abcd_data.groupby('NARGUID')['SSDDur'].value_counts(normalize=True)\n",
    "SSD_dist.name = 'proportion'\n",
    "SSD_dist = SSD_dist.reset_index()\n",
    "SSD_dist.to_csv(f'abcd_data/SSD_dist_by_subj_{MAX_SSD}Clip-False.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "SSD_dist_clipped = abcd_data.query(f\"(SSDDur <= {MAX_SSD}) | (trial_type=='GoTrial')\").groupby('NARGUID')['SSDDur'].value_counts(normalize=True)\n",
    "SSD_dist_clipped.name = 'proportion'\n",
    "SSD_dist_clipped = SSD_dist_clipped.reset_index()\n",
    "SSD_dist_clipped.to_csv(f'abcd_data/SSD_dist_by_subj_{MAX_SSD}Clip-True.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1892"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(SSD_dist.query(f\"SSDDur > {MAX_SSD}\").NARGUID.unique())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1.5 Sample Mus if necessary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "# MAGIC NUMBERS\n",
    "GORT_SCALE = goRT_var\n",
    "SSRT_SCALES = [85, 25, 5, 0]\n",
    "\n",
    "def convert_to_mu(rt, thresh=100, ndt=50):\n",
    "    # example solve:\n",
    "    # Threshold = mu * (rt - non-decision-time)\n",
    "    # Threshold / (rt - ndt) = mu\n",
    "    return thresh/(rt-ndt)\n",
    "\n",
    "def sample(mean, scale, ndt=50):\n",
    "    tmp = np.random.normal(loc=mean, scale=scale)\n",
    "    # if it is less that the non-decision-time, you get negative mus\n",
    "    # if it is close to the nondecision time, you get HUGE mus\n",
    "    while tmp < ndt+10: \n",
    "        tmp = np.random.normal(loc=mean, scale=scale)\n",
    "    return tmp\n",
    "\n",
    "def get_mu_suffix(SSRT_SCALE=ssrt_var, GORT_SCALE=goRT_var):\n",
    "    stop_str_suffix = 'SSRTscale-%d' % SSRT_SCALE\n",
    "    go_str_suffix = ''\n",
    "    if GORT_SCALE != goRT_var:\n",
    "        go_str_suffix='_RTscale-%d' % GORT_SCALE\n",
    "    return '%s%s' % (stop_str_suffix, go_str_suffix)\n",
    "\n",
    "def assign_mus(SSRT_SCALE=ssrt_var, GORT_SCALE=goRT_var):\n",
    "    \n",
    "    suffix = get_mu_suffix(SSRT_SCALE, GORT_SCALE)\n",
    "    \n",
    "    write_out_path = 'abcd_data/assigned_mus_%s.json' % suffix\n",
    "    \n",
    "    if not path.exists(write_out_path) or OVERWRITE_MUS:\n",
    "        print('assigning mus: %s' % suffix)\n",
    "        mus_dict = {}\n",
    "        for sub in SSD_dist.NARGUID.unique():\n",
    "            mus_dict[sub] = {}\n",
    "            mus_dict[sub]['goRT'] = sample(mean=goRT_loc, scale=GORT_SCALE)\n",
    "            mus_dict[sub]['go'] = convert_to_mu(mus_dict[sub]['goRT'])\n",
    "            mus_dict[sub]['SSRT'] = sample(mean=ssrt_loc, scale=SSRT_SCALE)\n",
    "            mus_dict[sub]['stop'] = convert_to_mu(mus_dict[sub]['SSRT']) \n",
    "        with open(write_out_path, 'w') as jp:\n",
    "            json.dump(mus_dict, jp)\n",
    "    return write_out_path\n",
    "\n",
    "def visualize_mu_dists(write_out_path):\n",
    "    with open(write_out_path, 'r') as jf:\n",
    "        mus_dict = json.load(jf)\n",
    "    \n",
    "    suffix_str = write_out_path.split('assigned_mus_')[-1].replace('.json', '')\n",
    "    mu_df = pd.DataFrame(mus_dict).T\n",
    "    fig = plt.figure(figsize=(8,4))\n",
    "    ax = sns.distplot(mu_df['go'], kde=False, norm_hist=True, label='go')\n",
    "    _ = sns.distplot(mu_df['stop'], kde=False,norm_hist=True, ax=ax, label='stop')\n",
    "    _ = plt.legend()\n",
    "    _ = plt.title(suffix_str)\n",
    "    plt.show()\n",
    "    plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "assigning mus: SSRTscale-5\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# MAIN    \n",
    "for SSRT_SCALE in SSRT_SCALES: \n",
    "    mu_path = assign_mus(SSRT_SCALE=SSRT_SCALE)\n",
    "    visualize_mu_dists(mu_path)\n",
    "    \n",
    "# Single Individual Case\n",
    "mu_path_no_var = assign_mus(SSRT_SCALE=0,GORT_SCALE=0)\n",
    "visualize_mu_dists(mu_path_no_var)\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exgauss metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data = pd.read_csv('abcd_data/minimal_abcd_clean.csv')\n",
    "SSD0_RTs = abcd_data.query(\n",
    "    \"SSDDur == 0.0 and correct_stop==0.0\"\n",
    "    ).stop_rt_adjusted.values\n",
    "\n",
    "FIT_K, FIT_LOC, FIT_SCALE = sstats.exponnorm.fit(SSD0_RTs)\n",
    "FIT_LAMBDA = 1/(FIT_K*FIT_SCALE)\n",
    "FIT_BETA = 1/FIT_LAMBDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "exgaus_params = {\n",
    "    'K': FIT_K,\n",
    "    'loc': FIT_LOC,\n",
    "    'scale': FIT_SCALE,\n",
    "    'lambda': FIT_LAMBDA,\n",
    "    'beta': FIT_BETA,\n",
    "}\n",
    "\n",
    "with open('abcd_data/exgauss_params.json', 'w') as f:\n",
    "    json.dump(exgaus_params, f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. P(guess|SSD) for mixture distributions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data = pd.read_csv('abcd_data/minimal_abcd_clean.csv')\n",
    "SSDs = abcd_data.SSDDur.unique()\n",
    "SSDs = [i for i in SSDs if i == i]\n",
    "SSDs.sort()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "acc_per_SSD = pd.DataFrame()\n",
    "for ssd in SSDs:\n",
    "    curr_means = abcd_data.query(\n",
    "        \"SSDDur == %s and correct_stop==0.0\" % ssd\n",
    "    ).groupby('NARGUID').mean()['choice_accuracy']\n",
    "    curr_means.name = ssd\n",
    "    acc_per_SSD = pd.concat([acc_per_SSD, curr_means], 1, sort=True)\n",
    "\n",
    "go_accs = abcd_data.query(\n",
    "        \"trial_type == 'GoTrial' and correct_go_response in [1.0, 0.0]\"\n",
    "    ).groupby('NARGUID').mean()['choice_accuracy']\n",
    "go_accs.name = -1\n",
    "acc_per_SSD = pd.concat([acc_per_SSD, go_accs], 1, sort=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "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>0.0</th>\n",
       "      <th>50.0</th>\n",
       "      <th>100.0</th>\n",
       "      <th>150.0</th>\n",
       "      <th>200.0</th>\n",
       "      <th>250.0</th>\n",
       "      <th>300.0</th>\n",
       "      <th>350.0</th>\n",
       "      <th>400.0</th>\n",
       "      <th>450.0</th>\n",
       "      <th>500.0</th>\n",
       "      <th>550.0</th>\n",
       "      <th>600.0</th>\n",
       "      <th>650.0</th>\n",
       "      <th>700.0</th>\n",
       "      <th>750.0</th>\n",
       "      <th>800.0</th>\n",
       "      <th>850.0</th>\n",
       "      <th>900.0</th>\n",
       "      <th>-1.0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2944.000000</td>\n",
       "      <td>4146.000000</td>\n",
       "      <td>4819.000000</td>\n",
       "      <td>5759.000000</td>\n",
       "      <td>6554.000000</td>\n",
       "      <td>6817.000000</td>\n",
       "      <td>6566.000000</td>\n",
       "      <td>5842.000000</td>\n",
       "      <td>4783.000000</td>\n",
       "      <td>3651.000000</td>\n",
       "      <td>2516.000000</td>\n",
       "      <td>1661.000000</td>\n",
       "      <td>1057.000000</td>\n",
       "      <td>630.000000</td>\n",
       "      <td>392.000000</td>\n",
       "      <td>239.000000</td>\n",
       "      <td>138.000000</td>\n",
       "      <td>117.000000</td>\n",
       "      <td>115.000000</td>\n",
       "      <td>8210.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.499410</td>\n",
       "      <td>0.561566</td>\n",
       "      <td>0.656870</td>\n",
       "      <td>0.723856</td>\n",
       "      <td>0.795379</td>\n",
       "      <td>0.830639</td>\n",
       "      <td>0.860157</td>\n",
       "      <td>0.884024</td>\n",
       "      <td>0.895152</td>\n",
       "      <td>0.906854</td>\n",
       "      <td>0.912347</td>\n",
       "      <td>0.919252</td>\n",
       "      <td>0.917789</td>\n",
       "      <td>0.914306</td>\n",
       "      <td>0.894193</td>\n",
       "      <td>0.904812</td>\n",
       "      <td>0.852174</td>\n",
       "      <td>0.831197</td>\n",
       "      <td>0.817350</td>\n",
       "      <td>0.919088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.309872</td>\n",
       "      <td>0.346036</td>\n",
       "      <td>0.328498</td>\n",
       "      <td>0.303200</td>\n",
       "      <td>0.263413</td>\n",
       "      <td>0.243262</td>\n",
       "      <td>0.230942</td>\n",
       "      <td>0.219983</td>\n",
       "      <td>0.223044</td>\n",
       "      <td>0.218917</td>\n",
       "      <td>0.224541</td>\n",
       "      <td>0.229288</td>\n",
       "      <td>0.229932</td>\n",
       "      <td>0.240663</td>\n",
       "      <td>0.276771</td>\n",
       "      <td>0.263033</td>\n",
       "      <td>0.319575</td>\n",
       "      <td>0.327071</td>\n",
       "      <td>0.333918</td>\n",
       "      <td>0.085973</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.833333</td>\n",
       "      <td>0.875000</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",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.833333</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.898613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.571429</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.875000</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",
       "      <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",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.941708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.880515</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",
       "      <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",
       "      <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>0.969900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\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",
       "      <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",
       "      <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",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             0.0          50.0         100.0        150.0        200.0  \\\n",
       "count  2944.000000  4146.000000  4819.000000  5759.000000  6554.000000   \n",
       "mean      0.499410     0.561566     0.656870     0.723856     0.795379   \n",
       "std       0.309872     0.346036     0.328498     0.303200     0.263413   \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "25%       0.333333     0.333333     0.500000     0.500000     0.666667   \n",
       "50%       0.500000     0.571429     0.666667     0.800000     0.875000   \n",
       "75%       0.666667     0.880515     1.000000     1.000000     1.000000   \n",
       "max       1.000000     1.000000     1.000000     1.000000     1.000000   \n",
       "\n",
       "             250.0        300.0        350.0        400.0        450.0  \\\n",
       "count  6817.000000  6566.000000  5842.000000  4783.000000  3651.000000   \n",
       "mean      0.830639     0.860157     0.884024     0.895152     0.906854   \n",
       "std       0.243262     0.230942     0.219983     0.223044     0.218917   \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "25%       0.750000     0.800000     0.833333     0.875000     1.000000   \n",
       "50%       1.000000     1.000000     1.000000     1.000000     1.000000   \n",
       "75%       1.000000     1.000000     1.000000     1.000000     1.000000   \n",
       "max       1.000000     1.000000     1.000000     1.000000     1.000000   \n",
       "\n",
       "             500.0        550.0        600.0       650.0       700.0  \\\n",
       "count  2516.000000  1661.000000  1057.000000  630.000000  392.000000   \n",
       "mean      0.912347     0.919252     0.917789    0.914306    0.894193   \n",
       "std       0.224541     0.229288     0.229932    0.240663    0.276771   \n",
       "min       0.000000     0.000000     0.000000    0.000000    0.000000   \n",
       "25%       1.000000     1.000000     1.000000    1.000000    1.000000   \n",
       "50%       1.000000     1.000000     1.000000    1.000000    1.000000   \n",
       "75%       1.000000     1.000000     1.000000    1.000000    1.000000   \n",
       "max       1.000000     1.000000     1.000000    1.000000    1.000000   \n",
       "\n",
       "            750.0       800.0       850.0       900.0       -1.0    \n",
       "count  239.000000  138.000000  117.000000  115.000000  8210.000000  \n",
       "mean     0.904812    0.852174    0.831197    0.817350     0.919088  \n",
       "std      0.263033    0.319575    0.327071    0.333918     0.085973  \n",
       "min      0.000000    0.000000    0.000000    0.000000     0.000000  \n",
       "25%      1.000000    1.000000    0.833333    0.666667     0.898613  \n",
       "50%      1.000000    1.000000    1.000000    1.000000     0.941708  \n",
       "75%      1.000000    1.000000    1.000000    1.000000     0.969900  \n",
       "max      1.000000    1.000000    1.000000    1.000000     1.000000  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "acc_per_SSD.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 0.0      0.499410\n",
       " 50.0     0.561566\n",
       " 100.0    0.656870\n",
       " 150.0    0.723856\n",
       " 200.0    0.795379\n",
       " 250.0    0.830639\n",
       " 300.0    0.860157\n",
       " 350.0    0.884024\n",
       " 400.0    0.895152\n",
       " 450.0    0.906854\n",
       " 500.0    0.912347\n",
       " 550.0    0.919252\n",
       " 600.0    0.917789\n",
       " 650.0    0.914306\n",
       " 700.0    0.894193\n",
       " 750.0    0.904812\n",
       " 800.0    0.852174\n",
       " 850.0    0.831197\n",
       " 900.0    0.817350\n",
       "-1.0      0.919088\n",
       "dtype: float64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "acc_per_SSD.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "p = Symbol('p')\n",
    "guess_mean = acc_per_SSD.mean()[0.0]\n",
    "go_mean = acc_per_SSD.mean()[-1]\n",
    "p_guess_per_SSD = {}\n",
    "for ssd in SSDs:\n",
    "    curr_mean = acc_per_SSD.mean()[ssd]\n",
    "    solution = solve(p*guess_mean + (1-p)*go_mean - curr_mean, p)\n",
    "    assert len(solution) == 1\n",
    "    p_guess_per_SSD[ssd] = solution[0]\n",
    "p_guess_df = pd.DataFrame(p_guess_per_SSD, index=['p_guess'])\n",
    "p_guess_df.to_csv('abcd_data/p_guess_per_ssd.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0.0: 1.0,\n",
       " 50.0: 0.8518966881510508,\n",
       " 100.0: 0.6248081425758761,\n",
       " 150.0: 0.46519470731379453,\n",
       " 200.0: 0.2947710624963631,\n",
       " 250.0: 0.21075535798552303,\n",
       " 300.0: 0.1404209047269027,\n",
       " 350.0: 0.08354920017624702,\n",
       " 400.0: 0.05703409119007931,\n",
       " 450.0: 0.029151024358994575,\n",
       " 500.0: 0.016063618909064378,\n",
       " 550.0: -0.0003910068585019575,\n",
       " 600.0: 0.0030962512655911116,\n",
       " 650.0: 0.01139533818314855,\n",
       " 700.0: 0.059318925632538,\n",
       " 750.0: 0.0340178091093458,\n",
       " 800.0: 0.15944204679687593,\n",
       " 850.0: 0.20942638758940402,\n",
       " 900.0: 0.24241997384504005}"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "{col: float(p_guess_df[col].values[0]) for col\n",
    "                      in p_guess_df.columns}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "p_guess_df2 = pd.read_csv('abcd_data/p_guess_per_ssd.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "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>0.0</th>\n",
       "      <th>50.0</th>\n",
       "      <th>100.0</th>\n",
       "      <th>150.0</th>\n",
       "      <th>200.0</th>\n",
       "      <th>250.0</th>\n",
       "      <th>300.0</th>\n",
       "      <th>350.0</th>\n",
       "      <th>400.0</th>\n",
       "      <th>450.0</th>\n",
       "      <th>500.0</th>\n",
       "      <th>550.0</th>\n",
       "      <th>600.0</th>\n",
       "      <th>650.0</th>\n",
       "      <th>700.0</th>\n",
       "      <th>750.0</th>\n",
       "      <th>800.0</th>\n",
       "      <th>850.0</th>\n",
       "      <th>900.0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.851897</td>\n",
       "      <td>0.624808</td>\n",
       "      <td>0.465195</td>\n",
       "      <td>0.294771</td>\n",
       "      <td>0.210755</td>\n",
       "      <td>0.140421</td>\n",
       "      <td>0.083549</td>\n",
       "      <td>0.057034</td>\n",
       "      <td>0.029151</td>\n",
       "      <td>0.016064</td>\n",
       "      <td>-0.000391</td>\n",
       "      <td>0.003096</td>\n",
       "      <td>0.011395</td>\n",
       "      <td>0.059319</td>\n",
       "      <td>0.034018</td>\n",
       "      <td>0.159442</td>\n",
       "      <td>0.209426</td>\n",
       "      <td>0.24242</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   0.0      50.0     100.0     150.0     200.0     250.0     300.0     350.0  \\\n",
       "0  1.0  0.851897  0.624808  0.465195  0.294771  0.210755  0.140421  0.083549   \n",
       "\n",
       "      400.0     450.0     500.0     550.0     600.0     650.0     700.0  \\\n",
       "0  0.057034  0.029151  0.016064 -0.000391  0.003096  0.011395  0.059319   \n",
       "\n",
       "      750.0     800.0     850.0    900.0  \n",
       "0  0.034018  0.159442  0.209426  0.24242  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p_guess_df2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Within subjects SD of RTs for stop-fail and go RTs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "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>go_rt_adjusted</th>\n",
       "      <th>stop_rt_adjusted</th>\n",
       "      <th>SSDDur</th>\n",
       "      <th>correct_go_response</th>\n",
       "      <th>correct_stop</th>\n",
       "      <th>SSD.RESP</th>\n",
       "      <th>Fix.RESP</th>\n",
       "      <th>StopSignal.RESP</th>\n",
       "      <th>Go.RESP</th>\n",
       "      <th>TrialNum</th>\n",
       "      <th>correct_stimulus_mapping_1</th>\n",
       "      <th>correct_stimulus_mapping_2</th>\n",
       "      <th>block</th>\n",
       "      <th>finger_press</th>\n",
       "      <th>correct_response</th>\n",
       "      <th>choice_accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2.292144e+06</td>\n",
       "      <td>238026.000000</td>\n",
       "      <td>494283.000000</td>\n",
       "      <td>2.471397e+06</td>\n",
       "      <td>494283.000000</td>\n",
       "      <td>19624.000000</td>\n",
       "      <td>236334.000000</td>\n",
       "      <td>171150.000000</td>\n",
       "      <td>2.203096e+06</td>\n",
       "      <td>2.965680e+06</td>\n",
       "      <td>2965680.0</td>\n",
       "      <td>2965680.0</td>\n",
       "      <td>2.965680e+06</td>\n",
       "      <td>2.530170e+06</td>\n",
       "      <td>2965680.0</td>\n",
       "      <td>2.530170e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.437762e+02</td>\n",
       "      <td>458.165058</td>\n",
       "      <td>236.154895</td>\n",
       "      <td>8.411251e-01</td>\n",
       "      <td>0.518442</td>\n",
       "      <td>1.591215</td>\n",
       "      <td>1.638694</td>\n",
       "      <td>1.602530</td>\n",
       "      <td>1.603930e+00</td>\n",
       "      <td>1.805000e+02</td>\n",
       "      <td>1.5</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.502778e+00</td>\n",
       "      <td>1.493806e+00</td>\n",
       "      <td>1.5</td>\n",
       "      <td>9.096041e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.135862e+02</td>\n",
       "      <td>207.786635</td>\n",
       "      <td>159.296667</td>\n",
       "      <td>3.655595e-01</td>\n",
       "      <td>0.499660</td>\n",
       "      <td>0.662964</td>\n",
       "      <td>0.685852</td>\n",
       "      <td>0.682406</td>\n",
       "      <td>6.827720e-01</td>\n",
       "      <td>1.039227e+02</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>4.999924e-01</td>\n",
       "      <td>4.999617e-01</td>\n",
       "      <td>0.5</td>\n",
       "      <td>2.867482e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>3.990000e+02</td>\n",
       "      <td>336.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>9.075000e+01</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.060000e+02</td>\n",
       "      <td>426.000000</td>\n",
       "      <td>200.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>1.805000e+02</td>\n",
       "      <td>1.5</td>\n",
       "      <td>3.5</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.5</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.460000e+02</td>\n",
       "      <td>530.000000</td>\n",
       "      <td>350.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>2.702500e+02</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2.793000e+03</td>\n",
       "      <td>2835.000000</td>\n",
       "      <td>900.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000e+00</td>\n",
       "      <td>3.600000e+02</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       go_rt_adjusted  stop_rt_adjusted         SSDDur  correct_go_response  \\\n",
       "count    2.292144e+06     238026.000000  494283.000000         2.471397e+06   \n",
       "mean     5.437762e+02        458.165058     236.154895         8.411251e-01   \n",
       "std      2.135862e+02        207.786635     159.296667         3.655595e-01   \n",
       "min      0.000000e+00          0.000000       0.000000         0.000000e+00   \n",
       "25%      3.990000e+02        336.000000     100.000000         1.000000e+00   \n",
       "50%      5.060000e+02        426.000000     200.000000         1.000000e+00   \n",
       "75%      6.460000e+02        530.000000     350.000000         1.000000e+00   \n",
       "max      2.793000e+03       2835.000000     900.000000         1.000000e+00   \n",
       "\n",
       "        correct_stop      SSD.RESP       Fix.RESP  StopSignal.RESP  \\\n",
       "count  494283.000000  19624.000000  236334.000000    171150.000000   \n",
       "mean        0.518442      1.591215       1.638694         1.602530   \n",
       "std         0.499660      0.662964       0.685852         0.682406   \n",
       "min         0.000000      1.000000       1.000000         1.000000   \n",
       "25%         0.000000      1.000000       1.000000         1.000000   \n",
       "50%         1.000000      2.000000       2.000000         2.000000   \n",
       "75%         1.000000      2.000000       2.000000         2.000000   \n",
       "max         1.000000      4.000000       4.000000         4.000000   \n",
       "\n",
       "            Go.RESP      TrialNum  correct_stimulus_mapping_1  \\\n",
       "count  2.203096e+06  2.965680e+06                   2965680.0   \n",
       "mean   1.603930e+00  1.805000e+02                         1.5   \n",
       "std    6.827720e-01  1.039227e+02                         0.5   \n",
       "min    1.000000e+00  1.000000e+00                         1.0   \n",
       "25%    1.000000e+00  9.075000e+01                         1.0   \n",
       "50%    2.000000e+00  1.805000e+02                         1.5   \n",
       "75%    2.000000e+00  2.702500e+02                         2.0   \n",
       "max    4.000000e+00  3.600000e+02                         2.0   \n",
       "\n",
       "       correct_stimulus_mapping_2         block  finger_press  \\\n",
       "count                   2965680.0  2.965680e+06  2.530170e+06   \n",
       "mean                          3.5  1.502778e+00  1.493806e+00   \n",
       "std                           0.5  4.999924e-01  4.999617e-01   \n",
       "min                           3.0  1.000000e+00  1.000000e+00   \n",
       "25%                           3.0  1.000000e+00  1.000000e+00   \n",
       "50%                           3.5  2.000000e+00  1.000000e+00   \n",
       "75%                           4.0  2.000000e+00  2.000000e+00   \n",
       "max                           4.0  2.000000e+00  2.000000e+00   \n",
       "\n",
       "       correct_response  choice_accuracy  \n",
       "count         2965680.0     2.530170e+06  \n",
       "mean                1.5     9.096041e-01  \n",
       "std                 0.5     2.867482e-01  \n",
       "min                 1.0     0.000000e+00  \n",
       "25%                 1.0     1.000000e+00  \n",
       "50%                 1.5     1.000000e+00  \n",
       "75%                 2.0     1.000000e+00  \n",
       "max                 2.0     1.000000e+00  "
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "abcd_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data = pd.read_csv('abcd_data/minimal_abcd_clean.csv')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "std\n",
       "go_rt_adjusted      190.272261\n",
       "stop_rt_adjusted    155.122888\n",
       "dtype: float64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "abcd_data[['NARGUID', 'go_rt_adjusted', 'stop_rt_adjusted']].groupby('NARGUID').apply(lambda x: x.describe().loc['std', :]).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"8\" halign=\"left\">go_rt_adjusted</th>\n",
       "      <th colspan=\"8\" halign=\"left\">stop_rt_adjusted</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NARGUID</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",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>003RTV85</th>\n",
       "      <td>285.0</td>\n",
       "      <td>634.652632</td>\n",
       "      <td>169.271353</td>\n",
       "      <td>333.0</td>\n",
       "      <td>515.00</td>\n",
       "      <td>595.0</td>\n",
       "      <td>728.00</td>\n",
       "      <td>1336.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>546.107143</td>\n",
       "      <td>96.794681</td>\n",
       "      <td>395.0</td>\n",
       "      <td>496.25</td>\n",
       "      <td>525.0</td>\n",
       "      <td>564.25</td>\n",
       "      <td>877.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>007W6H7B</th>\n",
       "      <td>297.0</td>\n",
       "      <td>653.491582</td>\n",
       "      <td>178.164613</td>\n",
       "      <td>318.0</td>\n",
       "      <td>515.00</td>\n",
       "      <td>609.0</td>\n",
       "      <td>764.00</td>\n",
       "      <td>1242.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>562.000000</td>\n",
       "      <td>121.130673</td>\n",
       "      <td>405.0</td>\n",
       "      <td>474.75</td>\n",
       "      <td>541.5</td>\n",
       "      <td>645.25</td>\n",
       "      <td>859.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00CY2MDM</th>\n",
       "      <td>284.0</td>\n",
       "      <td>467.901408</td>\n",
       "      <td>200.308095</td>\n",
       "      <td>4.0</td>\n",
       "      <td>331.00</td>\n",
       "      <td>414.0</td>\n",
       "      <td>566.25</td>\n",
       "      <td>1332.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>347.583333</td>\n",
       "      <td>92.117743</td>\n",
       "      <td>116.0</td>\n",
       "      <td>272.50</td>\n",
       "      <td>349.0</td>\n",
       "      <td>405.50</td>\n",
       "      <td>530.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00HEV6HB</th>\n",
       "      <td>289.0</td>\n",
       "      <td>450.200692</td>\n",
       "      <td>228.866837</td>\n",
       "      <td>6.0</td>\n",
       "      <td>313.00</td>\n",
       "      <td>440.0</td>\n",
       "      <td>527.00</td>\n",
       "      <td>1688.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>373.947368</td>\n",
       "      <td>289.887298</td>\n",
       "      <td>83.0</td>\n",
       "      <td>256.00</td>\n",
       "      <td>343.5</td>\n",
       "      <td>403.00</td>\n",
       "      <td>1884.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00J52GPG</th>\n",
       "      <td>281.0</td>\n",
       "      <td>405.978648</td>\n",
       "      <td>157.648749</td>\n",
       "      <td>104.0</td>\n",
       "      <td>284.00</td>\n",
       "      <td>350.0</td>\n",
       "      <td>494.00</td>\n",
       "      <td>1165.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>305.194444</td>\n",
       "      <td>67.751993</td>\n",
       "      <td>245.0</td>\n",
       "      <td>265.00</td>\n",
       "      <td>285.5</td>\n",
       "      <td>319.25</td>\n",
       "      <td>543.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vcahyykd</th>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ve4u5</th>\n",
       "      <td>283.0</td>\n",
       "      <td>451.784452</td>\n",
       "      <td>132.458823</td>\n",
       "      <td>196.0</td>\n",
       "      <td>367.00</td>\n",
       "      <td>432.0</td>\n",
       "      <td>515.50</td>\n",
       "      <td>1053.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>387.851852</td>\n",
       "      <td>101.128518</td>\n",
       "      <td>264.0</td>\n",
       "      <td>303.00</td>\n",
       "      <td>370.0</td>\n",
       "      <td>443.50</td>\n",
       "      <td>595.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wzrf2ge6</th>\n",
       "      <td>275.0</td>\n",
       "      <td>591.981818</td>\n",
       "      <td>186.033696</td>\n",
       "      <td>311.0</td>\n",
       "      <td>469.50</td>\n",
       "      <td>547.0</td>\n",
       "      <td>658.00</td>\n",
       "      <td>1520.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>502.111111</td>\n",
       "      <td>102.445232</td>\n",
       "      <td>247.0</td>\n",
       "      <td>429.50</td>\n",
       "      <td>481.0</td>\n",
       "      <td>573.00</td>\n",
       "      <td>692.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>x8k59</th>\n",
       "      <td>289.0</td>\n",
       "      <td>488.622837</td>\n",
       "      <td>143.985096</td>\n",
       "      <td>210.0</td>\n",
       "      <td>380.00</td>\n",
       "      <td>478.0</td>\n",
       "      <td>565.00</td>\n",
       "      <td>1081.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>385.892857</td>\n",
       "      <td>87.299119</td>\n",
       "      <td>239.0</td>\n",
       "      <td>321.00</td>\n",
       "      <td>369.5</td>\n",
       "      <td>460.75</td>\n",
       "      <td>561.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>xxmy9wd8</th>\n",
       "      <td>252.0</td>\n",
       "      <td>535.813492</td>\n",
       "      <td>242.239257</td>\n",
       "      <td>39.0</td>\n",
       "      <td>355.25</td>\n",
       "      <td>483.0</td>\n",
       "      <td>660.75</td>\n",
       "      <td>1653.0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>382.218750</td>\n",
       "      <td>122.849694</td>\n",
       "      <td>126.0</td>\n",
       "      <td>314.75</td>\n",
       "      <td>356.5</td>\n",
       "      <td>450.25</td>\n",
       "      <td>776.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8238 rows × 16 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         go_rt_adjusted                                                        \\\n",
       "                  count        mean         std    min     25%    50%     75%   \n",
       "NARGUID                                                                         \n",
       "003RTV85          285.0  634.652632  169.271353  333.0  515.00  595.0  728.00   \n",
       "007W6H7B          297.0  653.491582  178.164613  318.0  515.00  609.0  764.00   \n",
       "00CY2MDM          284.0  467.901408  200.308095    4.0  331.00  414.0  566.25   \n",
       "00HEV6HB          289.0  450.200692  228.866837    6.0  313.00  440.0  527.00   \n",
       "00J52GPG          281.0  405.978648  157.648749  104.0  284.00  350.0  494.00   \n",
       "...                 ...         ...         ...    ...     ...    ...     ...   \n",
       "vcahyykd            0.0         NaN         NaN    NaN     NaN    NaN     NaN   \n",
       "ve4u5             283.0  451.784452  132.458823  196.0  367.00  432.0  515.50   \n",
       "wzrf2ge6          275.0  591.981818  186.033696  311.0  469.50  547.0  658.00   \n",
       "x8k59             289.0  488.622837  143.985096  210.0  380.00  478.0  565.00   \n",
       "xxmy9wd8          252.0  535.813492  242.239257   39.0  355.25  483.0  660.75   \n",
       "\n",
       "                 stop_rt_adjusted                                         \\\n",
       "             max            count        mean         std    min     25%   \n",
       "NARGUID                                                                    \n",
       "003RTV85  1336.0             28.0  546.107143   96.794681  395.0  496.25   \n",
       "007W6H7B  1242.0             26.0  562.000000  121.130673  405.0  474.75   \n",
       "00CY2MDM  1332.0             36.0  347.583333   92.117743  116.0  272.50   \n",
       "00HEV6HB  1688.0             38.0  373.947368  289.887298   83.0  256.00   \n",
       "00J52GPG  1165.0             36.0  305.194444   67.751993  245.0  265.00   \n",
       "...          ...              ...         ...         ...    ...     ...   \n",
       "vcahyykd     NaN              0.0         NaN         NaN    NaN     NaN   \n",
       "ve4u5     1053.0             27.0  387.851852  101.128518  264.0  303.00   \n",
       "wzrf2ge6  1520.0             27.0  502.111111  102.445232  247.0  429.50   \n",
       "x8k59     1081.0             28.0  385.892857   87.299119  239.0  321.00   \n",
       "xxmy9wd8  1653.0             32.0  382.218750  122.849694  126.0  314.75   \n",
       "\n",
       "                                 \n",
       "            50%     75%     max  \n",
       "NARGUID                          \n",
       "003RTV85  525.0  564.25   877.0  \n",
       "007W6H7B  541.5  645.25   859.0  \n",
       "00CY2MDM  349.0  405.50   530.0  \n",
       "00HEV6HB  343.5  403.00  1884.0  \n",
       "00J52GPG  285.5  319.25   543.0  \n",
       "...         ...     ...     ...  \n",
       "vcahyykd    NaN     NaN     NaN  \n",
       "ve4u5     370.0  443.50   595.0  \n",
       "wzrf2ge6  481.0  573.00   692.0  \n",
       "x8k59     369.5  460.75   561.0  \n",
       "xxmy9wd8  356.5  450.25   776.0  \n",
       "\n",
       "[8238 rows x 16 columns]"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rt_desc = abcd_data[['NARGUID', 'go_rt_adjusted', 'stop_rt_adjusted']].groupby('NARGUID').describe()\n",
    "rt_desc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"8\" halign=\"left\">go_rt_adjusted</th>\n",
       "      <th colspan=\"8\" halign=\"left\">stop_rt_adjusted</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NARGUID</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",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>003RTV85</th>\n",
       "      <td>285.0</td>\n",
       "      <td>634.652632</td>\n",
       "      <td>169.271353</td>\n",
       "      <td>333.0</td>\n",
       "      <td>515.00</td>\n",
       "      <td>595.0</td>\n",
       "      <td>728.00</td>\n",
       "      <td>1336.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>546.107143</td>\n",
       "      <td>96.794681</td>\n",
       "      <td>395.0</td>\n",
       "      <td>496.25</td>\n",
       "      <td>525.0</td>\n",
       "      <td>564.25</td>\n",
       "      <td>877.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>007W6H7B</th>\n",
       "      <td>297.0</td>\n",
       "      <td>653.491582</td>\n",
       "      <td>178.164613</td>\n",
       "      <td>318.0</td>\n",
       "      <td>515.00</td>\n",
       "      <td>609.0</td>\n",
       "      <td>764.00</td>\n",
       "      <td>1242.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>562.000000</td>\n",
       "      <td>121.130673</td>\n",
       "      <td>405.0</td>\n",
       "      <td>474.75</td>\n",
       "      <td>541.5</td>\n",
       "      <td>645.25</td>\n",
       "      <td>859.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00CY2MDM</th>\n",
       "      <td>284.0</td>\n",
       "      <td>467.901408</td>\n",
       "      <td>200.308095</td>\n",
       "      <td>4.0</td>\n",
       "      <td>331.00</td>\n",
       "      <td>414.0</td>\n",
       "      <td>566.25</td>\n",
       "      <td>1332.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>347.583333</td>\n",
       "      <td>92.117743</td>\n",
       "      <td>116.0</td>\n",
       "      <td>272.50</td>\n",
       "      <td>349.0</td>\n",
       "      <td>405.50</td>\n",
       "      <td>530.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00HEV6HB</th>\n",
       "      <td>289.0</td>\n",
       "      <td>450.200692</td>\n",
       "      <td>228.866837</td>\n",
       "      <td>6.0</td>\n",
       "      <td>313.00</td>\n",
       "      <td>440.0</td>\n",
       "      <td>527.00</td>\n",
       "      <td>1688.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>373.947368</td>\n",
       "      <td>289.887298</td>\n",
       "      <td>83.0</td>\n",
       "      <td>256.00</td>\n",
       "      <td>343.5</td>\n",
       "      <td>403.00</td>\n",
       "      <td>1884.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00J52GPG</th>\n",
       "      <td>281.0</td>\n",
       "      <td>405.978648</td>\n",
       "      <td>157.648749</td>\n",
       "      <td>104.0</td>\n",
       "      <td>284.00</td>\n",
       "      <td>350.0</td>\n",
       "      <td>494.00</td>\n",
       "      <td>1165.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>305.194444</td>\n",
       "      <td>67.751993</td>\n",
       "      <td>245.0</td>\n",
       "      <td>265.00</td>\n",
       "      <td>285.5</td>\n",
       "      <td>319.25</td>\n",
       "      <td>543.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vcahyykd</th>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ve4u5</th>\n",
       "      <td>283.0</td>\n",
       "      <td>451.784452</td>\n",
       "      <td>132.458823</td>\n",
       "      <td>196.0</td>\n",
       "      <td>367.00</td>\n",
       "      <td>432.0</td>\n",
       "      <td>515.50</td>\n",
       "      <td>1053.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>387.851852</td>\n",
       "      <td>101.128518</td>\n",
       "      <td>264.0</td>\n",
       "      <td>303.00</td>\n",
       "      <td>370.0</td>\n",
       "      <td>443.50</td>\n",
       "      <td>595.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wzrf2ge6</th>\n",
       "      <td>275.0</td>\n",
       "      <td>591.981818</td>\n",
       "      <td>186.033696</td>\n",
       "      <td>311.0</td>\n",
       "      <td>469.50</td>\n",
       "      <td>547.0</td>\n",
       "      <td>658.00</td>\n",
       "      <td>1520.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>502.111111</td>\n",
       "      <td>102.445232</td>\n",
       "      <td>247.0</td>\n",
       "      <td>429.50</td>\n",
       "      <td>481.0</td>\n",
       "      <td>573.00</td>\n",
       "      <td>692.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>x8k59</th>\n",
       "      <td>289.0</td>\n",
       "      <td>488.622837</td>\n",
       "      <td>143.985096</td>\n",
       "      <td>210.0</td>\n",
       "      <td>380.00</td>\n",
       "      <td>478.0</td>\n",
       "      <td>565.00</td>\n",
       "      <td>1081.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>385.892857</td>\n",
       "      <td>87.299119</td>\n",
       "      <td>239.0</td>\n",
       "      <td>321.00</td>\n",
       "      <td>369.5</td>\n",
       "      <td>460.75</td>\n",
       "      <td>561.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>xxmy9wd8</th>\n",
       "      <td>252.0</td>\n",
       "      <td>535.813492</td>\n",
       "      <td>242.239257</td>\n",
       "      <td>39.0</td>\n",
       "      <td>355.25</td>\n",
       "      <td>483.0</td>\n",
       "      <td>660.75</td>\n",
       "      <td>1653.0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>382.218750</td>\n",
       "      <td>122.849694</td>\n",
       "      <td>126.0</td>\n",
       "      <td>314.75</td>\n",
       "      <td>356.5</td>\n",
       "      <td>450.25</td>\n",
       "      <td>776.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8238 rows × 16 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         go_rt_adjusted                                                        \\\n",
       "                  count        mean         std    min     25%    50%     75%   \n",
       "NARGUID                                                                         \n",
       "003RTV85          285.0  634.652632  169.271353  333.0  515.00  595.0  728.00   \n",
       "007W6H7B          297.0  653.491582  178.164613  318.0  515.00  609.0  764.00   \n",
       "00CY2MDM          284.0  467.901408  200.308095    4.0  331.00  414.0  566.25   \n",
       "00HEV6HB          289.0  450.200692  228.866837    6.0  313.00  440.0  527.00   \n",
       "00J52GPG          281.0  405.978648  157.648749  104.0  284.00  350.0  494.00   \n",
       "...                 ...         ...         ...    ...     ...    ...     ...   \n",
       "vcahyykd            0.0         NaN         NaN    NaN     NaN    NaN     NaN   \n",
       "ve4u5             283.0  451.784452  132.458823  196.0  367.00  432.0  515.50   \n",
       "wzrf2ge6          275.0  591.981818  186.033696  311.0  469.50  547.0  658.00   \n",
       "x8k59             289.0  488.622837  143.985096  210.0  380.00  478.0  565.00   \n",
       "xxmy9wd8          252.0  535.813492  242.239257   39.0  355.25  483.0  660.75   \n",
       "\n",
       "                 stop_rt_adjusted                                         \\\n",
       "             max            count        mean         std    min     25%   \n",
       "NARGUID                                                                    \n",
       "003RTV85  1336.0             28.0  546.107143   96.794681  395.0  496.25   \n",
       "007W6H7B  1242.0             26.0  562.000000  121.130673  405.0  474.75   \n",
       "00CY2MDM  1332.0             36.0  347.583333   92.117743  116.0  272.50   \n",
       "00HEV6HB  1688.0             38.0  373.947368  289.887298   83.0  256.00   \n",
       "00J52GPG  1165.0             36.0  305.194444   67.751993  245.0  265.00   \n",
       "...          ...              ...         ...         ...    ...     ...   \n",
       "vcahyykd     NaN              0.0         NaN         NaN    NaN     NaN   \n",
       "ve4u5     1053.0             27.0  387.851852  101.128518  264.0  303.00   \n",
       "wzrf2ge6  1520.0             27.0  502.111111  102.445232  247.0  429.50   \n",
       "x8k59     1081.0             28.0  385.892857   87.299119  239.0  321.00   \n",
       "xxmy9wd8  1653.0             32.0  382.218750  122.849694  126.0  314.75   \n",
       "\n",
       "                                 \n",
       "            50%     75%     max  \n",
       "NARGUID                          \n",
       "003RTV85  525.0  564.25   877.0  \n",
       "007W6H7B  541.5  645.25   859.0  \n",
       "00CY2MDM  349.0  405.50   530.0  \n",
       "00HEV6HB  343.5  403.00  1884.0  \n",
       "00J52GPG  285.5  319.25   543.0  \n",
       "...         ...     ...     ...  \n",
       "vcahyykd    NaN     NaN     NaN  \n",
       "ve4u5     370.0  443.50   595.0  \n",
       "wzrf2ge6  481.0  573.00   692.0  \n",
       "x8k59     369.5  460.75   561.0  \n",
       "xxmy9wd8  356.5  450.25   776.0  \n",
       "\n",
       "[8238 rows x 16 columns]"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rt_desc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Inhibition function (p(respond|SSD))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_p_resp_per_SSD(data, SSDs):\n",
    "    data = data.copy()\n",
    "    out_dict = {}\n",
    "    for ssd in SSDs:\n",
    "        curr_data = data.query(\n",
    "            \"SSDDur == %s\" % ssd\n",
    "        )\n",
    "        if len(curr_data) == 0:\n",
    "            out_dict[ssd] = np.nan\n",
    "        else:\n",
    "            out_dict[ssd] = len(curr_data.query(\"correct_stop == 0.0\")) / len(curr_data)\n",
    "    return out_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data = pd.read_csv('abcd_data/minimal_abcd_clean.csv')\n",
    "SSDs = [i for i in abcd_data.SSDDur.unique() if i==i]\n",
    "ssd_resp_dict = abcd_data.groupby('NARGUID').apply(get_p_resp_per_SSD, SSDs)\n",
    "ssd_resp_df = ssd_resp_dict.apply(pd.Series)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "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>50.0</th>\n",
       "      <th>100.0</th>\n",
       "      <th>150.0</th>\n",
       "      <th>0.0</th>\n",
       "      <th>200.0</th>\n",
       "      <th>250.0</th>\n",
       "      <th>300.0</th>\n",
       "      <th>350.0</th>\n",
       "      <th>400.0</th>\n",
       "      <th>450.0</th>\n",
       "      <th>500.0</th>\n",
       "      <th>550.0</th>\n",
       "      <th>600.0</th>\n",
       "      <th>650.0</th>\n",
       "      <th>700.0</th>\n",
       "      <th>750.0</th>\n",
       "      <th>800.0</th>\n",
       "      <th>850.0</th>\n",
       "      <th>900.0</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NARGUID</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",
       "      <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>003RTV85</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.090909</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.454545</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>007W6H7B</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.571429</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00CY2MDM</th>\n",
       "      <td>0.375000</td>\n",
       "      <td>0.625000</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.687500</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00HEV6HB</th>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.888889</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00J52GPG</th>\n",
       "      <td>0.631579</td>\n",
       "      <td>0.700000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vcahyykd</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ve4u5</th>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.444444</td>\n",
       "      <td>0.692308</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wzrf2ge6</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.272727</td>\n",
       "      <td>0.411765</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>x8k59</th>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.444444</td>\n",
       "      <td>0.363636</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.461538</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>xxmy9wd8</th>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.454545</td>\n",
       "      <td>0.625000</td>\n",
       "      <td>0.571429</td>\n",
       "      <td>0.285714</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8238 rows × 19 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             50.0      100.0     150.0     0.0       200.0     250.0  \\\n",
       "NARGUID                                                                \n",
       "003RTV85  0.000000  0.000000  0.000000       NaN  0.333333  0.090909   \n",
       "007W6H7B  0.000000  0.000000  0.250000       NaN  0.666667  0.000000   \n",
       "00CY2MDM  0.375000  0.625000  0.666667  0.687500  1.000000       NaN   \n",
       "00HEV6HB  0.666667  0.500000  0.250000  0.888889  0.500000  0.250000   \n",
       "00J52GPG  0.631579  0.700000  0.750000  0.500000  0.500000  1.000000   \n",
       "...            ...       ...       ...       ...       ...       ...   \n",
       "vcahyykd  0.000000  0.000000  0.000000       NaN  0.000000  0.000000   \n",
       "ve4u5     0.200000  0.500000  0.181818  0.000000  0.444444  0.692308   \n",
       "wzrf2ge6  0.000000  0.000000  0.000000       NaN  0.000000  0.272727   \n",
       "x8k59     0.166667  0.444444  0.363636  0.000000  0.461538  0.600000   \n",
       "xxmy9wd8  0.500000  0.454545  0.625000  0.571429  0.285714  0.666667   \n",
       "\n",
       "             300.0     350.0     400.0  450.0     500.0  550.0     600.0  \\\n",
       "NARGUID                                                                    \n",
       "003RTV85  0.666667  0.500000  0.454545    1.0       NaN    NaN       NaN   \n",
       "007W6H7B  0.666667  0.000000  0.250000    0.2  0.571429   0.75  0.666667   \n",
       "00CY2MDM       NaN       NaN       NaN    NaN       NaN    NaN       NaN   \n",
       "00HEV6HB  0.750000  1.000000       NaN    NaN       NaN    NaN       NaN   \n",
       "00J52GPG       NaN       NaN       NaN    NaN       NaN    NaN       NaN   \n",
       "...            ...       ...       ...    ...       ...    ...       ...   \n",
       "vcahyykd  0.000000  0.000000  0.000000    0.0  0.000000   0.00  0.000000   \n",
       "ve4u5     0.750000  0.000000  1.000000    NaN       NaN    NaN       NaN   \n",
       "wzrf2ge6  0.411765  0.600000  0.750000    1.0       NaN    NaN       NaN   \n",
       "x8k59     0.666667  0.666667  1.000000    NaN       NaN    NaN       NaN   \n",
       "xxmy9wd8  1.000000       NaN       NaN    NaN       NaN    NaN       NaN   \n",
       "\n",
       "          650.0  700.0  750.0  800.0  850.0  900.0  \n",
       "NARGUID                                             \n",
       "003RTV85    NaN    NaN    NaN    NaN    NaN    NaN  \n",
       "007W6H7B    0.5    1.0    NaN    NaN    NaN    NaN  \n",
       "00CY2MDM    NaN    NaN    NaN    NaN    NaN    NaN  \n",
       "00HEV6HB    NaN    NaN    NaN    NaN    NaN    NaN  \n",
       "00J52GPG    NaN    NaN    NaN    NaN    NaN    NaN  \n",
       "...         ...    ...    ...    ...    ...    ...  \n",
       "vcahyykd    0.0    0.0    0.0    0.0    0.0    0.0  \n",
       "ve4u5       NaN    NaN    NaN    NaN    NaN    NaN  \n",
       "wzrf2ge6    NaN    NaN    NaN    NaN    NaN    NaN  \n",
       "x8k59       NaN    NaN    NaN    NaN    NaN    NaN  \n",
       "xxmy9wd8    NaN    NaN    NaN    NaN    NaN    NaN  \n",
       "\n",
       "[8238 rows x 19 columns]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ssd_resp_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "ssd_resp_melt = ssd_resp_df.reset_index().melt(id_vars='NARGUID', value_vars=ssd_resp_df.columns, var_name='SSD', value_name='p_respond' )\n",
    "ssd_resp_melt['underlying distribution'] = 'ABCD data'\n",
    "ssd_resp_melt.to_csv('abcd_data/abcd_inhib_func_per_sub.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_inhib_func = pd.DataFrame(ssd_resp_df.mean())\n",
    "abcd_inhib_func.index.name = 'SSD'\n",
    "abcd_inhib_func.columns = ['p_respond']\n",
    "abcd_inhib_func = abcd_inhib_func.reset_index()\n",
    "abcd_inhib_func['underlying distribution'] = 'ABCD data'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_inhib_func.to_csv('abcd_data/abcd_inhib_func.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. build run_sims.sh"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "abcd_data = pd.read_csv('abcd_data/minimal_abcd_clean.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8207\n"
     ]
    }
   ],
   "source": [
    "narguids = abcd_data.NARGUID.unique()\n",
    "narguids = [sub for sub in narguids if sub not in problem_subs]\n",
    "print(len(narguids))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "171\n",
      "#SBATCH -N 5 # number of nodes requested - set to ceil(n rows in command script / 36)\n",
      "\n",
      "#SBATCH -n 171 # total number of mpi tasks requested - set to n rows in command script\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# TACC\n",
    "\n",
    "nsubs_per_job = 48\n",
    "njobs_per_node = 36\n",
    "nlines = 0\n",
    "with open('batch_files/TACC/run_sims.sh', 'w') as f:\n",
    "    for start_idx in range(0, len(narguids), nsubs_per_job):\n",
    "        end_idx = start_idx + nsubs_per_job\n",
    "        if end_idx > len(narguids):\n",
    "            end_idx = len(narguids)\n",
    "        substr = ' '.join(narguids[start_idx:end_idx])\n",
    "        f.write(f'python simulate_individuals.py --subjects {substr}\\n')\n",
    "        nlines += 1\n",
    "print(nlines)\n",
    "\n",
    "N_line_str = '#SBATCH -N %d # number of nodes requested - set to ceil(n rows in command script / 36)\\n' % int(np.ceil(nlines/njobs_per_node))\n",
    "n_line_str = '#SBATCH -n %s # total number of mpi tasks requested - set to n rows in command script\\n' % nlines\n",
    "\n",
    "replace(\n",
    "    'batch_files/TACC/launch_sim_cmds.slurm',\n",
    "    '#SBATCH -N',\n",
    "    N_line_str)\n",
    "replace(\n",
    "    'batch_files/TACC/launch_sim_cmds.slurm',\n",
    "    '#SBATCH -n',\n",
    "    n_line_str)\n",
    "# prints so you can compare the lines of the slurm file\n",
    "print(N_line_str)\n",
    "print(n_line_str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "# SHERLOCK\n",
    "\n",
    "nsubs_per_job = 72\n",
    "njobs_per_node = 24\n",
    "\n",
    "ABCD_LOC = '/oak/stanford/groups/russpold/users/henrymj/ABCD_simulations/abcd_data'\n",
    "OUT_LOC = '/oak/stanford/groups/russpold/users/henrymj/ABCD_simulations/simulated_data/individual_data'\n",
    "\n",
    "sher_header = '''#!/bin/bash\n",
    "#SBATCH --job-name=sims\n",
    "#SBATCH --output=.out/sims%d.out\n",
    "#SBATCH --error=.err/sims%d.err\n",
    "#SBATCH --time=3:00:00\n",
    "#SBATCH --mail-type=END\n",
    "#SBATCH --mail-user=henrymj@stanford.edu\n",
    "#SBATCH -N 1 # number of nodes requested - set to ceil(n rows in command script / 24)\n",
    "#SBATCH -n %d # total number of mpi tasks requested - set to n rows in command script\n",
    "#SBATCH -p russpold,normal\n",
    "# Job Submission\n",
    "#-----------\n",
    "export PYTHONPATH=\"\"\n",
    "\n",
    "source ~/miniconda3/etc/profile.d/conda.sh \n",
    "conda activate py3-env\n",
    "\n",
    "'''\n",
    "def make_sherlock_sim_batch_files(sher_sim_file, suffix):\n",
    "    nlines = 0\n",
    "    batch_counter = 0\n",
    "\n",
    "    file_str = sher_header\n",
    "    for start_idx in range(0, len(narguids), nsubs_per_job):\n",
    "        end_idx = start_idx + nsubs_per_job\n",
    "        if end_idx > len(narguids):\n",
    "            end_idx = len(narguids)\n",
    "        substr = ' '.join(narguids[start_idx:end_idx])\n",
    "        file_str += (f'eval \"python ../../scripts/simulate_individuals.py --mu_suffix {suffix} --abcd_dir {ABCD_LOC} --out_dir {OUT_LOC} --subjects {substr}\" &\\n')\n",
    "        nlines += 1\n",
    "        if nlines == 24:\n",
    "            with open(sher_sim_file % batch_counter, 'w') as f:\n",
    "                f.write(file_str % (batch_counter, batch_counter, nlines))\n",
    "                f.write(f'wait\\n')\n",
    "            # reset for new batch file\n",
    "            file_str = sher_header\n",
    "            nlines = 0\n",
    "            batch_counter += 1\n",
    "    # at end, if things didn't split out evenly, write out the remaining subs\n",
    "    if nlines > 0:\n",
    "        with open(sher_sim_file % batch_counter, 'w') as f:\n",
    "            f.write(file_str % (batch_counter, batch_counter, nlines))\n",
    "            f.write(f'wait\\n')\n",
    "            \n",
    "\n",
    "for SSRT_SCALE in SSRT_SCALES:\n",
    "    suffix=get_mu_suffix(SSRT_SCALE)\n",
    "    sher_file = 'batch_files/sherlock/%s/' % suffix + '/sherlock_run_sims_iter%d.batch'\n",
    "    make_sherlock_sim_batch_files(sher_file, suffix)\n",
    "    \n",
    "            \n",
    "\n",
    "# Single Individual Case\n",
    "suffix_noVar=get_mu_suffix(0,0)\n",
    "sher_file = 'batch_files/sherlock/%s/' % suffix_noVar + '/sherlock_run_sims_iter%d.batch'\n",
    "make_sherlock_sim_batch_files(sher_file, suffix_noVar)    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}