sentience_models.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Sentience Models Calculations\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import Data from Simulations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rmdir /Q /S sentience_estimates\n",
      "mkdir sentience_estimates\n",
      "### Testing to make sure score generation works ###\n",
      "All proxies with zero/one probabilities have scores equal to their expected values\n",
      "Number proxies whose proportion was outside 95% CI: 28\n",
      "Proportion of total proxies whose mean score was outside of the 95% CI: 0.052\n",
      "Probability of getting > 28/542 proxies outside of their 95% CI: 0.381\n",
      "Pass Test: \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import os\n",
    "import pickle\n",
    "import copy \n",
    "import collections\n",
    "import warnings\n",
    "import csv\n",
    "import platform\n",
    "import test_simulations\n",
    "import squigglepy as sq\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from logging import warning\n",
    "from operator import concat\n",
    "from struct import unpack\n",
    "from IPython.display import display\n",
    "\n",
    "warnings.filterwarnings('ignore')\n",
    "plt.rcParams['figure.figsize'] = (18.5 * 0.65, 10.5 * 0.65)\n",
    "\n",
    "SCENARIO_RANGES = [1, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 95, 99]\n",
    "\n",
    "params = pickle.load(open(os.path.join('input_data', \"Sentience Parameters.p\"), 'rb'))\n",
    "NUM_SCENARIOS = params['N_SCENARIOS']\n",
    "HC_WEIGHT = params['HC_WEIGHT']\n",
    "WEIGHT_NOS = params['WEIGHT_NOS']\n",
    "\n",
    "SPECIES = ['bees', 'cockroaches', 'fruit_flies', 'ants', \\\n",
    "            'c_elegans', 'crabs', 'crayfish', 'earthworms', \\\n",
    "            'sea_hares', 'spiders', 'octopuses', 'chickens', \\\n",
    "            'cows', 'sometimes_operates', 'bsf', \\\n",
    "            'carp', 'salmon', 'silkworms', 'pigs']\n",
    "\n",
    "# import simulated scores\n",
    "bee_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_bees\")), 'rb'))\n",
    "cockroach_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_cockroaches\")), 'rb'))\n",
    "fruit_fly_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_fruit_flies\")), 'rb'))\n",
    "ants_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_ants\")), 'rb'))\n",
    "c_elegans_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_c_elegans\")), 'rb'))\n",
    "crab_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_crabs\")), 'rb'))\n",
    "crayfish_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_crayfish\")), 'rb'))\n",
    "carp_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_carp\")), 'rb'))\n",
    "earthworm_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_earthworms\")), 'rb'))\n",
    "sea_hare_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_sea_hares\")), 'rb'))\n",
    "spiders_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_spiders\")), 'rb'))\n",
    "octopus_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_octopuses\")), 'rb'))\n",
    "chicken_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_chickens\")), 'rb'))\n",
    "cow_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_cows\")), 'rb'))\n",
    "sometimes_operates_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_sometimes_operates\")), 'rb'))\n",
    "bsf_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_bsf\")), 'rb'))\n",
    "salmon_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_salmon\")), 'rb'))\n",
    "silkworm_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_silkworms\")), 'rb'))\n",
    "pig_scores = pickle.load(open('{}_simulated_scores.p'.format(os.path.join('output_data', \"sent_pigs\")), 'rb'))\n",
    "\n",
    "unknown_probabilities = pickle.load(open(os.path.join('input_data', \"Sentience Unknown Probabilities.p\"), 'rb'))\n",
    "\n",
    "def run_cmd(cmd):\n",
    "    print(cmd)\n",
    "    os.system(cmd)\n",
    "\n",
    "if platform.system() == 'Darwin' or platform.system() == 'Linux':\n",
    "    run_cmd('rm -rf sentience_estimates')\n",
    "    run_cmd('mkdir sentience_estimates')\n",
    "elif platform.system() == 'Windows':\n",
    "    run_cmd('rmdir /Q /S sentience_estimates')\n",
    "    run_cmd('mkdir sentience_estimates')\n",
    "else:\n",
    "    raise ValueError('Platform `{}` not supported'.format(platform.system()))\n",
    "\n",
    "data = {'bees': {'Scores': bee_scores, 'Unknown Prob': unknown_probabilities['bees']}, \n",
    "        'cockroaches': {'Scores': cockroach_scores, 'Unknown Prob': unknown_probabilities['cockroaches']}, \n",
    "        'fruit_flies': {'Scores': fruit_fly_scores, 'Unknown Prob': unknown_probabilities['fruit_flies']}, \n",
    "        'ants': {'Scores': ants_scores, 'Unknown Prob': unknown_probabilities['ants']},\n",
    "        'c_elegans': {'Scores': c_elegans_scores, 'Unknown Prob': unknown_probabilities['c_elegans']}, \n",
    "        'crabs': {'Scores': crab_scores, 'Unknown Prob': unknown_probabilities['crabs']}, \n",
    "        'crayfish': {'Scores': crayfish_scores, 'Unknown Prob': unknown_probabilities['crayfish']}, \n",
    "        'earthworms': {'Scores': earthworm_scores, 'Unknown Prob': unknown_probabilities['earthworms']},\n",
    "        'sea_hares': {'Scores': sea_hare_scores, 'Unknown Prob': unknown_probabilities['sea_hares']}, \n",
    "        'spiders': {'Scores': spiders_scores, 'Unknown Prob': unknown_probabilities['spiders']},\n",
    "        'octopuses': {'Scores': octopus_scores, 'Unknown Prob': unknown_probabilities['octopuses']},\n",
    "        'chickens': {'Scores': chicken_scores, 'Unknown Prob': unknown_probabilities['chickens']},\n",
    "        'cows': {'Scores': cow_scores, 'Unknown Prob': unknown_probabilities['cows']},\n",
    "        'sometimes_operates': {'Scores': sometimes_operates_scores, 'Unknown Prob': unknown_probabilities['sometimes_operates']},\n",
    "        'bsf': {'Scores': bsf_scores, 'Unknown Prob': unknown_probabilities['bsf']},\n",
    "        'carp': {'Scores': carp_scores, 'Unknown Prob': unknown_probabilities['carp']},\n",
    "        'salmon': {'Scores': salmon_scores, 'Unknown Prob': unknown_probabilities['salmon']},\n",
    "        'silkworms': {'Scores': silkworm_scores, 'Unknown Prob': unknown_probabilities['silkworms']},\n",
    "        'pigs': {'Scores': pig_scores, 'Unknown Prob': unknown_probabilities['pigs']},\n",
    "        }\n",
    "\n",
    "print(\"### Testing to make sure score generation works ###\")\n",
    "test_simulations.test_sentience_scores(data, HC_WEIGHT, SPECIES)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Make Subsets of Proxies for Each Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import proxies lists for each model\n",
    "model_proxies_df = pd.read_csv(os.path.join('input_data', 'Sentience Model Proxies.csv'))\n",
    "\n",
    "# simple scoring proxies\n",
    "ss_proxies_lst = model_proxies_df['simple scoring'].dropna().values.tolist()\n",
    "ss_proxies = set(ss_proxies_lst)\n",
    "\n",
    "\n",
    "# first high-value proxies list\n",
    "hv1_proxies_lst = model_proxies_df['#1_high value proxies'].dropna().values.tolist()\n",
    "hv1_proxies = set(hv1_proxies_lst)\n",
    "\n",
    "# pleasure and pain centric proxies\n",
    "pp_proxies_lst = model_proxies_df['#1 pleasure and pain centric model'].dropna().values.tolist()\n",
    "pp_proxies = set(pp_proxies_lst)\n",
    "\n",
    "# Anna's high-value proxies\n",
    "hva_proxies_lst = model_proxies_df['high value proxies_anna'].dropna().values.tolist()\n",
    "hva_proxies = set(hva_proxies_lst)\n",
    "\n",
    "# Martina's high-value proxies\n",
    "hvm_proxies_lst = model_proxies_df['high value proxies_martina'].dropna().values.tolist()\n",
    "hvm_proxies = set(hvm_proxies_lst)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### General code for summing-based sentience ranges"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def filter_proxies(species_scores, model_proxies):\n",
    "    filtered_scores = {}\n",
    "    for proxy, scores_list in species_scores.items():\n",
    "        if proxy in model_proxies:\n",
    "            filtered_scores[proxy] = scores_list\n",
    "    return filtered_scores\n",
    "\n",
    "def get_sum(filtered_scores, sim_idx):\n",
    "    sum = 0\n",
    "    for scores_list in filtered_scores.values():\n",
    "        score_i = scores_list[sim_idx]\n",
    "        sum += score_i  \n",
    "    return sum\n",
    "\n",
    "def plot_range_distribution(species, ps_sentience_list):\n",
    "    ps_sentience_array = np.array(ps_sentience_list)\n",
    "    plt.hist(ps_sentience_array, bins=20, density=True)\n",
    "    plt.axvline(x=np.percentile(ps_sentience_array, SCENARIO_RANGES)[1], color='k', linestyle='dashed', linewidth=1)\n",
    "    plt.axvline(x=np.percentile(ps_sentience_array, SCENARIO_RANGES)[11], color='k', linestyle='dashed', linewidth=1)\n",
    "    plt.title(\"Distribution of {} Probabilities of Sentience\".format(species))\n",
    "    plt.show()\n",
    "    print('-')\n",
    "\n",
    "def one_species_summary_stats(species, ps_sentience_list, to_print=False):\n",
    "    ps_sentience_array = np.array(ps_sentience_list)\n",
    "    percentiles = np.percentile(ps_sentience_array, SCENARIO_RANGES)\n",
    "    mean = np.mean(ps_sentience_array)\n",
    "    fifth_percentile = percentiles[1]\n",
    "    ninty_fifth_percentile = percentiles[11]\n",
    "    median = percentiles[6]\n",
    "    stats_tuple = (mean, fifth_percentile, median, ninty_fifth_percentile)\n",
    "    if to_print:\n",
    "        print(species)\n",
    "        print(\"Mean probability of sentience: {}\".format(mean))\n",
    "        print(\"5th-percentile probability of sentience: {}\".format(fifth_percentile))\n",
    "        print(\"50th-percentile probability of sentience: {}\".format(median))\n",
    "        print(\"95th-percentile probability of sentience: {}\".format(ninty_fifth_percentile))\n",
    "    return stats_tuple\n",
    "\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Parameters for these Simulations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "For all the models below, the assumptions are that...\n",
      "'Likely no's and 'Lean no's are given probabilities between [0,0.25) and [0.25,0.5) of being true.\n",
      "Proxies we're higly confident matter for sentience are given 5.0 x the weight of other proxies.\n",
      "                    Unknown Prob.\n",
      "bees                            0\n",
      "cockroaches                     0\n",
      "fruit_flies                     0\n",
      "ants                            0\n",
      "c_elegans                       0\n",
      "crabs                           0\n",
      "crayfish                        0\n",
      "earthworms                      0\n",
      "sea_hares                       0\n",
      "spiders                         0\n",
      "octopuses                       0\n",
      "chickens                        0\n",
      "cows                            0\n",
      "sometimes_operates              0\n",
      "bsf                             0\n",
      "carp                            0\n",
      "salmon                          0\n",
      "silkworms                       0\n",
      "pigs                            0\n"
     ]
    }
   ],
   "source": [
    "print(\"For all the models below, the assumptions are that...\")\n",
    "\n",
    "if WEIGHT_NOS:\n",
    "    print(\"'Likely no's and 'Lean no's are given probabilities between [0,0.25) and [0.25,0.5) of being true.\")\n",
    "else:\n",
    "    print(\"'Likely no's and 'Lean no's are given no probability of being true.\")\n",
    "\n",
    "print(\"Proxies we're higly confident matter for sentience are given\", HC_WEIGHT, \"x the weight of other proxies.\")\n",
    "\n",
    "def unknown_probs_df():\n",
    "    unknown_probs = []\n",
    "    for species in SPECIES:\n",
    "        unknown_prob = data[species][\"Unknown Prob\"]\n",
    "        unknown_probs.append(unknown_prob)\n",
    "    cols = ['Unknown Prob.']\n",
    "    unknowns_df = pd.DataFrame(unknown_probs, columns=cols, index=SPECIES)\n",
    "    print(unknowns_df) \n",
    "\n",
    "unknown_probs_df()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Priors-Based Scoring"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Priors distribution reconstruction\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "species_lst = ['bees', 'cockroaches', 'fruit_flies', 'ants', 'c_elegans', 'crabs', 'crayfish', \\\n",
    "        'earthworms', 'sea_hares', 'spiders', 'octopuses', 'chickens', 'cows', 'bsf', \\\n",
    "        'carp', 'salmon', 'silkworms', 'pigs']\n",
    "\n",
    "species_caps = [\"Bees\", \"Cockroaches\", \"Fruit Flies\", \"Ants\", \\\n",
    "                \"C. elegans\", \"Crabs\", \"Crayfish\", \"Earthworms\", \\\n",
    "                    \"Sea Hares\", \"Spiders\", \"Octopuses\", \"Chickens\", \\\n",
    "                            \"Cows\", \"Black Soldier Flies\", \"Carp\", \"Salmon\", \"Silkworms\", \"Pigs\"]\n",
    "\n",
    "d_prob_map = {'very probably yes': {'lower': 0.9, 'upper': 1.0}, \n",
    "                           'probably yes': {'lower': 0.65, 'upper': 0.9},\n",
    "                           'possibly yes': {'lower': 0.5, 'upper': 0.65},\n",
    "                           'possibly no': {'lower': 0.35, 'upper': 0.5},\n",
    "                           'probably no': {'lower': 0.1, 'upper': 0.35},\n",
    "                           'very probably no': {'lower': 0.0, 'upper': 0.1}}\n",
    "\n",
    "d_judgments = {'bees': 'probably yes', 'cockroaches': 'possibly yes', 'fruit_flies': 'probably yes', \n",
    "                'ants': 'possibly yes', 'c_elegans': 'probably no', 'crayfish': 'probably yes', 'crabs': 'probably yes',\n",
    "                'earthworms': 'probably no', 'sea_hares': 'possibly no', 'spiders': 'possibly yes', \n",
    "                'octopuses': 'very probably yes', 'chickens': 'very probably yes', 'cows': 'very probably yes', \n",
    "                # this row was made up using similar results from like animals\n",
    "                'bsf': 'probably yes', 'carp': 'probably yes', 'salmon': 'probably yes', 'silkworms': 'probably no', 'pigs': 'very probably yes'}\n",
    "\n",
    "daniela_priors = {}\n",
    "for species in species_lst:\n",
    "        daniela_priors[species] = {'dist_type': 'normal', 'lower': d_prob_map[d_judgments[species]]['lower'], \n",
    "                                        'upper': d_prob_map[d_judgments[species]]['upper'], 'lclip': 0, 'rclip': 1}\n",
    "\n",
    "marcus_priors = {'bees': {'dist_type': 'lognormal', 'lower': 0.02, \n",
    "                'upper': 0.6, 'lclip': 0, 'rclip': 1}, \n",
    "        'cockroaches': {'dist_type': 'lognormal', 'lower': 0.01, \n",
    "                'upper': 0.4, 'lclip': 0, 'rclip': 1}, \n",
    "        'fruit_flies': {'dist_type': 'lognormal', 'lower': 0.04, \n",
    "                'upper': 0.55, 'lclip': 0, 'rclip': 1}, \n",
    "        'ants': {'dist_type': 'lognormal', 'lower': 0.02, \n",
    "                'upper': 0.6, 'lclip': 0, 'rclip': 1},\n",
    "        'c_elegans': {'dist_type': 'lognormal', 'lower': 0.001, \n",
    "                'upper': 0.01, 'lclip': 0, 'rclip': 0.01}, \n",
    "        'crabs': {'dist_type': 'lognormal', 'lower': 0.05, \n",
    "                'upper': 0.6, 'lclip': 0, 'rclip': 1}, \n",
    "        'crayfish': {'dist_type': 'lognormal', 'lower': 0.05, \n",
    "                'upper': 0.6, 'lclip': 0, 'rclip': 1}, \n",
    "        'earthworms': {'dist_type': 'lognormal', 'lower': 0.001, \n",
    "                'upper': 0.2, 'lclip': 0, 'rclip': 1},\n",
    "        'sea_hares': {'dist_type': 'lognormal', 'lower': 0.001, \n",
    "                'upper': 0.04, 'lclip': 0, 'rclip': 1}, \n",
    "        'spiders': {'dist_type': 'lognormal', 'lower': 0.01, \n",
    "                'upper': 0.4, 'lclip': 0, 'rclip': 1},\n",
    "        'octopuses': {'dist_type': 'normal', 'lower': 0.3, \n",
    "                'upper': 0.9, 'lclip': 0, 'rclip': 1},\n",
    "        'chickens': {'dist_type': 'normal', 'lower': 0.5, \n",
    "                'upper': 0.9, 'lclip': 0, 'rclip': 1},\n",
    "        'cows': {'dist_type': 'normal', 'lower': 0.6, \n",
    "                'upper': 0.9, 'lclip': 0, 'rclip': 1},\n",
    "        # put this as as same as fruit flies\n",
    "        'bsf': {'dist_type': 'normal', 'lower': 0.04, \n",
    "                'upper': 0.55, 'lclip': 0, 'rclip': 1},\n",
    "        # put this as same as crabs\n",
    "        'carp': {'dist_type': 'lognormal', 'lower': 0.05, \n",
    "                'upper': 0.6, 'lclip': 0, 'rclip': 1},\n",
    "        'salmon': {'dist_type': 'lognormal', 'lower': 0.05, \n",
    "                'upper': 0.6, 'lclip': 0, 'rclip': 1},\n",
    "        # put this as same as earthworms\n",
    "        'silkworms': {'dist_type': 'lognormal', 'lower': 0.001, \n",
    "                'upper': 0.2, 'lclip': 0, 'rclip': 1},\n",
    "        # put this as same as cows\n",
    "        'pigs': {'dist_type': 'normal', 'lower': 0.6, \n",
    "                'upper': 0.9, 'lclip': 0, 'rclip': 1},\n",
    "        \n",
    "        }\n",
    "\n",
    "peter_priors = {'bees': {'dist_type': 'normal', 'lower': 0.36, 'upper': 0.44, 'lclip': 0, 'rclip': 1}, \n",
    "        'cockroaches': {'dist_type': 'normal', 'lower': 0.18, 'upper': 0.22, 'lclip': 0, 'rclip': 1}, \n",
    "        'fruit_flies': {'dist_type': 'normal', 'lower': 0.27, 'upper': 0.33, 'lclip': 0, 'rclip': 1}, \n",
    "        'ants': {'dist_type': 'normal', 'lower': 0.225, 'upper': 0.275, 'lclip': 0, 'rclip': 1},\n",
    "        'c_elegans': {'dist_type': 'lognormal', 'lower': 0.0001, 'upper': 0.02, 'lclip': 0.0001, 'rclip': 1}, \n",
    "        'crabs': {'dist_type': 'normal', 'lower': 0.27, 'upper': 0.33, 'lclip': 0, 'rclip': 1}, \n",
    "        'crayfish': {'dist_type': 'normal', 'lower': 0.27, 'upper': 0.33, 'lclip': 0, 'rclip': 1}, \n",
    "        'earthworms': {'dist_type': 'normal', 'lower': 0.045, 'upper': 0.055, 'lclip': 0, 'rclip': 1},\n",
    "        'sea_hares': {'dist_type': 'normal', 'lower': 0.045, 'upper': 0.055, 'lclip': 0, 'rclip': 1}, \n",
    "        'spiders': {'dist_type': 'normal', 'lower': 0.225, 'upper': 0.275, 'lclip': 0, 'rclip': 1},\n",
    "        'octopuses': {'dist_type': 'normal', 'lower': 0.63, 'upper': 0.77, 'lclip': 0, 'rclip': 1},\n",
    "        'chickens': {'dist_type': 'normal', 'lower': 0.72, 'upper': 0.88, 'lclip': 0, 'rclip': 1},\n",
    "        'cows': {'dist_type': 'normal', 'lower': 0.765, 'upper': 0.935, 'lclip': 0, 'rclip': 1},\n",
    "        # same as fruit flies\n",
    "        'bsf': {'dist_type': 'normal', 'lower': 0.27, 'upper': 0.33, 'lclip': 0, 'rclip': 1},\n",
    "        # same as crabs\n",
    "        'carp': {'dist_type': 'normal', 'lower': 0.27, 'upper': 0.33, 'lclip': 0, 'rclip': 1},\n",
    "        'salmon': {'dist_type': 'normal', 'lower': 0.27, 'upper': 0.33, 'lclip': 0, 'rclip': 1},\n",
    "        # same as earthworms\n",
    "        'silkworms': {'dist_type': 'normal', 'lower': 0.045, 'upper': 0.055, 'lclip': 0, 'rclip': 1},\n",
    "        # same as cows\n",
    "        'pigs': {'dist_type': 'normal', 'lower': 0.765, 'upper': 0.935, 'lclip': 0, 'rclip': 1},\n",
    "        }\n",
    "\n",
    "priors_distributions = {'Daniela': daniela_priors, 'Marcus': marcus_priors, 'Peter': peter_priors}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Updating based on evidence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def simulate_priors(priors_distributions):\n",
    "    priors = {}\n",
    "    for species in species_lst:\n",
    "        models = []\n",
    "        for person in ['Daniela', 'Marcus', 'Peter']:\n",
    "            dist_type = priors_distributions[person][species]['dist_type']\n",
    "            lower = priors_distributions[person][species]['lower']\n",
    "            upper = priors_distributions[person][species]['upper']\n",
    "            lclip = priors_distributions[person][species]['lclip']\n",
    "            rclip = priors_distributions[person][species]['rclip']\n",
    "            if dist_type == 'lognormal':\n",
    "                model = sq.lognorm(x=lower, y=upper, credibility = 90, lclip=lclip, rclip=rclip)\n",
    "            elif dist_type == 'normal':\n",
    "                model = sq.norm(x=lower, y=upper, credibility = 90, lclip=lclip, rclip=rclip)\n",
    "            models.append(model)\n",
    "        prior_lst = sq.sample(sq.mixture(models, [1/3, 1/3, 1/3]), NUM_SCENARIOS)\n",
    "        priors[species] = prior_lst\n",
    "    return priors\n",
    "\n",
    "priors = simulate_priors(priors_distributions)\n",
    "\n",
    "def shrimp_probability_sentience(priors):\n",
    "    models = []\n",
    "    for person in ['Daniela', 'Marcus', 'Peter']:\n",
    "        dist_type = priors_distributions[person]['crabs']['dist_type']\n",
    "        lower = priors_distributions[person]['crabs']['lower']\n",
    "        upper = priors_distributions[person]['crabs']['upper']\n",
    "        lclip = priors_distributions[person]['crabs']['lclip']\n",
    "        rclip = priors_distributions[person]['crabs']['rclip']\n",
    "        if dist_type == 'lognormal':\n",
    "            model = sq.lognorm(x=lower, y=upper, credibility = 90, lclip=lclip, rclip=rclip)\n",
    "        elif dist_type == 'normal':\n",
    "            model = sq.norm(x=lower, y=upper, credibility = 90, lclip=lclip, rclip=rclip)\n",
    "        models.append(model)\n",
    "    shrimp_prior_lst = sq.sample(sq.mixture(models, [1/3, 1/3, 1/3]), NUM_SCENARIOS)\n",
    "    pickle.dump(shrimp_prior_lst, open(os.path.join('sentience_estimates', 'shrimp_assumed_psent.p'), 'wb'))\n",
    "    return shrimp_prior_lst  \n",
    "\n",
    "shrimp_probability_sentience(priors)\n",
    "\n",
    "def one_sim_p_sentience_priors_based(f, priors, species, model_proxies, species_scores, sometimes_operates_scores, sim_idx):\n",
    "    s_filtered_scores = filter_proxies(species_scores, model_proxies)\n",
    "    so_filtered_scores = filter_proxies(sometimes_operates_scores, model_proxies)\n",
    "    species_sum = get_sum(s_filtered_scores, sim_idx)\n",
    "    sometimes_operates_sum = get_sum(so_filtered_scores, sim_idx)\n",
    "    prior = priors[species][sim_idx]\n",
    "\n",
    "    p_sentience = min(prior*(species_sum/sometimes_operates_sum)**0.25,0.99)\n",
    "    p_sentience = max(p_sentience, 0)\n",
    "    return p_sentience\n",
    "\n",
    "def one_species_ps_sentience_priors_based(f, priors, species, species_scores, model_proxies, model_name): \n",
    "    ps_sentience_list = []\n",
    "    for i in range(NUM_SCENARIOS):\n",
    "        p_sentience_i = one_sim_p_sentience_priors_based(f, priors, species, model_proxies, species_scores, sometimes_operates_scores, i)\n",
    "        ps_sentience_list.append(p_sentience_i)\n",
    "\n",
    "    if model_name == \"#1_high value proxies\":\n",
    "        pickle.dump(ps_sentience_list, open(os.path.join('sentience_estimates', '{}_psent_hv1_model.p'.format(species)), 'wb'))\n",
    "    \n",
    "    return ps_sentience_list\n",
    "\n",
    "def plot_range_distribution(species, ps_sentience_list):\n",
    "    ps_sentience_array = np.array(ps_sentience_list)\n",
    "    plt.hist(ps_sentience_array, bins=20, density=True)\n",
    "    plt.axvline(x=np.percentile(ps_sentience_array, SCENARIO_RANGES)[1], color='k', linestyle='dashed', linewidth=1)\n",
    "    plt.axvline(x=np.percentile(ps_sentience_array, SCENARIO_RANGES)[11], color='k', linestyle='dashed', linewidth=1)\n",
    "    plt.title(\"Distribution of {} Probabilities of Sentience\".format(species))\n",
    "    plt.show()\n",
    "    print('-')\n",
    "\n",
    "def print_priors(priors, to_plot):\n",
    "    means = []\n",
    "    fifth_percentiles = []\n",
    "    medians = []\n",
    "    ninty_fifth_percentiles = []\n",
    "    species_sent_data = []\n",
    "\n",
    "    for species in species_lst: \n",
    "        means.append(round(np.mean(priors[species]),3))\n",
    "        fifth_percentiles.append(round(np.percentile(priors[species], 5),3))\n",
    "        medians.append(round(np.percentile(priors[species], 50),3))\n",
    "        ninty_fifth_percentiles.append(round(np.percentile(priors[species], 95),3))\n",
    "\n",
    "    cols = [\"Mean\",\"5th-pct\", \"50th-pct\", \"95th-pct\"]\n",
    "    p_sentience_stats_df = pd.DataFrame(list(zip(means, fifth_percentiles, medians, ninty_fifth_percentiles)), \\\n",
    "        columns=cols, index=species_lst)\n",
    "    p_sentience_stats_df = p_sentience_stats_df.sort_values(\"50th-pct\", ascending=False)\n",
    "    path = os.path.join('sentience_estimates', \"Priors Sentience Summary Statistics.csv\")\n",
    "    p_sentience_stats_df.to_csv(path, index_label=\"Species\")\n",
    "\n",
    "    return p_sentience_stats_df\n",
    "\n",
    "def all_species_p_sentience_priors_based(f, priors, model_name, data, model_proxies, to_plot=False):\n",
    "    means = []\n",
    "    fifth_percentiles = []\n",
    "    medians = []\n",
    "    ninty_fifth_percentiles = []\n",
    "    species_sent_data = []\n",
    "\n",
    "    for species in species_lst: \n",
    "        species_scores = data[species][\"Scores\"]\n",
    "        species_p_sentience_lst = one_species_ps_sentience_priors_based(f, priors, species, species_scores, model_proxies, model_name)\n",
    "        if to_plot:\n",
    "            plot_range_distribution(species, species_p_sentience_lst)\n",
    "        species_sent_data.append(species_p_sentience_lst)\n",
    "        species_stats = one_species_summary_stats(species, species_p_sentience_lst)\n",
    "        means.append(round(species_stats[0],3))\n",
    "        fifth_percentiles.append(round(species_stats[1],3))\n",
    "        medians.append(round(species_stats[2],3))\n",
    "        ninty_fifth_percentiles.append(round(species_stats[3],3))\n",
    "\n",
    "    cols = [\"Mean\",\"5th-pct\", \"50th-pct\", \"95th-pct\"]\n",
    "    p_sentience_stats_df = pd.DataFrame(list(zip(means, fifth_percentiles, medians, ninty_fifth_percentiles)), \\\n",
    "        columns=cols, index=species_lst)\n",
    "    p_sentience_stats_df = p_sentience_stats_df.sort_values(\"50th-pct\", ascending=False)\n",
    "    path = os.path.join('sentience_estimates', \"Sent {} Summary Statistics.csv\".format(model_name))\n",
    "\n",
    "    p_sentience_stats_df.to_csv(path, index_label=\"Species\")\n",
    "\n",
    "    return p_sentience_stats_df, species_sent_data\n",
    "\n",
    "def box_plot_adj_wr(model_name, species_adj_wrs, showfliers=True):\n",
    "    sns.set_style(style='white')\n",
    "    fig, ax = plt.subplots(figsize=(10, 6))\n",
    "    sns.boxplot(data=species_adj_wrs, orient='h', ax=ax, showfliers=showfliers)\n",
    "    ax.set_yticklabels(species_caps)\n",
    "    ax.set_title(\"Probability of Sentience - {} Model\".format(model_name))\n",
    "    fig.savefig(os.path.join('sentience_estimates', \"{} Probability of Sentience  - Box Plot.png\".format(model_name)), dpi=300, bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Priors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Priors\n",
      "              Mean  5th-pct  50th-pct  95th-pct\n",
      "pigs         0.850    0.659     0.864     0.983\n",
      "cows         0.850    0.657     0.863     0.983\n",
      "chickens     0.817    0.576     0.822     0.983\n",
      "octopuses    0.749    0.416     0.729     0.985\n",
      "bees         0.450    0.037     0.406     0.863\n",
      "bsf          0.457    0.135     0.325     0.853\n",
      "crabs        0.434    0.080     0.311     0.856\n",
      "crayfish     0.434    0.078     0.311     0.862\n",
      "salmon       0.433    0.075     0.311     0.861\n",
      "carp         0.432    0.077     0.310     0.858\n",
      "fruit_flies  0.425    0.064     0.309     0.859\n",
      "ants         0.334    0.036     0.258     0.632\n",
      "spiders      0.316    0.020     0.255     0.625\n",
      "cockroaches  0.297    0.020     0.205     0.630\n",
      "silkworms    0.110    0.003     0.052     0.317\n",
      "earthworms   0.107    0.003     0.051     0.311\n",
      "sea_hares    0.160    0.002     0.050     0.470\n",
      "c_elegans    0.078    0.000     0.005     0.306\n"
     ]
    }
   ],
   "source": [
    "priors_df = print_priors(priors, to_plot=False)\n",
    "print(\"Priors\")\n",
    "print(priors_df)\n",
    "priors_df.to_csv(os.path.join('sentience_estimates', \"Priors Sentience Summary Statistics.csv\"), index_label=\"Species\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Simple Scoring"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Simple Scoring Model:\n",
      "              Mean  5th-pct  50th-pct  95th-pct\n",
      "pigs         0.923    0.733     0.967     0.990\n",
      "cows         0.907    0.711     0.941     0.990\n",
      "chickens     0.873    0.625     0.898     0.990\n",
      "octopuses    0.783    0.439     0.774     0.990\n",
      "bees         0.467    0.038     0.423     0.905\n",
      "fruit_flies  0.449    0.067     0.327     0.914\n",
      "carp         0.452    0.080     0.327     0.906\n",
      "crayfish     0.452    0.082     0.325     0.903\n",
      "salmon       0.449    0.077     0.325     0.901\n",
      "crabs        0.429    0.078     0.309     0.855\n",
      "ants         0.305    0.033     0.238     0.589\n",
      "bsf          0.307    0.090     0.225     0.589\n",
      "spiders      0.264    0.017     0.213     0.532\n",
      "cockroaches  0.269    0.018     0.186     0.577\n",
      "earthworms   0.086    0.002     0.042     0.253\n",
      "sea_hares    0.133    0.002     0.042     0.396\n",
      "silkworms    0.084    0.002     0.040     0.243\n",
      "c_elegans    0.069    0.000     0.004     0.271\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# simple scoring function\n",
    "def ss_f(sentience_sum):\n",
    "    adjusted_sentience_sum = sentience_sum\n",
    "    return adjusted_sentience_sum\n",
    "\n",
    "ss_sent_stats, ss_sent_data = all_species_p_sentience_priors_based(ss_f, priors, \"simple scoring\", data, ss_proxies, to_plot=False)\n",
    "print(\"Simple Scoring Model:\")\n",
    "print(ss_sent_stats)\n",
    "box_plot_adj_wr(\"Simple Scoring\", ss_sent_data, showfliers=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### #1 High Value Proxies \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "#1 High-Value Proxies Model:\n",
      "              Mean  5th-pct  50th-pct  95th-pct\n",
      "pigs         0.925    0.737     0.973     0.990\n",
      "cows         0.910    0.712     0.945     0.990\n",
      "chickens     0.876    0.629     0.904     0.990\n",
      "octopuses    0.783    0.439     0.775     0.990\n",
      "bees         0.466    0.038     0.422     0.905\n",
      "fruit_flies  0.451    0.067     0.329     0.919\n",
      "carp         0.453    0.081     0.328     0.912\n",
      "crayfish     0.453    0.082     0.327     0.910\n",
      "salmon       0.451    0.078     0.327     0.909\n",
      "crabs        0.426    0.078     0.308     0.850\n",
      "ants         0.300    0.032     0.234     0.580\n",
      "bsf          0.294    0.079     0.218     0.574\n",
      "spiders      0.257    0.016     0.208     0.521\n",
      "cockroaches  0.265    0.018     0.184     0.570\n",
      "earthworms   0.083    0.002     0.041     0.244\n",
      "sea_hares    0.128    0.002     0.040     0.384\n",
      "silkworms    0.082    0.002     0.039     0.239\n",
      "c_elegans    0.068    0.000     0.004     0.268\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#1_High-Value Proxies Function\n",
    "def hv1_f(sentience_sum):\n",
    "    adjusted_sentience_sum = sentience_sum\n",
    "    return adjusted_sentience_sum\n",
    "\n",
    "hv1_sent_stats, hv1_sent_data = all_species_p_sentience_priors_based(hv1_f, priors, \"#1_high value proxies\", data, hv1_proxies, to_plot=False)\n",
    "print(\"#1 High-Value Proxies Model:\")\n",
    "print(hv1_sent_stats)\n",
    "box_plot_adj_wr(\"#1 High-Value Proxies\", hv1_sent_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Martina's High-Value Proxies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Martina's High-Value Proxies Model:\n",
      "              Mean  5th-pct  50th-pct  95th-pct\n",
      "pigs         0.926    0.739     0.975     0.990\n",
      "cows         0.911    0.714     0.947     0.990\n",
      "chickens     0.877    0.629     0.906     0.990\n",
      "octopuses    0.784    0.440     0.777     0.990\n",
      "bees         0.467    0.038     0.424     0.908\n",
      "fruit_flies  0.452    0.067     0.330     0.921\n",
      "carp         0.454    0.081     0.329     0.914\n",
      "crayfish     0.454    0.082     0.328     0.912\n",
      "salmon       0.452    0.078     0.327     0.911\n",
      "crabs        0.428    0.078     0.309     0.853\n",
      "ants         0.301    0.033     0.235     0.583\n",
      "bsf          0.298    0.084     0.221     0.580\n",
      "spiders      0.259    0.016     0.209     0.525\n",
      "cockroaches  0.266    0.018     0.184     0.573\n",
      "earthworms   0.084    0.002     0.041     0.245\n",
      "sea_hares    0.129    0.002     0.041     0.387\n",
      "silkworms    0.083    0.002     0.040     0.241\n",
      "c_elegans    0.068    0.000     0.004     0.270\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "hvm_sent_stats, hvm_sent_data = all_species_p_sentience_priors_based(ss_f, priors, \"Martina's High-Value Proxies\", data, hvm_proxies, to_plot=False)\n",
    "print(\"Martina's High-Value Proxies Model:\")\n",
    "print(hvm_sent_stats)\n",
    "box_plot_adj_wr(\"Martina's High-Value Proxies\", hvm_sent_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Anna's High-Value Proxies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Anna's High-Value Proxies Model:\n",
      "              Mean  5th-pct  50th-pct  95th-pct\n",
      "pigs         0.924    0.735     0.971     0.990\n",
      "cows         0.908    0.711     0.942     0.990\n",
      "chickens     0.874    0.626     0.901     0.990\n",
      "octopuses    0.783    0.439     0.775     0.990\n",
      "bees         0.466    0.038     0.422     0.905\n",
      "fruit_flies  0.450    0.067     0.328     0.916\n",
      "carp         0.452    0.080     0.327     0.908\n",
      "crayfish     0.452    0.082     0.326     0.906\n",
      "salmon       0.450    0.078     0.326     0.905\n",
      "crabs        0.427    0.078     0.308     0.851\n",
      "ants         0.301    0.033     0.235     0.583\n",
      "bsf          0.293    0.078     0.218     0.572\n",
      "spiders      0.257    0.016     0.208     0.521\n",
      "cockroaches  0.264    0.018     0.183     0.568\n",
      "earthworms   0.083    0.002     0.041     0.243\n",
      "sea_hares    0.128    0.002     0.040     0.383\n",
      "silkworms    0.082    0.002     0.039     0.238\n",
      "c_elegans    0.068    0.000     0.004     0.267\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "hva_sent_stats, hva_sent_data = all_species_p_sentience_priors_based(ss_f, priors, \"Anna's High-Value Proxies\", data, hva_proxies, to_plot=False)\n",
    "print(\"Anna's High-Value Proxies Model:\")\n",
    "print(hva_sent_stats)\n",
    "box_plot_adj_wr(\"Anna's High-Value Proxies\", hva_sent_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Birch Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rmdir /Q /S birch_estimates\n",
      "mkdir birch_estimates\n",
      "Birch Model Sentience Estimates\n",
      "              Mean  5th-pct  50th-pct  95th-pct\n",
      "pigs         0.921    0.732     0.966     0.990\n",
      "cows         0.901    0.703     0.931     0.990\n",
      "chickens     0.865    0.616     0.888     0.990\n",
      "octopuses    0.769    0.431     0.758     0.990\n",
      "bees         0.452    0.037     0.409     0.877\n",
      "carp         0.452    0.080     0.327     0.907\n",
      "fruit_flies  0.445    0.067     0.326     0.907\n",
      "salmon       0.446    0.076     0.323     0.897\n",
      "crayfish     0.445    0.081     0.321     0.894\n",
      "crabs        0.424    0.077     0.306     0.848\n",
      "bsf          0.289    0.083     0.219     0.575\n",
      "ants         0.268    0.029     0.209     0.519\n",
      "spiders      0.238    0.015     0.194     0.490\n",
      "cockroaches  0.259    0.017     0.180     0.558\n",
      "earthworms   0.085    0.002     0.042     0.249\n",
      "sea_hares    0.133    0.002     0.042     0.398\n",
      "silkworms    0.081    0.002     0.039     0.235\n",
      "c_elegans    0.069    0.000     0.004     0.272\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import platform\n",
    "import os\n",
    "\n",
    "def run_cmd(cmd):\n",
    "    print(cmd)\n",
    "    os.system(cmd)\n",
    "\n",
    "if platform.system() == 'Darwin' or platform.system() == 'Linux':\n",
    "    run_cmd('rm -rf birch_estimates')\n",
    "    run_cmd('mkdir birch_estimates')\n",
    "\n",
    "elif platform.system() == 'Windows':\n",
    "    run_cmd('rmdir /Q /S birch_estimates')\n",
    "    run_cmd('mkdir birch_estimates')\n",
    "\n",
    "birch_proxies = pd.read_csv(os.path.join('input_data', 'Birch Model Proxies.csv'))\n",
    "\n",
    "def get_overlap(criterion_proxies, species_scores):\n",
    "    criterion_scores = {}\n",
    "    for proxy in criterion_proxies:\n",
    "        criterion_scores[proxy] = species_scores[proxy]\n",
    "\n",
    "    return criterion_scores\n",
    "\n",
    "def one_species_birch_sum(birch_proxies, species_scores):\n",
    "    birch_sums = {}\n",
    "    for criterion in birch_proxies.columns:\n",
    "        birch_sums[criterion] = []\n",
    "        criterion_proxies = birch_proxies[criterion].dropna().values.tolist()\n",
    "        criterion_scores = get_overlap(criterion_proxies, species_scores)\n",
    "\n",
    "        for i in range(NUM_SCENARIOS):\n",
    "            criterion_sum_i = 0\n",
    "            for proxy in criterion_scores:\n",
    "                crit_score_i = criterion_scores[proxy][i]\n",
    "                criterion_sum_i += crit_score_i\n",
    "        \n",
    "            birch_sums[criterion].append(criterion_sum_i)\n",
    "\n",
    "    return birch_sums\n",
    "\n",
    "def one_species_birch_est(priors, birch_proxies, species_scores, species):\n",
    "    so_birch_sums = one_species_birch_sum(birch_proxies, sometimes_operates_scores)\n",
    "    species_birch_sums = one_species_birch_sum(birch_proxies, species_scores)\n",
    "    \n",
    "    species_birch_scores = {}\n",
    "\n",
    "    means = []\n",
    "    fifth_percentiles = []\n",
    "    medians = []\n",
    "    ninty_fifth_percentiles = []\n",
    "    criteria = ['overall'] \n",
    "\n",
    "    overall_scores = []\n",
    "    for i in range(NUM_SCENARIOS):\n",
    "        prior = priors[species][i]\n",
    "        score_i = 0\n",
    "        so_score_i = 0\n",
    "        for c in species_birch_sums:\n",
    "            score_i += species_birch_sums[c][i]\n",
    "            so_score_i += so_birch_sums[c][i]\n",
    "        posterior = min(prior*(score_i/(so_score_i))**0.25,0.99)\n",
    "        posterior = max(posterior,0)\n",
    "        overall_scores.append(posterior)\n",
    "    overall_birch_array = np.array(overall_scores)\n",
    "    overall_birch_percentiles = np.percentile(overall_birch_array, SCENARIO_RANGES)\n",
    "    \n",
    "    overall_mean = np.mean(overall_birch_array)\n",
    "    fifth_percentile = overall_birch_percentiles[1]\n",
    "    ninty_fifth_percentile = overall_birch_percentiles[11]\n",
    "    median = overall_birch_percentiles[6]\n",
    "\n",
    "    means.append(round(overall_mean,3))\n",
    "    fifth_percentiles.append(round(fifth_percentile,3))\n",
    "    medians.append(round(median,3))\n",
    "    ninty_fifth_percentiles.append(round(ninty_fifth_percentile,3))\n",
    "\n",
    "    for criterion in species_birch_sums:\n",
    "        criteria.append(criterion)\n",
    "        species_birch_scores[criterion] = []\n",
    "        crit_list = species_birch_sums[criterion]\n",
    "        for i in range(len(crit_list)):\n",
    "            species_birch_scores[criterion].append(species_birch_sums[criterion][i])\n",
    "        \n",
    "        criterion_birch_array = np.array(species_birch_scores[criterion])\n",
    "        criterion_birch_percentiles = np.percentile(criterion_birch_array, SCENARIO_RANGES)\n",
    "\n",
    "        mean = np.mean(criterion_birch_array)\n",
    "        fifth_percentile = criterion_birch_percentiles[1]\n",
    "        ninty_fifth_percentile = criterion_birch_percentiles[11]\n",
    "        median = criterion_birch_percentiles[6]\n",
    "\n",
    "        means.append(round(mean,3))\n",
    "        fifth_percentiles.append(round(fifth_percentile,3))\n",
    "        medians.append(round(median,3))\n",
    "        ninty_fifth_percentiles.append(round(ninty_fifth_percentile,3))\n",
    "\n",
    "    cols = [\"Mean\", \"5th-pct\", \"50th-pct\", \"95th-pct\"]\n",
    "    crit = species_birch_scores.keys()\n",
    "    birch_sentience_stats_df = pd.DataFrame(list(zip(means, fifth_percentiles, medians, ninty_fifth_percentiles)), \\\n",
    "        columns=cols, index=criteria)\n",
    "        \n",
    "    return birch_sentience_stats_df\n",
    "\n",
    "def all_species_birch_est():\n",
    "    mean = []\n",
    "    fifth_pctiles = []\n",
    "    med_pctiles = []\n",
    "    ninetyfifth_pctiles = []\n",
    "    species_sent_data = []\n",
    "\n",
    "    for species in species_lst:\n",
    "        species_scores = data[species]['Scores']\n",
    "        spec_birch_est = one_species_birch_est(priors, birch_proxies, species_scores, species)\n",
    "        path = os.path.join('birch_estimates', \"{}_birch_estimates.csv\".format(species))\n",
    "        spec_birch_est.to_csv(path, index_label=\"Species\")\n",
    "        spec_overall = spec_birch_est.loc[['overall']]\n",
    "        species_sent_data.append(spec_overall)\n",
    "        mean.append(spec_overall['Mean'].values[0])\n",
    "        fifth_pctiles.append(spec_overall['5th-pct'].values[0])\n",
    "        med_pctiles.append(spec_overall['50th-pct'].values[0])\n",
    "        ninetyfifth_pctiles.append(spec_overall['95th-pct'].values[0])\n",
    "        \n",
    "    birch_overalls = pd.DataFrame(list(zip(mean, fifth_pctiles, med_pctiles, ninetyfifth_pctiles)), \\\n",
    "        columns=[\"Mean\",'5th-pct', '50th-pct', '95th-pct'],index=species_lst)\n",
    "    birch_overalls = birch_overalls.sort_values(\"50th-pct\", ascending=False)\n",
    "    path = os.path.join('sentience_estimates', \"Sent {} Summary Statistics.csv\".format(\"Birch Model\"))\n",
    "    birch_overalls.to_csv(path, index_label=\"Species\")\n",
    "    return birch_overalls, species_sent_data\n",
    "\n",
    "birch_overalls, birch_sent_data = all_species_birch_est()\n",
    "print(\"Birch Model Sentience Estimates\")\n",
    "print(birch_overalls)\n",
    "box_plot_adj_wr(\"Birch Model\", birch_sent_data)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tests "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "test_one_species_birch_est (__main__.BirchModelTests) ... ok\n",
      "test_one_species_birch_sum (__main__.BirchModelTests) ... ok\n",
      "test_proxies (__main__.BirchModelTests) ... ok\n",
      "test_get_sum (__main__.TestSimpleFunctions) ... ok\n",
      "test_one_sim_p_sent_priors_based (__main__.TestSimpleFunctions) ... ok\n",
      "test_proxies (__main__.TestSimpleFunctions) ... ok\n",
      "\n",
      "----------------------------------------------------------------------\n",
      "Ran 6 tests in 1.691s\n",
      "\n",
      "OK\n"
     ]
    }
   ],
   "source": [
    "import unittest\n",
    "\n",
    "proxies = {'simple scoring': {'List': ss_proxies_lst, 'Set': ss_proxies}, \n",
    "            '#1_high value proxies': {'List': hv1_proxies_lst, 'Set': hv1_proxies}, \n",
    "            '#1 pleasure and pain centric model': {\"List\": pp_proxies_lst, 'Set': pp_proxies},\n",
    "            'high value proxies_anna': {\"List\": hva_proxies_lst, 'Set': hva_proxies}, \n",
    "            'high value proxies_martina': {'List': hvm_proxies_lst, 'Set': hvm_proxies}}\n",
    "\n",
    "models = {'Simple Scoring': {'Proxies': ss_proxies, 'Function': ss_f},\n",
    "        '#1 High-Value Proxies': {'Proxies': hv1_proxies, 'Function': hv1_f},\n",
    "        \"Martina's High-Value Proxies\": {'Proxies': hvm_proxies, 'Function': ss_f}, \n",
    "        \"Anna's High-Value Proxies\": {'Proxies': hva_proxies, 'Function': ss_f}}   \n",
    "\n",
    "class TestSimpleFunctions(unittest.TestCase):\n",
    "\n",
    "    def test_proxies(self):\n",
    "        pass_test = True\n",
    "        for model in proxies:\n",
    "            if set(proxies[model]['List']) != proxies[model]['Set']:\n",
    "                pass_test = False\n",
    "        self.assertTrue(pass_test)\n",
    "\n",
    "    def test_get_sum(self):\n",
    "        pass_test = True\n",
    "        for x in [0,1]:  \n",
    "            for model in proxies:\n",
    "                test_data = {}\n",
    "                model_proxies = proxies[model]['Set']\n",
    "                for proxy in model_proxies:\n",
    "                    test_data[proxy] = [x]*NUM_SCENARIOS\n",
    "                for i in range(NUM_SCENARIOS):\n",
    "                    expected = x*len(model_proxies)\n",
    "                    actual = get_sum(test_data, i)\n",
    "                    if expected != actual:\n",
    "                        pass_test = False\n",
    "        self.assertTrue(pass_test)\n",
    "\n",
    "    def test_one_sim_p_sent_priors_based(self):\n",
    "        pass_test = True\n",
    "        fake_so_scores = {}\n",
    "        fake_priors = {'pigs': np.array([0.5]*NUM_SCENARIOS)}\n",
    "        for proxy in ss_proxies:\n",
    "            fake_so_scores[proxy] = [1]*NUM_SCENARIOS\n",
    "        for x in [0, 1]:\n",
    "            test_scores = {}\n",
    "            for proxy in ss_proxies:\n",
    "                test_scores[proxy] = [x]*NUM_SCENARIOS\n",
    "            for model in models:\n",
    "                model_proxies = models[model]['Proxies']\n",
    "                f = models[model]['Function']\n",
    "                prediction = 0.5*(f(x)/f(1))**0.25\n",
    "                for i in range(NUM_SCENARIOS):\n",
    "                    p = one_sim_p_sentience_priors_based(f, fake_priors, species, model_proxies, test_scores, fake_so_scores, i)\n",
    "                    if p != prediction:\n",
    "                        pass_test = False\n",
    "                        print(\"prediction: {}\".format(prediction))\n",
    "                        print(\"p: {}\".format(p))\n",
    "        self.assertTrue(pass_test)\n",
    "\n",
    "birch_proxies_dict = {\"1: possession of nociceptors\": ['physiological responses to nociception or handling', \\\n",
    "                                                        'nociception', 'nociception (strict definition)', \\\n",
    "                                                        'noxious stimuli related vocalizations', 'movement away from noxious stimuli'],\n",
    "                    \"2: possession of integrative brain regions\": ['centralized information processing', \\\n",
    "                                                        'vertebrate midbrain-like function'],\n",
    "                    \"3: possession of neural pathways connecting nociceptiorrs to brain regions\": ['taste aversion behavior', \\\n",
    "                                                        'pain relief learning', 'long-term behavior alteration to avoid noxious stimulus (24+ hours)', \\\n",
    "                                                        'long-term behavior alteration to avoid noxious stimuli (30+ days)'], \n",
    "                    \"4: behavioural response modulated by chemical compounds\": ['affected by analgesics in a manner similar to humans', \\\n",
    "                                                        'opioid-like receptors', 'affected by recreational drugs in a similar manner to humans', \\\n",
    "                                                        'affected by antidepressants or anxiolytics in a similar manner to humans'], \n",
    "                    '5: evidence of motivational trade-offs': ['paying a cost to receive a reward', 'paying a cost to avoid a noxious stimulus', \\\n",
    "                                                        'self-control', 'predator avoidance tradeoffs', 'selective attention to noxious stimuli over other concurrent events'], \n",
    "                    \"6: flexible self-protective behaviour\": ['protective behavior', 'defensive behavior/fighting back'], \n",
    "                    \"7: evidence of associative learning\": ['classical conditioning', 'operant conditioning', 'operant conditioning with unfamiliar action', \\\n",
    "                                                        'contextual learning', 'observational or social learning', 'taste aversion behavior', 'pain relief learning', \\\n",
    "                                                        'long-term behavior alteration to avoid noxious stimulus (24+ hours)', 'long-term behavior alteration to avoid noxious stimuli (30+ days)'],\n",
    "                    \"8: evidence that the animal values a putative analgesic/anaesthetic when injured\": ['self-administers analgesics', 'self-administers recreational drugs'], \n",
    "                    \"9: evidence of affective states\": [\"anhedonia behavior\", \"learned helplessness behavior\", 'displacement behavior', 'adverse effects of social isolation', \\\n",
    "                                                        'stereotypic behavior', 'fear/anxiety behavior', 'play behavior']}\n",
    "\n",
    "class BirchModelTests(unittest.TestCase):\n",
    "    def test_proxies(self):\n",
    "        birch_proxies = pd.read_csv(os.path.join('input_data', 'Birch Model Proxies.csv'))\n",
    "        pass_test = True\n",
    "        for criterion in birch_proxies.columns:\n",
    "            criterion_proxies = birch_proxies[criterion].dropna().values.tolist()\n",
    "            if set(criterion_proxies) != set(birch_proxies_dict[criterion]):\n",
    "                pass_test = False\n",
    "                print(\"Failed at {}\".format(criterion))\n",
    "                print(\"Hard-coded proxies:\", birch_proxies_dict[criterion])\n",
    "                print(\"Downloaded proxies:\", criterion_proxies)\n",
    "        self.assertTrue(pass_test)\n",
    "\n",
    "    def test_one_species_birch_sum(self):\n",
    "        pass_test = True\n",
    "        birch_sums_2 = {}\n",
    "        \n",
    "        for x in [0,1]:\n",
    "            test_all_scores = {}\n",
    "            for proxy in ss_proxies:\n",
    "                if proxy in hv1_proxies:\n",
    "                    test_all_scores[proxy] = [x*HC_WEIGHT]*NUM_SCENARIOS\n",
    "                else:\n",
    "                    test_all_scores[proxy] = [x]*NUM_SCENARIOS\n",
    "\n",
    "            for criterion, lst in birch_proxies_dict.items():\n",
    "                test_data = {}\n",
    "                for proxy in lst:\n",
    "                    if proxy in hv1_proxies:\n",
    "                        test_data[proxy] = [x*HC_WEIGHT]*NUM_SCENARIOS\n",
    "                    else:\n",
    "                        test_data[proxy] = [x]*NUM_SCENARIOS\n",
    "                birch_sums = one_species_birch_sum(birch_proxies, test_all_scores)\n",
    "                prediction = sum([test_data[proxy][0] for proxy in test_data.keys()])\n",
    "                for i in range(NUM_SCENARIOS):\n",
    "                    p = birch_sums[criterion][i]\n",
    "                    if p != prediction:\n",
    "                        pass_test = False\n",
    "                        print(\"Failed on: \")\n",
    "                        print(\"x = {}\".format(x))\n",
    "                        print(\"Criterion: \", criterion)\n",
    "                        print(\"prediction: \", prediction)\n",
    "                        print(\"p:\", p)\n",
    "\n",
    "        self.assertTrue(pass_test)\n",
    "\n",
    "    def test_one_species_birch_est(self):\n",
    "        pass_test = True\n",
    "        for x in [0,1]:\n",
    "            human_all_scores = {}\n",
    "            species_all_scores = {}\n",
    "            for proxy in ss_proxies:\n",
    "                if proxy in hv1_proxies:\n",
    "                    human_all_scores[proxy] = [1*HC_WEIGHT]*NUM_SCENARIOS\n",
    "                    species_all_scores[proxy] = [x*HC_WEIGHT]*NUM_SCENARIOS\n",
    "                else:\n",
    "                    human_all_scores[proxy] = [1]*NUM_SCENARIOS\n",
    "                    species_all_scores[proxy] = [x]*NUM_SCENARIOS\n",
    "            species_sum = 0\n",
    "            human_sum = 0\n",
    "            for lst in birch_proxies_dict.values():\n",
    "                for proxy in lst:\n",
    "                    species_sum += species_all_scores[proxy][0]\n",
    "                    human_sum += human_all_scores[proxy][0]\n",
    "            prediction = species_sum/human_sum\n",
    "            # function\n",
    "            human_birch_sums = one_species_birch_sum(birch_proxies, human_all_scores)\n",
    "            species_birch_sums = one_species_birch_sum(birch_proxies, species_all_scores)\n",
    "\n",
    "            overall_scores = []\n",
    "            for i in range(NUM_SCENARIOS):\n",
    "                score_i = 0\n",
    "                hum_score_i = 0\n",
    "                for c in species_birch_sums:\n",
    "                    score_i += species_birch_sums[c][i]\n",
    "                    hum_score_i += human_birch_sums[c][i]\n",
    "                overall_scores.append(score_i/hum_score_i)\n",
    "\n",
    "            for i in range(NUM_SCENARIOS):\n",
    "                if overall_scores[i] != prediction:\n",
    "                    pass_test = False\n",
    "                    print(\"Failed at:\")\n",
    "                    print(\"x: \", x)\n",
    "                    print(\"prediction: \", prediction)\n",
    "                    print(\"p: \", overall_scores[i])\n",
    "        self.assertTrue(pass_test)\n",
    "\n",
    "\n",
    "res2 = unittest.main(argv=[''], verbosity=3, exit=False)\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  },
  "vscode": {
   "interpreter": {
    "hash": "25831d51e312d3fc9beb241e1c1dc2cc36c33e705bc782eae468a79335f361cf"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}