metrics.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "from utils import *\n",
    "import mrf\n",
    "import llm\n",
    "from importlib import reload\n",
    "from evaluate import *\n",
    "import utils\n",
    "import evaluate\n",
    "reload(evaluate)\n",
    "import json\n",
    "import os\n",
    "from tabulate import tabulate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# list all files in all subdirectories\n",
    "def list_files(directory):\n",
    "    for root, dirs, files in os.walk(directory):\n",
    "        for file in files:\n",
    "            if file.endswith(\".json\"):\n",
    "                yield os.path.join(root, file)\n",
    "\n",
    "k = [i for i in list_files('./Experiments/trial/bin0')]\n",
    "k"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "def parse_results(path):\n",
    "    js = json.load(open(path))\n",
    "    ground = js['ground']\n",
    "    cot = []\n",
    "    zero = []\n",
    "    mrf = []\n",
    "    question = js['question']['Text']\n",
    "    for key in js.keys():\n",
    "        if key in ['question', 'ground']:\n",
    "            continue\n",
    "        cot.append(js[key]['cot']['probability'])\n",
    "        zero.append(js[key]['zero']['probability'])\n",
    "        mrf.append(js[key]['mrf']['probability'])\n",
    "        query = js[key]['mrf']['query']\n",
    "    return ground, cot, zero, mrf, question, query\n",
    "        \n",
    "def metrics(ground, res):\n",
    "    return evaluate.total_variation_distance(ground, res).round(3)\n",
    "\n",
    "def metrics2(ground, res):\n",
    "    return evaluate.kl_divergence(ground, res).round(3)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "for j in list_files('./Experiments/constraints/bin1'):\n",
    "    ground, cot, zero, mrf, question, query = parse_results(j)\n",
    "    if len(cot) == 0:\n",
    "        continue\n",
    "    print(j, question)\n",
    "    print(query)\n",
    "    cot = np.array(cot[0])\n",
    "    zero = np.array(zero[0])\n",
    "    mrf = np.array(mrf[0])\n",
    "    ground = np.array(ground)\n",
    "    table = [\n",
    "        [\"Method\", \"total_variation_distance\", \"dist\"],\n",
    "        [\"CoT\", metrics(ground, cot), cot.round(2)],\n",
    "        [\"mrf\", metrics(ground, mrf), mrf.round(2)],\n",
    "        [\"ground\", metrics(ground, ground), ground.round(2)],\n",
    "        [\"zero\", metrics(ground, zero), zero.round(2)],\n",
    "    ]\n",
    "    table2 = [\n",
    "        [\"Method\", \"kl_divergence\", \"dist\"],\n",
    "        [\"CoT\", metrics2(ground, cot), cot.round(2)],\n",
    "        [\"mrf\", metrics2(ground, mrf), mrf.round(2)],\n",
    "        [\"ground\", metrics2(ground, ground), ground.round(2)],\n",
    "        [\"zero\", metrics2(ground, zero), zero.round(2)],\n",
    "    ]\n",
    "    print(tabulate(table))\n",
    "    print(\"#################################################\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "for j in list_files('./Experiments/trial/bin0'):\n",
    "    ground, cot, zero, mrf, question = parse_results(j)\n",
    "    print(j, question)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "What is the nightly price of an Airbnb listing in Rochester, NY with 2 bedrooms?\n",
      "What is the price of an Airbnb listing in Newark, NJ for the months of May to August?\n",
      "What is the price of an Airbnb listing with 2 beds in Columbus, OH?\n",
      "What is the nightly price of an Airbnb listing in Broward County, FL that accommodates 2 guests?\n",
      "What is the price of an Airbnb listing in San Diego, CA with 0 to 1 bathroom?\n",
      "What is the price of an Airbnb listing in Cambridge, MA for a host who is not a superhost?\n",
      "What is the price of an Airbnb listing hosted by a superhost in Santa Clara County, CA?\n",
      "What is the price of an Airbnb listing in Clark County, NV with no review scores available?\n",
      "What is the price of an Airbnb listing in Dallas, TX with a review scores rating of 4.9 or higher?\n",
      "What is the price of an Airbnb listing in Los Angeles, CA that accommodates 4 or more guests per night?\n",
      "What is the price of an Airbnb listing in Salem, OR, hosted by someone who has been hosting for 1-5 years?\n",
      "What is the nightly price of an Airbnb listing in San Mateo County, CA with an unspecified number of beds?\n",
      "What is the price of an Airbnb listing in Denver, CO, hosted by someone who has been hosting for 1-5 years?\n",
      "What is the price of an Airbnb listing in Jersey City, NJ with an unspecified number of bedrooms?\n",
      "What is the price of an Airbnb listing in Fort Worth, TX with a review score rating between 4.3 and 4.8?\n",
      "What is the nightly price of an Airbnb listing in Oakland, CA that has 4 or more beds?\n",
      "What is the price of an Airbnb listing in Nashville, TN that accommodates 1 guest per night?\n"
     ]
    }
   ],
   "source": [
    "for j in list_files('./Dataset/bin0'):\n",
    "    if j.endswith('q.json'):\n",
    "        file = json.load(open(j))\n",
    "        if 'Text' in file.keys():\n",
    "            print(file['Text'])\n",
    "        else:\n",
    "            print('no text')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "k = [i for i in list_files('./Experiments/trial/bin0')]\n",
    "b = ['./Experiments/trial/bin0/LosAngeles/LosAngeles_q.json']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./Experiments/trial/bin0/LosAngeles/LosAngeles_q.json\n"
     ]
    },
    {
     "ename": "FileNotFoundError",
     "evalue": "[Errno 2] No such file or directory: './Experiments/trial/bin0/LosAngeles/LosAngeles_q.json'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[41], line 10\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m j \u001b[38;5;129;01min\u001b[39;00m b[:\u001b[38;5;241m1\u001b[39m]:\n\u001b[1;32m      9\u001b[0m     \u001b[38;5;28mprint\u001b[39m(j)\n\u001b[0;32m---> 10\u001b[0m     js \u001b[38;5;241m=\u001b[39m json\u001b[38;5;241m.\u001b[39mload(\u001b[38;5;28;43mopen\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mj\u001b[49m\u001b[43m)\u001b[49m)\n\u001b[1;32m     11\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m key \u001b[38;5;129;01min\u001b[39;00m js\u001b[38;5;241m.\u001b[39mkeys():\n\u001b[1;32m     12\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m key \u001b[38;5;129;01min\u001b[39;00m [\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mquestion\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mground\u001b[39m\u001b[38;5;124m'\u001b[39m]:\n",
      "File \u001b[0;32m/opt/homebrew/Caskroom/miniconda/base/envs/dl/lib/python3.8/site-packages/IPython/core/interactiveshell.py:284\u001b[0m, in \u001b[0;36m_modified_open\u001b[0;34m(file, *args, **kwargs)\u001b[0m\n\u001b[1;32m    277\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m file \u001b[38;5;129;01min\u001b[39;00m {\u001b[38;5;241m0\u001b[39m, \u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m2\u001b[39m}:\n\u001b[1;32m    278\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[1;32m    279\u001b[0m         \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mIPython won\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mt let you open fd=\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mfile\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m by default \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    280\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mas it is likely to crash IPython. If you know what you are doing, \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    281\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124myou can use builtins\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m open.\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    282\u001b[0m     )\n\u001b[0;32m--> 284\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mio_open\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfile\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
      "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: './Experiments/trial/bin0/LosAngeles/LosAngeles_q.json'"
     ]
    }
   ],
   "source": [
    "import mrf\n",
    "import importlib\n",
    "importlib.reload(mrf)\n",
    "\n",
    "k = [i for i in list_files('./Experiments/trial/bin0')]\n",
    "\n",
    "\n",
    "for j in b[:1]:\n",
    "    print(j)\n",
    "    js = json.load(open(j))\n",
    "    for key in js.keys():\n",
    "        if key in ['question', 'ground']:\n",
    "            continue\n",
    "        #print(key)\n",
    "        query = js[key]['mrf']['query']\n",
    "        query = json.loads(query)\n",
    "        query['Queries'][0]['Condition'] = [query['Queries'][0]['Condition'][0]]\n",
    "        print(query)\n",
    "        test = mrf.Brute(verbose=True)\n",
    "        test.from_json(vars=js[key]['mrf']['json']['Variables'], constraints=pp)\n",
    "        #test.from_json(pp)\n",
    "        test.update(10000)\n",
    "        test.query(query=query)\n",
    "        break\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ".DS_Store\n",
      "BrowardCounty\n",
      "LosAngeles\n"
     ]
    }
   ],
   "source": [
    "reload(utils)\n",
    "p = './Experiments/35/'\n",
    "\n",
    "for j in os.listdir(p):\n",
    "    # check if is folder\n",
    "    if os.path.isdir(p+j):\n",
    "        dat = None\n",
    "        for file in os.listdir(p+j):\n",
    "            if file.endswith('.csv'):\n",
    "                dat = utils.DatasetQ(p+j+'/'+file)\n",
    "            elif file.endswith('.json'):\n",
    "                constraints = json.load(open(p+j+'/'+file))\n",
    "        table = [[\"Constraint\", \"Ground\", \"Zero\", \"Joint\", \"Change\", \"Zero dist\", \"Joint dist\"],]\n",
    "        tot_change = 0\n",
    "        #print(constraints['Variables'])\n",
    "        for cons in tmpp['Constraints']:\n",
    "            er = False\n",
    "            ground = dat.marg(cons)\n",
    "            pred = cons['Probability']\n",
    "            if np.isnan(ground).any():\n",
    "                print(\"Error: nan in ground\", cons['Question'])\n",
    "                print(\"#################################################\")\n",
    "                continue\n",
    "            query = {}\n",
    "            query['Queries'] = [{'Condition': cons['Condition'], 'Target': cons['Target']}]\n",
    "            after = mrfff.query(query)\n",
    "            ground = np.array(ground)\n",
    "            zero = np.array(pred)\n",
    "            after = np.array(after)\n",
    "            tar_name = cons['Target'][0]['Name']\n",
    "            cons_val = '[]'\n",
    "            if len(cons['Condition']) > 0:\n",
    "                cons_val = cons['Condition'][0]['Value']\n",
    "            try:\n",
    "                zero_metric = metrics2(ground, zero)\n",
    "            except:\n",
    "                er = True\n",
    "                zero_metric = float('nan')\n",
    "            after_metric = metrics2(ground, after)\n",
    "            table.append([f\"{tar_name}|{cons_val}\", str(ground.round(2)), str(zero.round(2)), str(after.round(2)), f\"{(after_metric-zero_metric):+.3f}\", zero_metric, after_metric])\n",
    "            if not er:\n",
    "                tot_change += after_metric-zero_metric\n",
    "        print(tabulate(table))\n",
    "        print(\"#################################################\")\n",
    "        print(f\"Total change: {tot_change}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "ename": "FileNotFoundError",
     "evalue": "[Errno 2] No such file or directory: './Experiments/35/BrowardCounty'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[13], line 3\u001b[0m\n\u001b[1;32m      1\u001b[0m dat \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m      2\u001b[0m pa \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124m./Experiments/35/BrowardCounty\u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[0;32m----> 3\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m file \u001b[38;5;129;01min\u001b[39;00m \u001b[43mos\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mlistdir\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpa\u001b[49m\u001b[43m)\u001b[49m:\n\u001b[1;32m      4\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m file\u001b[38;5;241m.\u001b[39mendswith(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m.csv\u001b[39m\u001b[38;5;124m'\u001b[39m):\n\u001b[1;32m      5\u001b[0m         dat \u001b[38;5;241m=\u001b[39m utils\u001b[38;5;241m.\u001b[39mDatasetQ(pa\u001b[38;5;241m+\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m/\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;241m+\u001b[39mfile)\n",
      "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: './Experiments/35/BrowardCounty'"
     ]
    }
   ],
   "source": [
    "dat = None\n",
    "pa = './Experiments/35/BrowardCounty'\n",
    "for file in os.listdir(pa):\n",
    "    if file.endswith('.csv'):\n",
    "        dat = utils.DatasetQ(pa+'/'+file)\n",
    "    elif file.endswith('.json'):\n",
    "        constraints = json.load(open(pa+'/'+file))\n",
    "table = [[\"Constraint\", \"Ground\", \"Zero\", \"Joint\", \"Change\", \"Zero dist\", \"Joint dist\"],]\n",
    "tot_change = 0\n",
    "#con = remove_vars(constraints, ['Month'])\n",
    "#print(constraints['Variables'])\n",
    "for cons in constraints['Constraints']:\n",
    "    er = False\n",
    "    ground = dat.marg(cons)\n",
    "    pred = cons['Probability']\n",
    "    if np.isnan(ground).any():\n",
    "        print(\"Error: nan in ground\", cons)\n",
    "        print(\"#################################################\")\n",
    "        continue\n",
    "    query = {}\n",
    "    query['Queries'] = [{'Condition': cons['Condition'], 'Target': cons['Target']}]\n",
    "    after = model.query(query)\n",
    "    ground = np.array(ground)\n",
    "    zero = np.array(pred)\n",
    "    after = np.array(after)\n",
    "    tar_name = cons['Target'][0]['Name']\n",
    "    cons_val = '[]'\n",
    "    if len(cons['Condition']) > 0:\n",
    "        cons_val = cons['Condition'][0]['Value']\n",
    "    try:\n",
    "        zero_metric = metrics(ground, zero)\n",
    "    except:\n",
    "        er = True\n",
    "        zero_metric = float('nan')\n",
    "    after_metric = metrics(ground, after)\n",
    "    table.append([f\"{tar_name}|{cons_val}\", str(ground.round(2)), str(zero.round(2)), str(after.round(2)), f\"{(after_metric-zero_metric):+.3f}\", zero_metric, after_metric])\n",
    "    if not er:\n",
    "        tot_change += after_metric-zero_metric\n",
    "print(tabulate(table))\n",
    "print(\"#################################################\")\n",
    "print(f\"Total change: {tot_change}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-----------------------------------------------\n",
      "Finished building\n",
      "Features:  113\n",
      "Constraints:  112\n",
      "-----------------------------------------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress:   0%|          | 10/10000 [00:00<01:43, 96.93it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Entropy:  -2.23779559135437\n",
      "Loss:  tensor(7.3527, grad_fn=<AddBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.7935, 0.7415, 0.6387, 0.6348, 0.5360], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([27, 41, 11, 94, 78]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress:  20%|██        | 2005/10000 [00:20<01:15, 106.44it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Entropy:  -3.4043216705322266\n",
      "Loss:  tensor(-3.1953, grad_fn=<AddBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.1373, 0.1026, 0.0898, 0.0892, 0.0743], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([  0,  23, 100,   4,  22]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress:  40%|████      | 4016/10000 [00:39<00:54, 109.16it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Entropy:  -3.404226064682007\n",
      "Loss:  tensor(-3.1953, grad_fn=<AddBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.1373, 0.1026, 0.0898, 0.0891, 0.0744], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([  0,  23, 100,   4,  22]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress:  60%|██████    | 6015/10000 [00:58<00:37, 107.23it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Entropy:  -3.4042255878448486\n",
      "Loss:  tensor(-3.1953, grad_fn=<AddBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.1373, 0.1026, 0.0898, 0.0891, 0.0744], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([  0,  23, 100,   4,  22]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress:  80%|████████  | 8022/10000 [01:17<00:20, 95.97it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Entropy:  -3.4042255878448486\n",
      "Loss:  tensor(-3.1953, grad_fn=<AddBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.1373, 0.1026, 0.0898, 0.0891, 0.0744], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([  0,  23, 100,   4,  22]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress: 100%|██████████| 10000/10000 [01:36<00:00, 104.10it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Entropy:  -3.4042255878448486\n",
      "Loss:  tensor(-3.1953, grad_fn=<AddBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.1373, 0.1026, 0.0898, 0.0891, 0.0744], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([  0,  23, 100,   4,  22]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "dat = None\n",
    "pa = './Experiments3/LosAngeles'\n",
    "dat = utils.DatasetQ('./data/LosAngeles.csv')\n",
    "for file in os.listdir(pa):\n",
    "    if file.endswith('.csv'):\n",
    "        dat = utils.DatasetQ(pa+'/'+file)\n",
    "    elif file.endswith('agents2.json'):\n",
    "        constraints = json.load(open(pa+'/'+file))\n",
    "\n",
    "\n",
    "import mrf\n",
    "reload(mrf)\n",
    "\n",
    "model = mrf.Brute(verbose=True)\n",
    "ret = model.from_json(constraints['schema'])\n",
    "model.set_w0(0.4)\n",
    "model.update(10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dat = utils.DatasetQ('./data/LosAngeles.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "prob_file = json.load(open('./Experiments3/LosAngeles/prob_cache.json'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "var_dict = {}\n",
    "for var in constraints['schema']['Variables']:\n",
    "    var_dict[var['Name']] = var['Value']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-----------------------------------------------\n",
      "Finished building\n",
      "Features:  113\n",
      "Constraints:  112\n",
      "-----------------------------------------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress:   0%|          | 12/10000 [00:00<01:27, 114.42it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  tensor(10.1424, grad_fn=<SumBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.8624, 0.7190, 0.6365, 0.6014, 0.5197], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([108,  77,   4,  74,  27]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress:  20%|██        | 2013/10000 [00:18<01:12, 110.84it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  tensor(0.0103, grad_fn=<SumBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.0474, 0.0163, 0.0150, 0.0138, 0.0138], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([24, 81, 21, 22, 23]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress:  40%|████      | 4020/10000 [00:35<00:51, 115.63it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  tensor(0.0082, grad_fn=<SumBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.0456, 0.0169, 0.0165, 0.0128, 0.0125], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([24, 96, 21, 22, 23]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress:  60%|██████    | 6020/10000 [00:52<00:31, 127.57it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  tensor(0.0077, grad_fn=<SumBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.0449, 0.0188, 0.0170, 0.0124, 0.0120], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([24, 96, 21, 22, 23]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress:  80%|████████  | 8016/10000 [01:09<00:17, 111.16it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  tensor(0.0076, grad_fn=<SumBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.0446, 0.0197, 0.0172, 0.0123, 0.0117], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([24, 96, 21, 22, 23]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Progress: 100%|██████████| 10000/10000 [01:26<00:00, 115.97it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  tensor(0.0075, grad_fn=<SumBackward0>) torch.return_types.topk(\n",
      "values=tensor([0.0444, 0.0202, 0.0173, 0.0123, 0.0115], grad_fn=<TopkBackward0>),\n",
      "indices=tensor([24, 96, 21, 22, 23]))\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "modell = mrf.Brute(verbose=True)\n",
    "ret = modell.from_json(schema)\n",
    "modell.update(10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Question': {'Target': [{'Name': 'Price'}],\n",
       "  'Text': 'What is the nightly price of an Airbnb listing in Los Angeles, CA?'},\n",
       " 'Variables': [{'Name': 'Number of Bedrooms',\n",
       "   'Value': ['0/1 bedroom', '2 bedrooms', '3 bedrooms', '4+ bedrooms']},\n",
       "  {'Name': 'Number of Bathrooms',\n",
       "   'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']},\n",
       "  {'Name': 'Host Experience',\n",
       "   'Value': ['<1 year', '1-5 years', '5-10 years', '10+ years']},\n",
       "  {'Name': 'Superhost', 'Value': ['Yes', 'No']},\n",
       "  {'Name': 'Room Type',\n",
       "   'Value': ['Entire home/apt', 'Private room', 'Shared/Hotel room']},\n",
       "  {'Name': 'Total Beds', 'Value': ['0/1 bed', '2 beds', '3 beds', '4+ beds']},\n",
       "  {'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']},\n",
       "  {'Name': 'Price',\n",
       "   'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       " 'Constraints': [{'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "   'Condition': [],\n",
       "   'Probability': [0.3297555944642986, 0.6702444055357014]},\n",
       "  {'Target': [{'Name': 'Room Type',\n",
       "     'Value': ['Entire home/apt', 'Private room', 'Shared/Hotel room']}],\n",
       "   'Condition': [],\n",
       "   'Probability': [0.7294291944800493,\n",
       "    0.2522209090051548,\n",
       "    0.018349896514795887]},\n",
       "  {'Target': [{'Name': 'Number of Bathrooms',\n",
       "     'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "   'Condition': [],\n",
       "   'Probability': [0.6677771115108713,\n",
       "    0.22307171252803873,\n",
       "    0.10915117596109002]},\n",
       "  {'Target': [{'Name': 'Host Experience',\n",
       "     'Value': ['<1 year', '1-5 years', '5-10 years', '10+ years']}],\n",
       "   'Condition': [],\n",
       "   'Probability': [0.07762615359856158,\n",
       "    0.26707436860769046,\n",
       "    0.5176390283966992,\n",
       "    0.13766044939704877]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Condition': [],\n",
       "   'Probability': [0.0687830380144335,\n",
       "    0.2384261547676508,\n",
       "    0.3571391187536709,\n",
       "    0.24099650502148742,\n",
       "    0.09465518344275738]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Condition': [],\n",
       "   'Probability': [0.08130983447230875,\n",
       "    0.4435193192166706,\n",
       "    0.47517084631102063]},\n",
       "  {'Target': [{'Name': 'Total Beds',\n",
       "     'Value': ['0/1 bed', '2 beds', '3 beds', '4+ beds']}],\n",
       "   'Condition': [],\n",
       "   'Probability': [0.45769291759413155,\n",
       "    0.24386992194611992,\n",
       "    0.12577892369043914,\n",
       "    0.1726582367693094]},\n",
       "  {'Target': [{'Name': 'Number of Bedrooms',\n",
       "     'Value': ['0/1 bedroom', '2 bedrooms', '3 bedrooms', '4+ bedrooms']}],\n",
       "   'Condition': [],\n",
       "   'Probability': [0.48824558182679023,\n",
       "    0.2442773907414028,\n",
       "    0.14744958596233562,\n",
       "    0.12002744146947136]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Condition': [{'Name': 'Superhost', 'Value': ['Yes'], 'Other': ['No']}],\n",
       "   'Probability': [0.012179856543940869,\n",
       "    0.41405551471318724,\n",
       "    0.5737646287428719]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Condition': [{'Name': 'Superhost', 'Value': ['No'], 'Other': ['Yes']}],\n",
       "   'Probability': [0.12889075748020462,\n",
       "    0.462378249597085,\n",
       "    0.40873099292271037]},\n",
       "  {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "   'Condition': [{'Name': 'Host Experience',\n",
       "     'Value': ['<1 year'],\n",
       "     'Other': ['1-5 years', '5-10 years', '10+ years']}],\n",
       "   'Probability': [0.17798954810943743, 0.8220104518905625]},\n",
       "  {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "   'Condition': [{'Name': 'Host Experience',\n",
       "     'Value': ['1-5 years'],\n",
       "     'Other': ['<1 year', '5-10 years', '10+ years']}],\n",
       "   'Probability': [0.30046522183138547, 0.6995347781686145]},\n",
       "  {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "   'Condition': [{'Name': 'Host Experience',\n",
       "     'Value': ['5-10 years'],\n",
       "     'Other': ['<1 year', '1-5 years', '10+ years']}],\n",
       "   'Probability': [0.3409640515276938, 0.6590359484723062]},\n",
       "  {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "   'Condition': [{'Name': 'Host Experience',\n",
       "     'Value': ['10+ years'],\n",
       "     'Other': ['<1 year', '1-5 years', '5-10 years']}],\n",
       "   'Probability': [0.4271828247823514, 0.5728171752176485]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Condition': [{'Name': 'Room Type',\n",
       "     'Value': ['Entire home/apt'],\n",
       "     'Other': ['Private room', 'Shared/Hotel room']}],\n",
       "   'Probability': [0.005870176658344658,\n",
       "    0.14037069810068578,\n",
       "    0.4265793382422653,\n",
       "    0.3070031286540467,\n",
       "    0.12017665834465759]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Condition': [{'Name': 'Room Type',\n",
       "     'Value': ['Private room'],\n",
       "     'Other': ['Entire home/apt', 'Shared/Hotel room']}],\n",
       "   'Probability': [0.21448789923102327,\n",
       "    0.534320894466198,\n",
       "    0.17011841298296929,\n",
       "    0.05790913808557815,\n",
       "    0.023163655234231258]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Condition': [{'Name': 'Room Type',\n",
       "     'Value': ['Shared/Hotel room'],\n",
       "     'Other': ['Entire home/apt', 'Private room']}],\n",
       "   'Probability': [0.6651085141903172,\n",
       "    0.19398998330550918,\n",
       "    0.06978297161936561,\n",
       "    0.04307178631051753,\n",
       "    0.028046744574290485]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Condition': [{'Name': 'Review Score',\n",
       "     'Value': ['<4.3'],\n",
       "     'Other': ['4.3-4.8', '4.9+']}],\n",
       "   'Probability': [0.12790697674418605,\n",
       "    0.2738024249652157,\n",
       "    0.34873782548201154,\n",
       "    0.19141323792486584,\n",
       "    0.05813953488372093]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Condition': [{'Name': 'Review Score',\n",
       "     'Value': ['4.3-4.8'],\n",
       "     'Other': ['<4.3', '4.9+']}],\n",
       "   'Probability': [0.07017481014471988,\n",
       "    0.26653173807135694,\n",
       "    0.37324473420260784,\n",
       "    0.23611907150021494,\n",
       "    0.053929646081100444]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Condition': [{'Name': 'Review Score',\n",
       "     'Value': ['4.9+'],\n",
       "     'Other': ['<4.3', '4.3-4.8']}],\n",
       "   'Probability': [0.05750830899385638,\n",
       "    0.2204317319636083,\n",
       "    0.3560445832074395,\n",
       "    0.2697653338704804,\n",
       "    0.09625004196461544]},\n",
       "  {'Target': [{'Name': 'Number of Bathrooms',\n",
       "     'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms',\n",
       "     'Value': ['0/1 bedroom'],\n",
       "     'Other': ['2 bedrooms', '3 bedrooms', '4+ bedrooms']}],\n",
       "   'Probability': [0.9647128712871287,\n",
       "    0.027306930693069307,\n",
       "    0.00798019801980198]},\n",
       "  {'Target': [{'Name': 'Number of Bathrooms',\n",
       "     'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms',\n",
       "     'Value': ['2 bedrooms'],\n",
       "     'Other': ['0/1 bedroom', '3 bedrooms', '4+ bedrooms']}],\n",
       "   'Probability': [0.4799192814466031,\n",
       "    0.5082894788905156,\n",
       "    0.011791239662881335]},\n",
       "  {'Target': [{'Name': 'Number of Bathrooms',\n",
       "     'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms',\n",
       "     'Value': ['3 bedrooms'],\n",
       "     'Other': ['0/1 bedroom', '2 bedrooms', '4+ bedrooms']}],\n",
       "   'Probability': [0.12267365661861075,\n",
       "    0.6410878112712975,\n",
       "    0.23623853211009174]},\n",
       "  {'Target': [{'Name': 'Number of Bathrooms',\n",
       "     'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms',\n",
       "     'Value': ['4+ bedrooms'],\n",
       "     'Other': ['0/1 bedroom', '2 bedrooms', '3 bedrooms']}],\n",
       "   'Probability': [0.018515536950571565,\n",
       "    0.21276767026243762,\n",
       "    0.7687167927869908]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Condition': [{'Name': 'Host Experience',\n",
       "     'Value': ['<1 year'],\n",
       "     'Other': ['1-5 years', '5-10 years', '10+ years']}],\n",
       "   'Probability': [0.1431090214516603,\n",
       "    0.3385248310314428,\n",
       "    0.5183661475168968]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Condition': [{'Name': 'Host Experience',\n",
       "     'Value': ['1-5 years'],\n",
       "     'Other': ['<1 year', '5-10 years', '10+ years']}],\n",
       "   'Probability': [0.10893557508132924,\n",
       "    0.4518865343102976,\n",
       "    0.4391778906083731]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Condition': [{'Name': 'Host Experience',\n",
       "     'Value': ['5-10 years'],\n",
       "     'Other': ['<1 year', '1-5 years', '10+ years']}],\n",
       "   'Probability': [0.07303155474172805,\n",
       "    0.4490482476685262,\n",
       "    0.4779201975897458]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Condition': [{'Name': 'Host Experience',\n",
       "     'Value': ['10+ years'],\n",
       "     'Other': ['<1 year', '1-5 years', '5-10 years']}],\n",
       "   'Probability': [0.04484742127327679,\n",
       "    0.4461343472750317,\n",
       "    0.5090182314516916]},\n",
       "  {'Target': [{'Name': 'Total Beds',\n",
       "     'Value': ['0/1 bed', '2 beds', '3 beds', '4+ beds']}],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms',\n",
       "     'Value': ['0/1 bedroom'],\n",
       "     'Other': ['2 bedrooms', '3 bedrooms', '4+ bedrooms']}],\n",
       "   'Probability': [0.7318394536025971,\n",
       "    0.2271807411779586,\n",
       "    0.027825793667523926,\n",
       "    0.013154011551920402]},\n",
       "  {'Target': [{'Name': 'Total Beds',\n",
       "     'Value': ['0/1 bed', '2 beds', '3 beds', '4+ beds']}],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms',\n",
       "     'Value': ['2 bedrooms'],\n",
       "     'Other': ['0/1 bedroom', '3 bedrooms', '4+ bedrooms']}],\n",
       "   'Probability': [0.037118087703055296,\n",
       "    0.6065987711472098,\n",
       "    0.2536823499705412,\n",
       "    0.10260079117919367]},\n",
       "  {'Target': [{'Name': 'Total Beds',\n",
       "     'Value': ['0/1 bed', '2 beds', '3 beds', '4+ beds']}],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms',\n",
       "     'Value': ['3 bedrooms'],\n",
       "     'Other': ['0/1 bedroom', '2 bedrooms', '4+ bedrooms']}],\n",
       "   'Probability': [0.015863247863247863,\n",
       "    0.029675213675213676,\n",
       "    0.5184273504273504,\n",
       "    0.43603418803418803]},\n",
       "  {'Target': [{'Name': 'Total Beds',\n",
       "     'Value': ['0/1 bed', '2 beds', '3 beds', '4+ beds']}],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms',\n",
       "     'Value': ['4+ bedrooms'],\n",
       "     'Other': ['0/1 bedroom', '2 bedrooms', '3 bedrooms']}],\n",
       "   'Probability': [0.019288829251928884,\n",
       "    0.004696410600469641,\n",
       "    0.02331432405233143,\n",
       "    0.95270043609527]}]}"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "schema"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "schema = constraints['schema']\n",
    "schema['Constraints'] = golds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "  'Condition': [],\n",
       "  'Probability': [0.3297555944642986, 0.6702444055357014]},\n",
       " {'Target': [{'Name': 'Room Type',\n",
       "    'Value': ['Entire home/apt', 'Private room', 'Shared/Hotel room']}],\n",
       "  'Condition': [],\n",
       "  'Probability': [0.7294291944800493,\n",
       "   0.2522209090051548,\n",
       "   0.018349896514795887]},\n",
       " {'Target': [{'Name': 'Number of Bathrooms',\n",
       "    'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "  'Condition': [],\n",
       "  'Probability': [0.6677771115108713,\n",
       "   0.22307171252803873,\n",
       "   0.10915117596109002]},\n",
       " {'Target': [{'Name': 'Host Experience',\n",
       "    'Value': ['<1 year', '1-5 years', '5-10 years', '10+ years']}],\n",
       "  'Condition': [],\n",
       "  'Probability': [0.07762615359856158,\n",
       "   0.26707436860769046,\n",
       "   0.5176390283966992,\n",
       "   0.13766044939704877]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Condition': [],\n",
       "  'Probability': [0.0687830380144335,\n",
       "   0.2384261547676508,\n",
       "   0.3571391187536709,\n",
       "   0.24099650502148742,\n",
       "   0.09465518344275738]},\n",
       " {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "  'Condition': [],\n",
       "  'Probability': [0.08130983447230875,\n",
       "   0.4435193192166706,\n",
       "   0.47517084631102063]},\n",
       " {'Target': [{'Name': 'Total Beds',\n",
       "    'Value': ['0/1 bed', '2 beds', '3 beds', '4+ beds']}],\n",
       "  'Condition': [],\n",
       "  'Probability': [0.45769291759413155,\n",
       "   0.24386992194611992,\n",
       "   0.12577892369043914,\n",
       "   0.1726582367693094]},\n",
       " {'Target': [{'Name': 'Number of Bedrooms',\n",
       "    'Value': ['0/1 bedroom', '2 bedrooms', '3 bedrooms', '4+ bedrooms']}],\n",
       "  'Condition': [],\n",
       "  'Probability': [0.48824558182679023,\n",
       "   0.2442773907414028,\n",
       "   0.14744958596233562,\n",
       "   0.12002744146947136]},\n",
       " {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "  'Condition': [{'Name': 'Superhost', 'Value': ['Yes'], 'Other': ['No']}],\n",
       "  'Probability': [0.012179856543940869,\n",
       "   0.41405551471318724,\n",
       "   0.5737646287428719]},\n",
       " {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "  'Condition': [{'Name': 'Superhost', 'Value': ['No'], 'Other': ['Yes']}],\n",
       "  'Probability': [0.12889075748020462,\n",
       "   0.462378249597085,\n",
       "   0.40873099292271037]},\n",
       " {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "  'Condition': [{'Name': 'Host Experience',\n",
       "    'Value': ['<1 year'],\n",
       "    'Other': ['1-5 years', '5-10 years', '10+ years']}],\n",
       "  'Probability': [0.17798954810943743, 0.8220104518905625]},\n",
       " {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "  'Condition': [{'Name': 'Host Experience',\n",
       "    'Value': ['1-5 years'],\n",
       "    'Other': ['<1 year', '5-10 years', '10+ years']}],\n",
       "  'Probability': [0.30046522183138547, 0.6995347781686145]},\n",
       " {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "  'Condition': [{'Name': 'Host Experience',\n",
       "    'Value': ['5-10 years'],\n",
       "    'Other': ['<1 year', '1-5 years', '10+ years']}],\n",
       "  'Probability': [0.3409640515276938, 0.6590359484723062]},\n",
       " {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "  'Condition': [{'Name': 'Host Experience',\n",
       "    'Value': ['10+ years'],\n",
       "    'Other': ['<1 year', '1-5 years', '5-10 years']}],\n",
       "  'Probability': [0.4271828247823514, 0.5728171752176485]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Condition': [{'Name': 'Room Type',\n",
       "    'Value': ['Entire home/apt'],\n",
       "    'Other': ['Private room', 'Shared/Hotel room']}],\n",
       "  'Probability': [0.005870176658344658,\n",
       "   0.14037069810068578,\n",
       "   0.4265793382422653,\n",
       "   0.3070031286540467,\n",
       "   0.12017665834465759]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Condition': [{'Name': 'Room Type',\n",
       "    'Value': ['Private room'],\n",
       "    'Other': ['Entire home/apt', 'Shared/Hotel room']}],\n",
       "  'Probability': [0.21448789923102327,\n",
       "   0.534320894466198,\n",
       "   0.17011841298296929,\n",
       "   0.05790913808557815,\n",
       "   0.023163655234231258]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Condition': [{'Name': 'Room Type',\n",
       "    'Value': ['Shared/Hotel room'],\n",
       "    'Other': ['Entire home/apt', 'Private room']}],\n",
       "  'Probability': [0.6651085141903172,\n",
       "   0.19398998330550918,\n",
       "   0.06978297161936561,\n",
       "   0.04307178631051753,\n",
       "   0.028046744574290485]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Condition': [{'Name': 'Review Score',\n",
       "    'Value': ['<4.3'],\n",
       "    'Other': ['4.3-4.8', '4.9+']}],\n",
       "  'Probability': [0.12790697674418605,\n",
       "   0.2738024249652157,\n",
       "   0.34873782548201154,\n",
       "   0.19141323792486584,\n",
       "   0.05813953488372093]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Condition': [{'Name': 'Review Score',\n",
       "    'Value': ['4.3-4.8'],\n",
       "    'Other': ['<4.3', '4.9+']}],\n",
       "  'Probability': [0.07017481014471988,\n",
       "   0.26653173807135694,\n",
       "   0.37324473420260784,\n",
       "   0.23611907150021494,\n",
       "   0.053929646081100444]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Condition': [{'Name': 'Review Score',\n",
       "    'Value': ['4.9+'],\n",
       "    'Other': ['<4.3', '4.3-4.8']}],\n",
       "  'Probability': [0.05750830899385638,\n",
       "   0.2204317319636083,\n",
       "   0.3560445832074395,\n",
       "   0.2697653338704804,\n",
       "   0.09625004196461544]},\n",
       " {'Target': [{'Name': 'Number of Bathrooms',\n",
       "    'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms',\n",
       "    'Value': ['0/1 bedroom'],\n",
       "    'Other': ['2 bedrooms', '3 bedrooms', '4+ bedrooms']}],\n",
       "  'Probability': [0.9647128712871287,\n",
       "   0.027306930693069307,\n",
       "   0.00798019801980198]},\n",
       " {'Target': [{'Name': 'Number of Bathrooms',\n",
       "    'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms',\n",
       "    'Value': ['2 bedrooms'],\n",
       "    'Other': ['0/1 bedroom', '3 bedrooms', '4+ bedrooms']}],\n",
       "  'Probability': [0.4799192814466031,\n",
       "   0.5082894788905156,\n",
       "   0.011791239662881335]},\n",
       " {'Target': [{'Name': 'Number of Bathrooms',\n",
       "    'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms',\n",
       "    'Value': ['3 bedrooms'],\n",
       "    'Other': ['0/1 bedroom', '2 bedrooms', '4+ bedrooms']}],\n",
       "  'Probability': [0.12267365661861075,\n",
       "   0.6410878112712975,\n",
       "   0.23623853211009174]},\n",
       " {'Target': [{'Name': 'Number of Bathrooms',\n",
       "    'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms',\n",
       "    'Value': ['4+ bedrooms'],\n",
       "    'Other': ['0/1 bedroom', '2 bedrooms', '3 bedrooms']}],\n",
       "  'Probability': [0.018515536950571565,\n",
       "   0.21276767026243762,\n",
       "   0.7687167927869908]},\n",
       " {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "  'Condition': [{'Name': 'Host Experience',\n",
       "    'Value': ['<1 year'],\n",
       "    'Other': ['1-5 years', '5-10 years', '10+ years']}],\n",
       "  'Probability': [0.1431090214516603, 0.3385248310314428, 0.5183661475168968]},\n",
       " {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "  'Condition': [{'Name': 'Host Experience',\n",
       "    'Value': ['1-5 years'],\n",
       "    'Other': ['<1 year', '5-10 years', '10+ years']}],\n",
       "  'Probability': [0.10893557508132924,\n",
       "   0.4518865343102976,\n",
       "   0.4391778906083731]},\n",
       " {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "  'Condition': [{'Name': 'Host Experience',\n",
       "    'Value': ['5-10 years'],\n",
       "    'Other': ['<1 year', '1-5 years', '10+ years']}],\n",
       "  'Probability': [0.07303155474172805,\n",
       "   0.4490482476685262,\n",
       "   0.4779201975897458]},\n",
       " {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "  'Condition': [{'Name': 'Host Experience',\n",
       "    'Value': ['10+ years'],\n",
       "    'Other': ['<1 year', '1-5 years', '5-10 years']}],\n",
       "  'Probability': [0.04484742127327679,\n",
       "   0.4461343472750317,\n",
       "   0.5090182314516916]},\n",
       " {'Target': [{'Name': 'Total Beds',\n",
       "    'Value': ['0/1 bed', '2 beds', '3 beds', '4+ beds']}],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms',\n",
       "    'Value': ['0/1 bedroom'],\n",
       "    'Other': ['2 bedrooms', '3 bedrooms', '4+ bedrooms']}],\n",
       "  'Probability': [0.7318394536025971,\n",
       "   0.2271807411779586,\n",
       "   0.027825793667523926,\n",
       "   0.013154011551920402]},\n",
       " {'Target': [{'Name': 'Total Beds',\n",
       "    'Value': ['0/1 bed', '2 beds', '3 beds', '4+ beds']}],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms',\n",
       "    'Value': ['2 bedrooms'],\n",
       "    'Other': ['0/1 bedroom', '3 bedrooms', '4+ bedrooms']}],\n",
       "  'Probability': [0.037118087703055296,\n",
       "   0.6065987711472098,\n",
       "   0.2536823499705412,\n",
       "   0.10260079117919367]},\n",
       " {'Target': [{'Name': 'Total Beds',\n",
       "    'Value': ['0/1 bed', '2 beds', '3 beds', '4+ beds']}],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms',\n",
       "    'Value': ['3 bedrooms'],\n",
       "    'Other': ['0/1 bedroom', '2 bedrooms', '4+ bedrooms']}],\n",
       "  'Probability': [0.015863247863247863,\n",
       "   0.029675213675213676,\n",
       "   0.5184273504273504,\n",
       "   0.43603418803418803]},\n",
       " {'Target': [{'Name': 'Total Beds',\n",
       "    'Value': ['0/1 bed', '2 beds', '3 beds', '4+ beds']}],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms',\n",
       "    'Value': ['4+ bedrooms'],\n",
       "    'Other': ['0/1 bedroom', '2 bedrooms', '3 bedrooms']}],\n",
       "  'Probability': [0.019288829251928884,\n",
       "   0.004696410600469641,\n",
       "   0.02331432405233143,\n",
       "   0.95270043609527]}]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "golds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-----------------------------------  --------------------------  --------------------------  --------------------------  ------  ---------  ----------\n",
      "Constraint                           Ground                      Zero                        Joint                       Change  Zero dist  Joint dist\n",
      "Superhost|[]                         [0.33 0.67]                 [0.33 0.67]                 [0.29 0.71]                 +0.042  0.0        0.042\n",
      "Room Type|[]                         [0.73 0.25 0.02]            [0.73 0.25 0.01]            [0.51 0.3  0.19]            +0.219  0.004      0.223\n",
      "Number of Bathrooms|[]               [0.67 0.22 0.11]            [0.65 0.23 0.12]            [0.47 0.32 0.2 ]            +0.177  0.018      0.195\n",
      "Host Experience|[]                   [0.08 0.27 0.52 0.14]       [0.08 0.27 0.52 0.14]       [0.27 0.39 0.23 0.1 ]       +0.319  0.001      0.32\n",
      "Price|[]                             [0.07 0.24 0.36 0.24 0.09]  [0.07 0.24 0.36 0.24 0.08]  [0.08 0.18 0.34 0.28 0.12]  +0.066  0.01       0.076\n",
      "Review Score|[]                      [0.08 0.44 0.48]            [0.09 0.44 0.47]            [0.15 0.48 0.37]            +0.100  0.005      0.105\n",
      "Total Beds|[]                        [0.46 0.24 0.13 0.17]       [0.42 0.26 0.14 0.18]       [0.32 0.25 0.22 0.2 ]       +0.097  0.041      0.138\n",
      "Number of Bedrooms|[]                [0.49 0.24 0.15 0.12]       [0.53 0.24 0.13 0.1 ]       [0.37 0.33 0.17 0.13]       +0.076  0.044      0.12\n",
      "Review Score|['Yes']                 [0.01 0.41 0.57]            [0.01 0.41 0.58]            [0.07 0.27 0.66]            +0.143  0.004      0.147\n",
      "Review Score|['No']                  [0.13 0.46 0.41]            [0.12 0.46 0.42]            [0.18 0.57 0.25]            +0.145  0.01       0.155\n",
      "Superhost|['<1 year']                [0.18 0.82]                 [0.18 0.82]                 [0.15 0.85]                 +0.024  0.0        0.024\n",
      "Superhost|['1-5 years']              [0.3 0.7]                   [0.3 0.7]                   [0.24 0.76]                 +0.063  0.0        0.063\n",
      "Superhost|['5-10 years']             [0.34 0.66]                 [0.34 0.66]                 [0.33 0.67]                 +0.009  0.001      0.01\n",
      "Superhost|['10+ years']              [0.43 0.57]                 [0.43 0.57]                 [0.73 0.27]                 +0.300  0.0        0.3\n",
      "Price|['Entire home/apt']            [0.01 0.14 0.43 0.31 0.12]  [0.01 0.14 0.43 0.31 0.11]  [0.08 0.15 0.32 0.32 0.13]  +0.095  0.012      0.107\n",
      "Price|['Private room']               [0.21 0.53 0.17 0.06 0.02]  [0.21 0.54 0.17 0.06 0.02]  [0.07 0.18 0.37 0.25 0.12]  +0.487  0.005      0.492\n",
      "Price|['Shared/Hotel room']          [0.67 0.19 0.07 0.04 0.03]  [0.66 0.19 0.07 0.04 0.03]  [0.1  0.27 0.34 0.21 0.08]  +0.563  0.0        0.563\n",
      "Price|['<4.3']                       [0.13 0.27 0.35 0.19 0.06]  [0.13 0.27 0.35 0.19 0.06]  [0.1  0.29 0.39 0.16 0.07]  +0.059  0.002      0.061\n",
      "Price|['4.3-4.8']                    [0.07 0.27 0.37 0.24 0.05]  [0.07 0.26 0.37 0.23 0.06]  [0.08 0.16 0.33 0.31 0.11]  +0.136  0.012      0.148\n",
      "Price|['4.9+']                       [0.06 0.22 0.36 0.27 0.1 ]  [0.06 0.22 0.35 0.26 0.11]  [0.08 0.16 0.33 0.28 0.15]  +0.067  0.013      0.08\n",
      "Number of Bathrooms|['0/1 bedroom']  [0.96 0.03 0.01]            [0.97 0.02 0.01]            [0.77 0.15 0.09]            +0.191  0.008      0.199\n",
      "Number of Bathrooms|['2 bedrooms']   [0.48 0.51 0.01]            [0.48 0.5  0.01]            [0.48 0.38 0.15]            +0.130  0.004      0.134\n",
      "Number of Bathrooms|['3 bedrooms']   [0.12 0.64 0.24]            [0.13 0.64 0.24]            [0.13 0.66 0.21]            +0.027  0.002      0.029\n",
      "Number of Bathrooms|['4+ bedrooms']  [0.02 0.21 0.77]            [0.02 0.21 0.77]            [0.08 0.25 0.67]            +0.093  0.004      0.097\n",
      "Review Score|['<1 year']             [0.14 0.34 0.52]            [0.14 0.34 0.52]            [0.21 0.51 0.28]            +0.239  0.001      0.24\n",
      "Review Score|['1-5 years']           [0.11 0.45 0.44]            [0.11 0.45 0.44]            [0.14 0.51 0.35]            +0.084  0.003      0.087\n",
      "Review Score|['5-10 years']          [0.07 0.45 0.48]            [0.08 0.45 0.47]            [0.12 0.5  0.37]            +0.100  0.005      0.105\n",
      "Review Score|['10+ years']           [0.04 0.45 0.51]            [0.05 0.45 0.51]            [0.07 0.26 0.68]            +0.188  0.001      0.189\n",
      "Total Beds|['0/1 bedroom']           [0.73 0.23 0.03 0.01]       [0.75 0.22 0.02 0.01]       [0.66 0.19 0.09 0.06]       +0.089  0.021      0.11\n",
      "Total Beds|['2 bedrooms']            [0.04 0.61 0.25 0.1 ]       [0.05 0.6  0.25 0.1 ]       [0.14 0.45 0.26 0.14]       +0.145  0.01       0.155\n",
      "Total Beds|['3 bedrooms']            [0.02 0.03 0.52 0.44]       [0.02 0.03 0.52 0.43]       [0.1  0.12 0.46 0.32]       +0.170  0.006      0.176\n",
      "Total Beds|['4+ bedrooms']           [0.02 0.   0.02 0.95]       [0.02 0.01 0.02 0.95]       [0.09 0.11 0.19 0.61]       +0.341  0.006      0.347\n",
      "-----------------------------------  --------------------------  --------------------------  --------------------------  ------  ---------  ----------\n",
      "#################################################\n",
      "Average change: 0.15575\n"
     ]
    }
   ],
   "source": [
    "golds = []\n",
    "\n",
    "table = [[\"Constraint\", \"Ground\", \"Zero\", \"Joint\", \"Change\", \"Zero dist\", \"Joint dist\"],]\n",
    "tot_change = 0\n",
    "cnt = 0\n",
    "\n",
    "for i, cons in enumerate(constraints['schema']['Constraints']):\n",
    "    er = False\n",
    "    ground = dat.marg(cons, target_values=var_dict[cons['Target'][0]['Name']])\n",
    "    pred = cons['Probability']\n",
    "    if np.isnan(ground).any():\n",
    "        print(\"Error: nan in ground\", cons)\n",
    "        print(\"#################################################\")\n",
    "        continue\n",
    "    query = {}\n",
    "    query['Queries'] = [{'Target': cons['Target'], 'Condition': cons['Condition']}]\n",
    "    after = model.query(query)\n",
    "    zero = np.array(modell.query(query))\n",
    "\n",
    "    ground = np.array(ground)\n",
    "    #golds.append({'Target': cons['Target'], 'Condition': cons['Condition'], 'Probability': ground.tolist()})\n",
    "    #zero = np.array(j['Probability'])\n",
    "    after = np.array(after)\n",
    "    tar_name = cons['Target'][0]['Name']\n",
    "    cons_val = '[]'\n",
    "    if len(cons['Condition']) > 0:\n",
    "        cons_val = cons['Condition'][0]['Value']\n",
    "    try:\n",
    "        zero_metric = metrics(ground, zero)\n",
    "    except:\n",
    "        er = True\n",
    "        zero_metric = float('nan')\n",
    "    after_metric = metrics(ground, after)\n",
    "    table.append([f\"{tar_name}|{cons_val}\", str(ground.round(2)), str(zero.round(2)), str(after.round(2)), f\"{(after_metric-zero_metric):+.3f}\", zero_metric, after_metric])\n",
    "    if not er:\n",
    "        tot_change += after_metric-zero_metric\n",
    "        cnt += 1\n",
    "print(tabulate(table))\n",
    "print(\"#################################################\")\n",
    "print(f\"Average change: {tot_change/cnt}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-----------------------------------  --------------------------  --------------------------  --------------------------  ------  ---------  ----------\n",
      "Constraint                           Ground                      Zero                        Joint                       Change  Zero dist  Joint dist\n",
      "Superhost|[]                         [0.33 0.67]                 [0.25 0.75]                 [0.29 0.71]                 -0.038  0.08       0.042\n",
      "Superhost|[]                         [0.33 0.67]                 [0.2 0.8]                   [0.29 0.71]                 -0.088  0.13       0.042\n",
      "Superhost|[]                         [0.33 0.67]                 [0.18 0.82]                 [0.29 0.71]                 -0.108  0.15       0.042\n",
      "Superhost|[]                         [0.33 0.67]                 [0.25 0.75]                 [0.29 0.71]                 -0.038  0.08       0.042\n",
      "Superhost|[]                         [0.33 0.67]                 [0.15 0.85]                 [0.29 0.71]                 -0.138  0.18       0.042\n",
      "Room Type|[]                         [0.73 0.25 0.02]            [0.6 0.3 0.1]               [0.51 0.3  0.19]            +0.094  0.129      0.223\n",
      "Room Type|[]                         [0.73 0.25 0.02]            [0.6 0.3 0.1]               [0.51 0.3  0.19]            +0.094  0.129      0.223\n",
      "Room Type|[]                         [0.73 0.25 0.02]            [0.6 0.3 0.1]               [0.51 0.3  0.19]            +0.094  0.129      0.223\n",
      "Room Type|[]                         [0.73 0.25 0.02]            [0.65 0.25 0.1 ]            [0.51 0.3  0.19]            +0.141  0.082      0.223\n",
      "Room Type|[]                         [0.73 0.25 0.02]            [0.6 0.3 0.1]               [0.51 0.3  0.19]            +0.094  0.129      0.223\n",
      "Number of Bathrooms|[]               [0.67 0.22 0.11]            [0.6  0.25 0.15]            [0.47 0.32 0.2 ]            +0.127  0.068      0.195\n",
      "Number of Bathrooms|[]               [0.67 0.22 0.11]            [0.6 0.3 0.1]               [0.47 0.32 0.2 ]            +0.118  0.077      0.195\n",
      "Number of Bathrooms|[]               [0.67 0.22 0.11]            [0.5 0.3 0.2]               [0.47 0.32 0.2 ]            +0.027  0.168      0.195\n",
      "Number of Bathrooms|[]               [0.67 0.22 0.11]            [0.5 0.3 0.2]               [0.47 0.32 0.2 ]            +0.027  0.168      0.195\n",
      "Number of Bathrooms|[]               [0.67 0.22 0.11]            [0.5 0.3 0.2]               [0.47 0.32 0.2 ]            +0.027  0.168      0.195\n",
      "Host Experience|[]                   [0.08 0.27 0.52 0.14]       [0.25 0.4  0.25 0.1 ]       [0.27 0.39 0.23 0.1 ]       +0.015  0.305      0.32\n",
      "Host Experience|[]                   [0.08 0.27 0.52 0.14]       [0.2 0.5 0.2 0.1]           [0.27 0.39 0.23 0.1 ]       -0.035  0.355      0.32\n",
      "Host Experience|[]                   [0.08 0.27 0.52 0.14]       [0.25 0.45 0.2  0.1 ]       [0.27 0.39 0.23 0.1 ]       -0.035  0.355      0.32\n",
      "Host Experience|[]                   [0.08 0.27 0.52 0.14]       [0.25 0.4  0.2  0.15]       [0.27 0.39 0.23 0.1 ]       +0.002  0.318      0.32\n",
      "Host Experience|[]                   [0.08 0.27 0.52 0.14]       [0.25 0.45 0.2  0.1 ]       [0.27 0.39 0.23 0.1 ]       -0.035  0.355      0.32\n",
      "Price|[]                             [0.07 0.24 0.36 0.24 0.09]  [0.05 0.15 0.35 0.35 0.1 ]  [0.08 0.18 0.34 0.28 0.12]  -0.038  0.114      0.076\n",
      "Price|[]                             [0.07 0.24 0.36 0.24 0.09]  [0.05 0.15 0.35 0.35 0.1 ]  [0.08 0.18 0.34 0.28 0.12]  -0.038  0.114      0.076\n",
      "Price|[]                             [0.07 0.24 0.36 0.24 0.09]  [0.05 0.15 0.35 0.3  0.15]  [0.08 0.18 0.34 0.28 0.12]  -0.038  0.114      0.076\n",
      "Price|[]                             [0.07 0.24 0.36 0.24 0.09]  [0.05 0.15 0.35 0.3  0.15]  [0.08 0.18 0.34 0.28 0.12]  -0.038  0.114      0.076\n",
      "Price|[]                             [0.07 0.24 0.36 0.24 0.09]  [0.05 0.15 0.35 0.3  0.15]  [0.08 0.18 0.34 0.28 0.12]  -0.038  0.114      0.076\n",
      "Review Score|[]                      [0.08 0.44 0.48]            [0.1 0.4 0.5]               [0.15 0.48 0.37]            +0.061  0.044      0.105\n",
      "Review Score|[]                      [0.08 0.44 0.48]            [0.1 0.5 0.4]               [0.15 0.48 0.37]            +0.030  0.075      0.105\n",
      "Review Score|[]                      [0.08 0.44 0.48]            [0.1 0.6 0.3]               [0.15 0.48 0.37]            -0.070  0.175      0.105\n",
      "Review Score|[]                      [0.08 0.44 0.48]            [0.1 0.4 0.5]               [0.15 0.48 0.37]            +0.061  0.044      0.105\n",
      "Review Score|[]                      [0.08 0.44 0.48]            [0.1 0.6 0.3]               [0.15 0.48 0.37]            -0.070  0.175      0.105\n",
      "Total Beds|[]                        [0.46 0.24 0.13 0.17]       [0.4 0.3 0.2 0.1]           [0.32 0.25 0.22 0.2 ]       +0.008  0.13       0.138\n",
      "Total Beds|[]                        [0.46 0.24 0.13 0.17]       [0.4 0.3 0.2 0.1]           [0.32 0.25 0.22 0.2 ]       +0.008  0.13       0.138\n",
      "Total Beds|[]                        [0.46 0.24 0.13 0.17]       [0.5  0.3  0.15 0.05]       [0.32 0.25 0.22 0.2 ]       +0.015  0.123      0.138\n",
      "Total Beds|[]                        [0.46 0.24 0.13 0.17]       [0.5  0.3  0.15 0.05]       [0.32 0.25 0.22 0.2 ]       +0.015  0.123      0.138\n",
      "Total Beds|[]                        [0.46 0.24 0.13 0.17]       [0.5  0.25 0.15 0.1 ]       [0.32 0.25 0.22 0.2 ]       +0.065  0.073      0.138\n",
      "Number of Bedrooms|[]                [0.49 0.24 0.15 0.12]       [0.5  0.3  0.15 0.05]       [0.37 0.33 0.17 0.13]       +0.050  0.07       0.12\n",
      "Number of Bedrooms|[]                [0.49 0.24 0.15 0.12]       [0.5  0.3  0.15 0.05]       [0.37 0.33 0.17 0.13]       +0.050  0.07       0.12\n",
      "Number of Bedrooms|[]                [0.49 0.24 0.15 0.12]       [0.5  0.3  0.15 0.05]       [0.37 0.33 0.17 0.13]       +0.050  0.07       0.12\n",
      "Number of Bedrooms|[]                [0.49 0.24 0.15 0.12]       [0.5  0.3  0.15 0.05]       [0.37 0.33 0.17 0.13]       +0.050  0.07       0.12\n",
      "Number of Bedrooms|[]                [0.49 0.24 0.15 0.12]       [0.4  0.35 0.15 0.1 ]       [0.37 0.33 0.17 0.13]       +0.012  0.108      0.12\n",
      "Review Score|['Yes']                 [0.01 0.41 0.57]            [0.05 0.35 0.6 ]            [0.07 0.27 0.66]            +0.083  0.064      0.147\n",
      "Review Score|['Yes']                 [0.01 0.41 0.57]            [0.05 0.35 0.6 ]            [0.07 0.27 0.66]            +0.083  0.064      0.147\n",
      "Review Score|['Yes']                 [0.01 0.41 0.57]            [0.1 0.4 0.5]               [0.07 0.27 0.66]            +0.059  0.088      0.147\n",
      "Review Score|['Yes']                 [0.01 0.41 0.57]            [0.1 0.4 0.5]               [0.07 0.27 0.66]            +0.059  0.088      0.147\n",
      "Review Score|['Yes']                 [0.01 0.41 0.57]            [0.05 0.25 0.7 ]            [0.07 0.27 0.66]            -0.017  0.164      0.147\n",
      "Review Score|['No']                  [0.13 0.46 0.41]            [0.2 0.6 0.2]               [0.18 0.57 0.25]            -0.054  0.209      0.155\n",
      "Review Score|['No']                  [0.13 0.46 0.41]            [0.2 0.6 0.2]               [0.18 0.57 0.25]            -0.054  0.209      0.155\n",
      "Review Score|['No']                  [0.13 0.46 0.41]            [0.2 0.6 0.2]               [0.18 0.57 0.25]            -0.054  0.209      0.155\n",
      "Review Score|['No']                  [0.13 0.46 0.41]            [0.2 0.6 0.2]               [0.18 0.57 0.25]            -0.054  0.209      0.155\n",
      "Review Score|['No']                  [0.13 0.46 0.41]            [0.2 0.6 0.2]               [0.18 0.57 0.25]            -0.054  0.209      0.155\n",
      "Superhost|['<1 year']                [0.18 0.82]                 [0.15 0.85]                 [0.15 0.85]                 -0.004  0.028      0.024\n",
      "Superhost|['<1 year']                [0.18 0.82]                 [0.15 0.85]                 [0.15 0.85]                 -0.004  0.028      0.024\n",
      "Superhost|['<1 year']                [0.18 0.82]                 [0.1 0.9]                   [0.15 0.85]                 -0.054  0.078      0.024\n",
      "Superhost|['<1 year']                [0.18 0.82]                 [0.1 0.9]                   [0.15 0.85]                 -0.054  0.078      0.024\n",
      "Superhost|['<1 year']                [0.18 0.82]                 [0.15 0.85]                 [0.15 0.85]                 -0.004  0.028      0.024\n",
      "Superhost|['1-5 years']              [0.3 0.7]                   [0.25 0.75]                 [0.24 0.76]                 +0.013  0.05       0.063\n",
      "Superhost|['1-5 years']              [0.3 0.7]                   [0.3 0.7]                   [0.24 0.76]                 +0.063  0.0        0.063\n",
      "Superhost|['1-5 years']              [0.3 0.7]                   [0.25 0.75]                 [0.24 0.76]                 +0.013  0.05       0.063\n",
      "Superhost|['1-5 years']              [0.3 0.7]                   [0.3 0.7]                   [0.24 0.76]                 +0.063  0.0        0.063\n",
      "Superhost|['1-5 years']              [0.3 0.7]                   [0.25 0.75]                 [0.24 0.76]                 +0.013  0.05       0.063\n",
      "Superhost|['5-10 years']             [0.34 0.66]                 [0.5 0.5]                   [0.33 0.67]                 -0.149  0.159      0.01\n",
      "Superhost|['5-10 years']             [0.34 0.66]                 [0.35 0.65]                 [0.33 0.67]                 +0.001  0.009      0.01\n",
      "Superhost|['5-10 years']             [0.34 0.66]                 [0.7 0.3]                   [0.33 0.67]                 -0.349  0.359      0.01\n",
      "Superhost|['5-10 years']             [0.34 0.66]                 [0.35 0.65]                 [0.33 0.67]                 +0.001  0.009      0.01\n",
      "Superhost|['5-10 years']             [0.34 0.66]                 [0.35 0.65]                 [0.33 0.67]                 +0.001  0.009      0.01\n",
      "Superhost|['10+ years']              [0.43 0.57]                 [0.7 0.3]                   [0.73 0.27]                 +0.027  0.273      0.3\n",
      "Superhost|['10+ years']              [0.43 0.57]                 [0.5 0.5]                   [0.73 0.27]                 +0.227  0.073      0.3\n",
      "Superhost|['10+ years']              [0.43 0.57]                 [0.2 0.8]                   [0.73 0.27]                 +0.073  0.227      0.3\n",
      "Superhost|['10+ years']              [0.43 0.57]                 [0.5 0.5]                   [0.73 0.27]                 +0.227  0.073      0.3\n",
      "Superhost|['10+ years']              [0.43 0.57]                 [0.75 0.25]                 [0.73 0.27]                 -0.023  0.323      0.3\n",
      "Price|['Entire home/apt']            [0.01 0.14 0.43 0.31 0.12]  [0.05 0.1  0.3  0.4  0.15]  [0.08 0.15 0.32 0.32 0.13]  -0.060  0.167      0.107\n",
      "Price|['Entire home/apt']            [0.01 0.14 0.43 0.31 0.12]  [0.01 0.04 0.25 0.5  0.2 ]  [0.08 0.15 0.32 0.32 0.13]  -0.170  0.277      0.107\n",
      "Price|['Entire home/apt']            [0.01 0.14 0.43 0.31 0.12]  [0.05 0.1  0.35 0.4  0.1 ]  [0.08 0.15 0.32 0.32 0.13]  -0.030  0.137      0.107\n",
      "Price|['Entire home/apt']            [0.01 0.14 0.43 0.31 0.12]  [0.05 0.15 0.4  0.3  0.1 ]  [0.08 0.15 0.32 0.32 0.13]  +0.053  0.054      0.107\n",
      "Price|['Entire home/apt']            [0.01 0.14 0.43 0.31 0.12]  [0.05 0.15 0.35 0.35 0.1 ]  [0.08 0.15 0.32 0.32 0.13]  +0.010  0.097      0.107\n",
      "Price|['Private room']               [0.21 0.53 0.17 0.06 0.02]  [0.05 0.25 0.4  0.25 0.05]  [0.07 0.18 0.37 0.25 0.12]  +0.043  0.449      0.492\n",
      "Price|['Private room']               [0.21 0.53 0.17 0.06 0.02]  [0.1  0.4  0.3  0.15 0.05]  [0.07 0.18 0.37 0.25 0.12]  +0.243  0.249      0.492\n",
      "Price|['Private room']               [0.21 0.53 0.17 0.06 0.02]  [0.05 0.25 0.5  0.15 0.05]  [0.07 0.18 0.37 0.25 0.12]  +0.043  0.449      0.492\n",
      "Price|['Private room']               [0.21 0.53 0.17 0.06 0.02]  [0.05 0.25 0.4  0.25 0.05]  [0.07 0.18 0.37 0.25 0.12]  +0.043  0.449      0.492\n",
      "Price|['Private room']               [0.21 0.53 0.17 0.06 0.02]  [0.05 0.2  0.4  0.25 0.1 ]  [0.07 0.18 0.37 0.25 0.12]  -0.007  0.499      0.492\n",
      "Price|['Shared/Hotel room']          [0.67 0.19 0.07 0.04 0.03]  [0.1  0.3  0.4  0.15 0.05]  [0.1  0.27 0.34 0.21 0.08]  -0.002  0.565      0.563\n",
      "Price|['Shared/Hotel room']          [0.67 0.19 0.07 0.04 0.03]  [0.3  0.4  0.2  0.08 0.02]  [0.1  0.27 0.34 0.21 0.08]  +0.190  0.373      0.563\n",
      "Price|['Shared/Hotel room']          [0.67 0.19 0.07 0.04 0.03]  [0.1  0.4  0.3  0.15 0.05]  [0.1  0.27 0.34 0.21 0.08]  -0.002  0.565      0.563\n",
      "Price|['Shared/Hotel room']          [0.67 0.19 0.07 0.04 0.03]  [0.3  0.4  0.2  0.08 0.02]  [0.1  0.27 0.34 0.21 0.08]  +0.190  0.373      0.563\n",
      "Price|['Shared/Hotel room']          [0.67 0.19 0.07 0.04 0.03]  [0.1  0.3  0.35 0.2  0.05]  [0.1  0.27 0.34 0.21 0.08]  -0.002  0.565      0.563\n",
      "Price|['<4.3']                       [0.13 0.27 0.35 0.19 0.06]  [0.1  0.3  0.4  0.15 0.05]  [0.1  0.29 0.39 0.16 0.07]  -0.016  0.077      0.061\n",
      "Price|['<4.3']                       [0.13 0.27 0.35 0.19 0.06]  [0.1  0.3  0.4  0.15 0.05]  [0.1  0.29 0.39 0.16 0.07]  -0.016  0.077      0.061\n",
      "Price|['<4.3']                       [0.13 0.27 0.35 0.19 0.06]  [0.1  0.3  0.4  0.15 0.05]  [0.1  0.29 0.39 0.16 0.07]  -0.016  0.077      0.061\n",
      "Price|['<4.3']                       [0.13 0.27 0.35 0.19 0.06]  [0.05 0.35 0.4  0.15 0.05]  [0.1  0.29 0.39 0.16 0.07]  -0.066  0.127      0.061\n",
      "Price|['<4.3']                       [0.13 0.27 0.35 0.19 0.06]  [0.1  0.3  0.4  0.15 0.05]  [0.1  0.29 0.39 0.16 0.07]  -0.016  0.077      0.061\n",
      "Price|['4.3-4.8']                    [0.07 0.27 0.37 0.24 0.05]  [0.05 0.2  0.4  0.25 0.1 ]  [0.08 0.16 0.33 0.31 0.11]  +0.061  0.087      0.148\n",
      "Price|['4.3-4.8']                    [0.07 0.27 0.37 0.24 0.05]  [0.05 0.15 0.35 0.35 0.1 ]  [0.08 0.16 0.33 0.31 0.11]  -0.012  0.16       0.148\n",
      "Price|['4.3-4.8']                    [0.07 0.27 0.37 0.24 0.05]  [0.05 0.15 0.4  0.3  0.1 ]  [0.08 0.16 0.33 0.31 0.11]  +0.011  0.137      0.148\n",
      "Price|['4.3-4.8']                    [0.07 0.27 0.37 0.24 0.05]  [0.05 0.15 0.35 0.3  0.15]  [0.08 0.16 0.33 0.31 0.11]  -0.012  0.16       0.148\n",
      "Price|['4.3-4.8']                    [0.07 0.27 0.37 0.24 0.05]  [0.05 0.15 0.35 0.35 0.1 ]  [0.08 0.16 0.33 0.31 0.11]  -0.012  0.16       0.148\n",
      "Price|['4.9+']                       [0.06 0.22 0.36 0.27 0.1 ]  [0.05 0.1  0.35 0.4  0.1 ]  [0.08 0.16 0.33 0.28 0.15]  -0.054  0.134      0.08\n",
      "Price|['4.9+']                       [0.06 0.22 0.36 0.27 0.1 ]  [0.05 0.1  0.35 0.4  0.1 ]  [0.08 0.16 0.33 0.28 0.15]  -0.054  0.134      0.08\n",
      "Price|['4.9+']                       [0.06 0.22 0.36 0.27 0.1 ]  [0.02 0.08 0.25 0.5  0.15]  [0.08 0.16 0.33 0.28 0.15]  -0.204  0.284      0.08\n",
      "Price|['4.9+']                       [0.06 0.22 0.36 0.27 0.1 ]  [0.05 0.1  0.35 0.4  0.1 ]  [0.08 0.16 0.33 0.28 0.15]  -0.054  0.134      0.08\n",
      "Price|['4.9+']                       [0.06 0.22 0.36 0.27 0.1 ]  [0.05 0.15 0.35 0.3  0.15]  [0.08 0.16 0.33 0.28 0.15]  -0.004  0.084      0.08\n",
      "Number of Bathrooms|['0/1 bedroom']  [0.96 0.03 0.01]            [0.85 0.1  0.05]            [0.77 0.15 0.09]            +0.084  0.115      0.199\n",
      "Number of Bathrooms|['0/1 bedroom']  [0.96 0.03 0.01]            [0.85 0.1  0.05]            [0.77 0.15 0.09]            +0.084  0.115      0.199\n",
      "Number of Bathrooms|['0/1 bedroom']  [0.96 0.03 0.01]            [0.85 0.1  0.05]            [0.77 0.15 0.09]            +0.084  0.115      0.199\n",
      "Number of Bathrooms|['0/1 bedroom']  [0.96 0.03 0.01]            [0.85 0.13 0.02]            [0.77 0.15 0.09]            +0.084  0.115      0.199\n",
      "Number of Bathrooms|['0/1 bedroom']  [0.96 0.03 0.01]            [0.85 0.13 0.02]            [0.77 0.15 0.09]            +0.084  0.115      0.199\n",
      "Number of Bathrooms|['2 bedrooms']   [0.48 0.51 0.01]            [0.1  0.75 0.15]            [0.48 0.38 0.15]            -0.246  0.38       0.134\n",
      "Number of Bathrooms|['2 bedrooms']   [0.48 0.51 0.01]            [0.6  0.35 0.05]            [0.48 0.38 0.15]            -0.024  0.158      0.134\n",
      "Number of Bathrooms|['2 bedrooms']   [0.48 0.51 0.01]            [0.5 0.4 0.1]               [0.48 0.38 0.15]            +0.026  0.108      0.134\n",
      "Number of Bathrooms|['2 bedrooms']   [0.48 0.51 0.01]            [0.5 0.4 0.1]               [0.48 0.38 0.15]            +0.026  0.108      0.134\n",
      "Number of Bathrooms|['2 bedrooms']   [0.48 0.51 0.01]            [0.5 0.4 0.1]               [0.48 0.38 0.15]            +0.026  0.108      0.134\n",
      "Number of Bathrooms|['3 bedrooms']   [0.12 0.64 0.24]            [0.1 0.6 0.3]               [0.13 0.66 0.21]            -0.035  0.064      0.029\n",
      "Number of Bathrooms|['3 bedrooms']   [0.12 0.64 0.24]            [0.1 0.6 0.3]               [0.13 0.66 0.21]            -0.035  0.064      0.029\n",
      "Number of Bathrooms|['3 bedrooms']   [0.12 0.64 0.24]            [0.1 0.6 0.3]               [0.13 0.66 0.21]            -0.035  0.064      0.029\n",
      "Number of Bathrooms|['3 bedrooms']   [0.12 0.64 0.24]            [0.1 0.7 0.2]               [0.13 0.66 0.21]            -0.030  0.059      0.029\n",
      "Number of Bathrooms|['3 bedrooms']   [0.12 0.64 0.24]            [0.1 0.7 0.2]               [0.13 0.66 0.21]            -0.030  0.059      0.029\n",
      "Number of Bathrooms|['4+ bedrooms']  [0.02 0.21 0.77]            [0.05 0.25 0.7 ]            [0.08 0.25 0.67]            +0.028  0.069      0.097\n",
      "Number of Bathrooms|['4+ bedrooms']  [0.02 0.21 0.77]            [0.05 0.25 0.7 ]            [0.08 0.25 0.67]            +0.028  0.069      0.097\n",
      "Number of Bathrooms|['4+ bedrooms']  [0.02 0.21 0.77]            [0.1 0.3 0.6]               [0.08 0.25 0.67]            -0.072  0.169      0.097\n",
      "Number of Bathrooms|['4+ bedrooms']  [0.02 0.21 0.77]            [0.05 0.25 0.7 ]            [0.08 0.25 0.67]            +0.028  0.069      0.097\n",
      "Number of Bathrooms|['4+ bedrooms']  [0.02 0.21 0.77]            [0.05 0.25 0.7 ]            [0.08 0.25 0.67]            +0.028  0.069      0.097\n",
      "Review Score|['<1 year']             [0.14 0.34 0.52]            [0.2 0.5 0.3]               [0.21 0.51 0.28]            +0.022  0.218      0.24\n",
      "Review Score|['<1 year']             [0.14 0.34 0.52]            [0.2 0.5 0.3]               [0.21 0.51 0.28]            +0.022  0.218      0.24\n",
      "Review Score|['<1 year']             [0.14 0.34 0.52]            [0.3 0.5 0.2]               [0.21 0.51 0.28]            -0.078  0.318      0.24\n",
      "Review Score|['<1 year']             [0.14 0.34 0.52]            [0.3 0.5 0.2]               [0.21 0.51 0.28]            -0.078  0.318      0.24\n",
      "Review Score|['<1 year']             [0.14 0.34 0.52]            [0.2 0.5 0.3]               [0.21 0.51 0.28]            +0.022  0.218      0.24\n",
      "Review Score|['1-5 years']           [0.11 0.45 0.44]            [0.1 0.5 0.4]               [0.14 0.51 0.35]            +0.039  0.048      0.087\n",
      "Review Score|['1-5 years']           [0.11 0.45 0.44]            [0.1 0.5 0.4]               [0.14 0.51 0.35]            +0.039  0.048      0.087\n",
      "Review Score|['1-5 years']           [0.11 0.45 0.44]            [0.1 0.4 0.5]               [0.14 0.51 0.35]            +0.026  0.061      0.087\n",
      "Review Score|['1-5 years']           [0.11 0.45 0.44]            [0.1 0.6 0.3]               [0.14 0.51 0.35]            -0.061  0.148      0.087\n",
      "Review Score|['1-5 years']           [0.11 0.45 0.44]            [0.1 0.5 0.4]               [0.14 0.51 0.35]            +0.039  0.048      0.087\n",
      "Review Score|['5-10 years']          [0.07 0.45 0.48]            [0.1  0.35 0.55]            [0.12 0.5  0.37]            +0.006  0.099      0.105\n",
      "Review Score|['5-10 years']          [0.07 0.45 0.48]            [0.1 0.4 0.5]               [0.12 0.5  0.37]            +0.056  0.049      0.105\n",
      "Review Score|['5-10 years']          [0.07 0.45 0.48]            [0.1 0.4 0.5]               [0.12 0.5  0.37]            +0.056  0.049      0.105\n",
      "Review Score|['5-10 years']          [0.07 0.45 0.48]            [0.1 0.5 0.4]               [0.12 0.5  0.37]            +0.027  0.078      0.105\n",
      "Review Score|['5-10 years']          [0.07 0.45 0.48]            [0.1 0.5 0.4]               [0.12 0.5  0.37]            +0.027  0.078      0.105\n",
      "Review Score|['10+ years']           [0.04 0.45 0.51]            [0.1 0.4 0.5]               [0.07 0.26 0.68]            +0.134  0.055      0.189\n",
      "Review Score|['10+ years']           [0.04 0.45 0.51]            [0.1 0.3 0.6]               [0.07 0.26 0.68]            +0.043  0.146      0.189\n",
      "Review Score|['10+ years']           [0.04 0.45 0.51]            [0.05 0.25 0.7 ]            [0.07 0.26 0.68]            -0.007  0.196      0.189\n",
      "Review Score|['10+ years']           [0.04 0.45 0.51]            [0.1 0.3 0.6]               [0.07 0.26 0.68]            +0.043  0.146      0.189\n",
      "Review Score|['10+ years']           [0.04 0.45 0.51]            [0.05 0.25 0.7 ]            [0.07 0.26 0.68]            -0.007  0.196      0.189\n",
      "Total Beds|['0/1 bedroom']           [0.73 0.23 0.03 0.01]       [0.6  0.25 0.1  0.05]       [0.66 0.19 0.09 0.06]       -0.022  0.132      0.11\n",
      "Total Beds|['0/1 bedroom']           [0.73 0.23 0.03 0.01]       [0.6  0.3  0.08 0.02]       [0.66 0.19 0.09 0.06]       -0.022  0.132      0.11\n",
      "Total Beds|['0/1 bedroom']           [0.73 0.23 0.03 0.01]       [0.7  0.2  0.08 0.02]       [0.66 0.19 0.09 0.06]       +0.051  0.059      0.11\n",
      "Total Beds|['0/1 bedroom']           [0.73 0.23 0.03 0.01]       [0.6  0.3  0.08 0.02]       [0.66 0.19 0.09 0.06]       -0.022  0.132      0.11\n",
      "Total Beds|['0/1 bedroom']           [0.73 0.23 0.03 0.01]       [0.7  0.2  0.08 0.02]       [0.66 0.19 0.09 0.06]       +0.051  0.059      0.11\n",
      "Total Beds|['2 bedrooms']            [0.04 0.61 0.25 0.1 ]       [0.05 0.5  0.3  0.15]       [0.14 0.45 0.26 0.14]       +0.048  0.107      0.155\n",
      "Total Beds|['2 bedrooms']            [0.04 0.61 0.25 0.1 ]       [0.05 0.5  0.3  0.15]       [0.14 0.45 0.26 0.14]       +0.048  0.107      0.155\n",
      "Total Beds|['2 bedrooms']            [0.04 0.61 0.25 0.1 ]       [0.05 0.5  0.3  0.15]       [0.14 0.45 0.26 0.14]       +0.048  0.107      0.155\n",
      "Total Beds|['2 bedrooms']            [0.04 0.61 0.25 0.1 ]       [0.05 0.5  0.3  0.15]       [0.14 0.45 0.26 0.14]       +0.048  0.107      0.155\n",
      "Total Beds|['2 bedrooms']            [0.04 0.61 0.25 0.1 ]       [0.05 0.5  0.3  0.15]       [0.14 0.45 0.26 0.14]       +0.048  0.107      0.155\n",
      "Total Beds|['3 bedrooms']            [0.02 0.03 0.52 0.44]       [0.05 0.1  0.4  0.45]       [0.1  0.12 0.46 0.32]       +0.058  0.118      0.176\n",
      "Total Beds|['3 bedrooms']            [0.02 0.03 0.52 0.44]       [0.05 0.1  0.5  0.35]       [0.1  0.12 0.46 0.32]       +0.072  0.104      0.176\n",
      "Total Beds|['3 bedrooms']            [0.02 0.03 0.52 0.44]       [0.05 0.15 0.4  0.4 ]       [0.1  0.12 0.46 0.32]       +0.022  0.154      0.176\n",
      "Total Beds|['3 bedrooms']            [0.02 0.03 0.52 0.44]       [0.05 0.1  0.5  0.35]       [0.1  0.12 0.46 0.32]       +0.072  0.104      0.176\n",
      "Total Beds|['3 bedrooms']            [0.02 0.03 0.52 0.44]       [0.05 0.1  0.5  0.35]       [0.1  0.12 0.46 0.32]       +0.072  0.104      0.176\n",
      "Total Beds|['4+ bedrooms']           [0.02 0.   0.02 0.95]       [0.05 0.1  0.2  0.65]       [0.09 0.11 0.19 0.61]       +0.044  0.303      0.347\n",
      "Total Beds|['4+ bedrooms']           [0.02 0.   0.02 0.95]       [0.05 0.1  0.25 0.6 ]       [0.09 0.11 0.19 0.61]       -0.006  0.353      0.347\n",
      "Total Beds|['4+ bedrooms']           [0.02 0.   0.02 0.95]       [0.05 0.15 0.3  0.5 ]       [0.09 0.11 0.19 0.61]       -0.106  0.453      0.347\n",
      "Total Beds|['4+ bedrooms']           [0.02 0.   0.02 0.95]       [0.05 0.1  0.2  0.65]       [0.09 0.11 0.19 0.61]       +0.044  0.303      0.347\n",
      "Total Beds|['4+ bedrooms']           [0.02 0.   0.02 0.95]       [0.05 0.1  0.2  0.65]       [0.09 0.11 0.19 0.61]       +0.044  0.303      0.347\n",
      "-----------------------------------  --------------------------  --------------------------  --------------------------  ------  ---------  ----------\n",
      "#################################################\n",
      "Average change: 0.006437499999999999\n"
     ]
    }
   ],
   "source": [
    "table = [[\"Constraint\", \"Ground\", \"Zero\", \"Joint\", \"Change\", \"Zero dist\", \"Joint dist\"],]\n",
    "tot_change = 0\n",
    "cnt = 0\n",
    "\n",
    "for i, cons in enumerate(constraints['schema']['Constraints']):\n",
    "    er = False\n",
    "    ground = dat.marg(cons, target_values=var_dict[cons['Target'][0]['Name']])\n",
    "    pred = cons['Probability']\n",
    "    if np.isnan(ground).any():\n",
    "        print(\"Error: nan in ground\", cons)\n",
    "        print(\"#################################################\")\n",
    "        continue\n",
    "    query = {}\n",
    "    query['Queries'] = [{'Target': cons['Target'], 'Condition': cons['Condition']}]\n",
    "    after = model.query(query)\n",
    "    for j in prob_file[json.dumps(query['Queries'][0])]:\n",
    "        ground = np.array(ground)\n",
    "        zero = np.array(j['Probability'])\n",
    "        after = np.array(after)\n",
    "        tar_name = cons['Target'][0]['Name']\n",
    "        cons_val = '[]'\n",
    "        if len(cons['Condition']) > 0:\n",
    "            cons_val = cons['Condition'][0]['Value']\n",
    "        try:\n",
    "            zero_metric = metrics(ground, zero)\n",
    "        except:\n",
    "            er = True\n",
    "            zero_metric = float('nan')\n",
    "        after_metric = metrics(ground, after)\n",
    "        table.append([f\"{tar_name}|{cons_val}\", str(ground.round(2)), str(zero.round(2)), str(after.round(2)), f\"{(after_metric-zero_metric):+.3f}\", zero_metric, after_metric])\n",
    "    if not er:\n",
    "        tot_change += after_metric-zero_metric\n",
    "        cnt += 1\n",
    "print(tabulate(table))\n",
    "print(\"#################################################\")\n",
    "print(f\"Average change: {tot_change/cnt}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['{\"Target\": [{\"Name\": \"Superhost\", \"Value\": [\"Yes\", \"No\"]}], \"Condition\": []}', '{\"Target\": [{\"Name\": \"Number of Bathrooms\", \"Value\": [\"0/1 bathroom\", \"2 bathrooms\", \"3+ bathrooms\"]}], \"Condition\": []}', '{\"Target\": [{\"Name\": \"Host Experience\", \"Value\": [\"<1 year\", \"1-5 years\", \"5-10 years\", \"10+ years\"]}], \"Condition\": []}', '{\"Target\": [{\"Name\": \"Number of Bedrooms\", \"Value\": [\"0/1 bedroom\", \"2 bedrooms\", \"3 bedrooms\", \"4+ bedrooms\"]}], \"Condition\": []}', '{\"Target\": [{\"Name\": \"Review Score\", \"Value\": [\"<4.3\", \"4.3-4.8\", \"4.9+\"]}], \"Condition\": []}', '{\"Target\": [{\"Name\": \"Price\", \"Value\": [\"<$50\", \"$51-$100\", \"$101-$200\", \"$201-$500\", \"$500+\"]}], \"Condition\": []}', '{\"Target\": [{\"Name\": \"Total Beds\", \"Value\": [\"0/1 bed\", \"2 beds\", \"3 beds\", \"4+ beds\"]}], \"Condition\": []}', '{\"Target\": [{\"Name\": \"Room Type\", \"Value\": [\"Entire home/apt\", \"Private room\", \"Shared/Hotel room\"]}], \"Condition\": []}', '{\"Target\": [{\"Name\": \"Review Score\", \"Value\": [\"<4.3\", \"4.3-4.8\", \"4.9+\"]}], \"Condition\": [{\"Name\": \"Superhost\", \"Value\": [\"Yes\"], \"Other\": [\"No\"]}]}', '{\"Target\": [{\"Name\": \"Number of Bathrooms\", \"Value\": [\"0/1 bathroom\", \"2 bathrooms\", \"3+ bathrooms\"]}], \"Condition\": [{\"Name\": \"Number of Bedrooms\", \"Value\": [\"0/1 bedroom\"], \"Other\": [\"2 bedrooms\", \"3 bedrooms\", \"4+ bedrooms\"]}]}', '{\"Target\": [{\"Name\": \"Price\", \"Value\": [\"<$50\", \"$51-$100\", \"$101-$200\", \"$201-$500\", \"$500+\"]}], \"Condition\": [{\"Name\": \"Room Type\", \"Value\": [\"Entire home/apt\"], \"Other\": [\"Private room\", \"Shared/Hotel room\"]}]}', '{\"Target\": [{\"Name\": \"Superhost\", \"Value\": [\"Yes\", \"No\"]}], \"Condition\": [{\"Name\": \"Host Experience\", \"Value\": [\"<1 year\"], \"Other\": [\"1-5 years\", \"5-10 years\", \"10+ years\"]}]}', '{\"Target\": [{\"Name\": \"Price\", \"Value\": [\"<$50\", \"$51-$100\", \"$101-$200\", \"$201-$500\", \"$500+\"]}], \"Condition\": [{\"Name\": \"Review Score\", \"Value\": [\"<4.3\"], \"Other\": [\"4.3-4.8\", \"4.9+\"]}]}', '{\"Target\": [{\"Name\": \"Total Beds\", \"Value\": [\"0/1 bed\", \"2 beds\", \"3 beds\", \"4+ beds\"]}], \"Condition\": [{\"Name\": \"Number of Bedrooms\", \"Value\": [\"0/1 bedroom\"], \"Other\": [\"2 bedrooms\", \"3 bedrooms\", \"4+ bedrooms\"]}]}', '{\"Target\": [{\"Name\": \"Review Score\", \"Value\": [\"<4.3\", \"4.3-4.8\", \"4.9+\"]}], \"Condition\": [{\"Name\": \"Host Experience\", \"Value\": [\"<1 year\"], \"Other\": [\"1-5 years\", \"5-10 years\", \"10+ years\"]}]}', '{\"Target\": [{\"Name\": \"Superhost\", \"Value\": [\"Yes\", \"No\"]}], \"Condition\": [{\"Name\": \"Host Experience\", \"Value\": [\"1-5 years\"], \"Other\": [\"<1 year\", \"5-10 years\", \"10+ years\"]}]}', '{\"Target\": [{\"Name\": \"Review Score\", \"Value\": [\"<4.3\", \"4.3-4.8\", \"4.9+\"]}], \"Condition\": [{\"Name\": \"Superhost\", \"Value\": [\"No\"], \"Other\": [\"Yes\"]}]}', '{\"Target\": [{\"Name\": \"Number of Bathrooms\", \"Value\": [\"0/1 bathroom\", \"2 bathrooms\", \"3+ bathrooms\"]}], \"Condition\": [{\"Name\": \"Number of Bedrooms\", \"Value\": [\"2 bedrooms\"], \"Other\": [\"0/1 bedroom\", \"3 bedrooms\", \"4+ bedrooms\"]}]}', '{\"Target\": [{\"Name\": \"Review Score\", \"Value\": [\"<4.3\", \"4.3-4.8\", \"4.9+\"]}], \"Condition\": [{\"Name\": \"Host Experience\", \"Value\": [\"1-5 years\"], \"Other\": [\"<1 year\", \"5-10 years\", \"10+ years\"]}]}', '{\"Target\": [{\"Name\": \"Price\", \"Value\": [\"<$50\", \"$51-$100\", \"$101-$200\", \"$201-$500\", \"$500+\"]}], \"Condition\": [{\"Name\": \"Review Score\", \"Value\": [\"4.3-4.8\"], \"Other\": [\"<4.3\", \"4.9+\"]}]}', '{\"Target\": [{\"Name\": \"Price\", \"Value\": [\"<$50\", \"$51-$100\", \"$101-$200\", \"$201-$500\", \"$500+\"]}], \"Condition\": [{\"Name\": \"Room Type\", \"Value\": [\"Private room\"], \"Other\": [\"Entire home/apt\", \"Shared/Hotel room\"]}]}', '{\"Target\": [{\"Name\": \"Total Beds\", \"Value\": [\"0/1 bed\", \"2 beds\", \"3 beds\", \"4+ beds\"]}], \"Condition\": [{\"Name\": \"Number of Bedrooms\", \"Value\": [\"2 bedrooms\"], \"Other\": [\"0/1 bedroom\", \"3 bedrooms\", \"4+ bedrooms\"]}]}', '{\"Target\": [{\"Name\": \"Superhost\", \"Value\": [\"Yes\", \"No\"]}], \"Condition\": [{\"Name\": \"Host Experience\", \"Value\": [\"5-10 years\"], \"Other\": [\"<1 year\", \"1-5 years\", \"10+ years\"]}]}', '{\"Target\": [{\"Name\": \"Number of Bathrooms\", \"Value\": [\"0/1 bathroom\", \"2 bathrooms\", \"3+ bathrooms\"]}], \"Condition\": [{\"Name\": \"Number of Bedrooms\", \"Value\": [\"3 bedrooms\"], \"Other\": [\"0/1 bedroom\", \"2 bedrooms\", \"4+ bedrooms\"]}]}', '{\"Target\": [{\"Name\": \"Price\", \"Value\": [\"<$50\", \"$51-$100\", \"$101-$200\", \"$201-$500\", \"$500+\"]}], \"Condition\": [{\"Name\": \"Room Type\", \"Value\": [\"Shared/Hotel room\"], \"Other\": [\"Entire home/apt\", \"Private room\"]}]}', '{\"Target\": [{\"Name\": \"Review Score\", \"Value\": [\"<4.3\", \"4.3-4.8\", \"4.9+\"]}], \"Condition\": [{\"Name\": \"Host Experience\", \"Value\": [\"5-10 years\"], \"Other\": [\"<1 year\", \"1-5 years\", \"10+ years\"]}]}', '{\"Target\": [{\"Name\": \"Price\", \"Value\": [\"<$50\", \"$51-$100\", \"$101-$200\", \"$201-$500\", \"$500+\"]}], \"Condition\": [{\"Name\": \"Review Score\", \"Value\": [\"4.9+\"], \"Other\": [\"<4.3\", \"4.3-4.8\"]}]}', '{\"Target\": [{\"Name\": \"Total Beds\", \"Value\": [\"0/1 bed\", \"2 beds\", \"3 beds\", \"4+ beds\"]}], \"Condition\": [{\"Name\": \"Number of Bedrooms\", \"Value\": [\"3 bedrooms\"], \"Other\": [\"0/1 bedroom\", \"2 bedrooms\", \"4+ bedrooms\"]}]}', '{\"Target\": [{\"Name\": \"Superhost\", \"Value\": [\"Yes\", \"No\"]}], \"Condition\": [{\"Name\": \"Host Experience\", \"Value\": [\"10+ years\"], \"Other\": [\"<1 year\", \"1-5 years\", \"5-10 years\"]}]}', '{\"Target\": [{\"Name\": \"Number of Bathrooms\", \"Value\": [\"0/1 bathroom\", \"2 bathrooms\", \"3+ bathrooms\"]}], \"Condition\": [{\"Name\": \"Number of Bedrooms\", \"Value\": [\"4+ bedrooms\"], \"Other\": [\"0/1 bedroom\", \"2 bedrooms\", \"3 bedrooms\"]}]}', '{\"Target\": [{\"Name\": \"Review Score\", \"Value\": [\"<4.3\", \"4.3-4.8\", \"4.9+\"]}], \"Condition\": [{\"Name\": \"Host Experience\", \"Value\": [\"10+ years\"], \"Other\": [\"<1 year\", \"1-5 years\", \"5-10 years\"]}]}', '{\"Target\": [{\"Name\": \"Total Beds\", \"Value\": [\"0/1 bed\", \"2 beds\", \"3 beds\", \"4+ beds\"]}], \"Condition\": [{\"Name\": \"Number of Bedrooms\", \"Value\": [\"4+ bedrooms\"], \"Other\": [\"0/1 bedroom\", \"2 bedrooms\", \"3 bedrooms\"]}]}'])"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prob_file.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Variables': [{'Name': 'Number of Bedrooms',\n",
       "   'Value': ['0/1 bedroom', '2 bedrooms', '3 bedrooms', '4+ bedrooms']},\n",
       "  {'Name': 'Number of Bathrooms',\n",
       "   'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']},\n",
       "  {'Name': 'Host Experience',\n",
       "   'Value': ['<1 year', '1-5 years', '5-10 years', '10+ years']},\n",
       "  {'Name': 'Superhost', 'Value': ['Yes', 'No']},\n",
       "  {'Name': 'Room Type',\n",
       "   'Value': ['Entire home/apt', 'Private room', 'Shared room']},\n",
       "  {'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']},\n",
       "  {'Name': 'Price',\n",
       "   'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       " 'Constraints': [{'Target': [{'Name': 'Number of Bedrooms',\n",
       "     'Value': ['0/1 bedroom', '2 bedrooms', '3 bedrooms', '4+ bedrooms']}],\n",
       "   'Probability': [0.3, 0.4, 0.2, 0.1],\n",
       "   'Condition': []},\n",
       "  {'Target': [{'Name': 'Number of Bathrooms',\n",
       "     'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "   'Probability': [0.4, 0.35, 0.25],\n",
       "   'Condition': []},\n",
       "  {'Target': [{'Name': 'Host Experience',\n",
       "     'Value': ['<1 year', '1-5 years', '5-10 years', '10+ years']}],\n",
       "   'Probability': [0.2, 0.5, 0.25, 0.05],\n",
       "   'Condition': []},\n",
       "  {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "   'Probability': [0.3, 0.7],\n",
       "   'Condition': []},\n",
       "  {'Target': [{'Name': 'Room Type',\n",
       "     'Value': ['Entire home/apt', 'Private room', 'Shared room']}],\n",
       "   'Probability': [0.6, 0.35, 0.05],\n",
       "   'Condition': []},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Probability': [0.15, 0.5, 0.35],\n",
       "   'Condition': []},\n",
       "  {'Target': [{'Name': 'Number of Bathrooms',\n",
       "     'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "   'Probability': [0.7, 0.25, 0.05],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['0/1 bedroom']}]},\n",
       "  {'Target': [{'Name': 'Number of Bathrooms',\n",
       "     'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "   'Probability': [0.15, 0.6, 0.25],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['3 bedrooms']}]},\n",
       "  {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "   'Probability': [0.25, 0.75],\n",
       "   'Condition': [{'Name': 'Host Experience', 'Value': ['<1 year']}]},\n",
       "  {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "   'Probability': [0.3, 0.7],\n",
       "   'Condition': [{'Name': 'Host Experience', 'Value': ['1-5 years']}]},\n",
       "  {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "   'Probability': [0.4, 0.6],\n",
       "   'Condition': [{'Name': 'Host Experience', 'Value': ['5-10 years']}]},\n",
       "  {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "   'Probability': [0.4, 0.6],\n",
       "   'Condition': [{'Name': 'Host Experience', 'Value': ['10+ years']}]},\n",
       "  {'Target': [{'Name': 'Room Type',\n",
       "     'Value': ['Entire home/apt', 'Private room', 'Shared room']}],\n",
       "   'Probability': [0.6, 0.3, 0.1],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['0/1 bedroom']}]},\n",
       "  {'Target': [{'Name': 'Room Type',\n",
       "     'Value': ['Entire home/apt', 'Private room', 'Shared room']}],\n",
       "   'Probability': [0.75, 0.15, 0.1],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['2 bedrooms']}]},\n",
       "  {'Target': [{'Name': 'Room Type',\n",
       "     'Value': ['Entire home/apt', 'Private room', 'Shared room']}],\n",
       "   'Probability': [0.9, 0.075, 0.025],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['4+ bedrooms']}]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Probability': [0.05, 0.3, 0.65],\n",
       "   'Condition': [{'Name': 'Superhost', 'Value': ['Yes']}]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Probability': [0.15, 0.6, 0.25],\n",
       "   'Condition': [{'Name': 'Superhost', 'Value': ['No']}]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Probability': [0.1, 0.3, 0.6],\n",
       "   'Condition': [{'Name': 'Host Experience', 'Value': ['<1 year']}]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Probability': [0.15, 0.6, 0.25],\n",
       "   'Condition': [{'Name': 'Host Experience', 'Value': ['1-5 years']}]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Probability': [0.1, 0.4, 0.5],\n",
       "   'Condition': [{'Name': 'Host Experience', 'Value': ['5-10 years']}]},\n",
       "  {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "   'Probability': [0.1, 0.3, 0.6],\n",
       "   'Condition': [{'Name': 'Host Experience', 'Value': ['10+ years']}]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Probability': [0.05, 0.2, 0.4, 0.3, 0.05],\n",
       "   'Condition': [{'Name': 'Room Type', 'Value': ['Entire home/apt']}]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Probability': [0.25, 0.5, 0.1667, 0.05, 0.0333],\n",
       "   'Condition': [{'Name': 'Room Type', 'Value': ['Shared room']}]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Probability': [0.05, 0.3, 0.4, 0.2, 0.05],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['0/1 bedroom']}]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Probability': [0.05, 0.2, 0.4, 0.3, 0.05],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['2 bedrooms']}]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Probability': [0.05, 0.15, 0.4, 0.3, 0.1],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['3 bedrooms']}]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Probability': [0.05, 0.15, 0.3, 0.4, 0.1],\n",
       "   'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['4+ bedrooms']}]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Probability': [0.1, 0.3, 0.4, 0.15, 0.05],\n",
       "   'Condition': [{'Name': 'Number of Bathrooms', 'Value': ['0/1 bathroom']}]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Probability': [0.05, 0.2, 0.4, 0.3, 0.05],\n",
       "   'Condition': [{'Name': 'Number of Bathrooms', 'Value': ['2 bathrooms']}]},\n",
       "  {'Target': [{'Name': 'Price',\n",
       "     'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "   'Probability': [0.05, 0.15, 0.4, 0.3, 0.1],\n",
       "   'Condition': [{'Name': 'Number of Bathrooms', 'Value': ['3+ bathrooms']}]}]}"
      ]
     },
     "execution_count": 387,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tmp = con.copy()\n",
    "tmp = remove_vars(tmp, [], \"Price\")\n",
    "tmp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'Target': [{'Name': 'Number of Bedrooms',\n",
       "    'Value': ['0/1 bedroom', '2 bedrooms', '3 bedrooms', '4+ bedrooms']}],\n",
       "  'Probability': [0.3, 0.4, 0.2, 0.1],\n",
       "  'Condition': []},\n",
       " {'Target': [{'Name': 'Number of Bathrooms',\n",
       "    'Value': ['0/1 bathroom', '2 bathrooms', '3+ bathrooms']}],\n",
       "  'Probability': [0.5, 0.3, 0.2],\n",
       "  'Condition': []},\n",
       " {'Target': [{'Name': 'Host Experience',\n",
       "    'Value': ['<1 year', '1-5 years', '5-10 years', '10+ years']}],\n",
       "  'Probability': [0.2727, 0.4545, 0.1818, 0.0909],\n",
       "  'Condition': []},\n",
       " {'Target': [{'Name': 'Superhost', 'Value': ['Yes', 'No']}],\n",
       "  'Probability': [0.3, 0.7],\n",
       "  'Condition': []},\n",
       " {'Target': [{'Name': 'Room Type',\n",
       "    'Value': ['Entire home/apt', 'Private room', 'Shared room']}],\n",
       "  'Probability': [0.6, 0.35, 0.05],\n",
       "  'Condition': []},\n",
       " {'Target': [{'Name': 'Review Score', 'Value': ['<4.3', '4.3-4.8', '4.9+']}],\n",
       "  'Probability': [0.15, 0.6, 0.25],\n",
       "  'Condition': []},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.3, 0.4, 0.15, 0.05],\n",
       "  'Condition': []},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.3, 0.4, 0.15, 0.05],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['0/1 bedroom']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.2, 0.4, 0.2, 0.1],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['2 bedrooms']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.05, 0.2, 0.5, 0.2, 0.05],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['3 bedrooms']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.2, 0.3, 0.3, 0.1],\n",
       "  'Condition': [{'Name': 'Number of Bedrooms', 'Value': ['4+ bedrooms']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.3, 0.4, 0.15, 0.05],\n",
       "  'Condition': [{'Name': 'Number of Bathrooms', 'Value': ['0/1 bathroom']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.15, 0.25, 0.35, 0.2, 0.05],\n",
       "  'Condition': [{'Name': 'Number of Bathrooms', 'Value': ['2 bathrooms']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.05, 0.15, 0.5, 0.25, 0.05],\n",
       "  'Condition': [{'Name': 'Number of Bathrooms', 'Value': ['3+ bathrooms']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.3, 0.4, 0.15, 0.05],\n",
       "  'Condition': [{'Name': 'Host Experience', 'Value': ['<1 year']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.3, 0.4, 0.15, 0.05],\n",
       "  'Condition': [{'Name': 'Host Experience', 'Value': ['1-5 years']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.15, 0.25, 0.35, 0.2, 0.05],\n",
       "  'Condition': [{'Name': 'Host Experience', 'Value': ['5-10 years']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1667, 0.3333, 0.5, 0.25, 0.0833],\n",
       "  'Condition': [{'Name': 'Host Experience', 'Value': ['10+ years']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.3, 0.4, 0.15, 0.05],\n",
       "  'Condition': [{'Name': 'Superhost', 'Value': ['Yes']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.15, 0.25, 0.35, 0.2, 0.05],\n",
       "  'Condition': [{'Name': 'Superhost', 'Value': ['No']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.3, 0.4, 0.15, 0.05],\n",
       "  'Condition': [{'Name': 'Room Type', 'Value': ['Entire home/apt']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.3, 0.4, 0.16, 0.04],\n",
       "  'Condition': [{'Name': 'Room Type', 'Value': ['Private room']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.2, 0.4, 0.3, 0.08, 0.02],\n",
       "  'Condition': [{'Name': 'Room Type', 'Value': ['Shared room']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.3, 0.25, 0.2, 0.15],\n",
       "  'Condition': [{'Name': 'Review Score', 'Value': ['<4.3']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.3, 0.4, 0.15, 0.05],\n",
       "  'Condition': [{'Name': 'Review Score', 'Value': ['4.3-4.8']}]},\n",
       " {'Target': [{'Name': 'Price',\n",
       "    'Value': ['<$50', '$51-$100', '$101-$200', '$201-$500', '$500+']}],\n",
       "  'Probability': [0.1, 0.2, 0.3, 0.3, 0.1],\n",
       "  'Condition': [{'Name': 'Review Score', 'Value': ['4.9+']}]}]"
      ]
     },
     "execution_count": 376,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "con['Constraints']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def remove_vars(constraints, vars, one=''):\n",
    "    ret = {}\n",
    "    ret['Variables'] = []\n",
    "    ret['Constraints'] = []\n",
    "    check = set()\n",
    "    for var in constraints['Variables']:\n",
    "        if var['Name'] not in vars:\n",
    "            ret['Variables'].append(var)\n",
    "    for cons in constraints['Constraints']:\n",
    "        if cons is None:\n",
    "            print(\"None\")\n",
    "            continue\n",
    "        if cons['Target'][0]['Name'] not in vars and (len(cons['Condition']) == 0 or (cons['Condition'][0]['Name'] not in vars)):\n",
    "            if len(cons['Condition']) != 0 or cons['Target'][0]['Name'] != one:\n",
    "                tt = {}\n",
    "                tt['Target'] = cons['Target']\n",
    "                tt['Probability'] = cons['Probability']\n",
    "                #tt['Question'] = cons['Question']\n",
    "                tt['Condition'] = []\n",
    "                for cond in cons['Condition']:\n",
    "                    if cond['Name'] == 'City':\n",
    "                        continue\n",
    "                    tt['Condition'].append(cond)\n",
    "                ret['Constraints'].append(tt)\n",
    "    return ret"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "dl",
   "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": "3.8.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}