08_conditional_manatees-2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/ksachdeva/rethinking-tensorflow-probability/blob/master/notebooks/08_conditional_manatees.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 8. Conditional Manatees"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import arviz as az\n",
    "import pandas as pd\n",
    "import xarray as xr\n",
    "import matplotlib.pyplot as plt\n",
    "from collections import namedtuple\n",
    "import statsmodels.formula.api as smf\n",
    "\n",
    "import random\n",
    "import jax\n",
    "import jax.numpy as jnp\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow_probability.substrates import jax as tfp\n",
    "tfd = tfp.distributions\n",
    "tfb = tfp.bijectors\n",
    "# to use 'sample_distributions'\n",
    "import tensorflow_probability as tfpx\n",
    "tfdx = tfpx.distributions\n",
    "tfbx = tfpx.bijectors\n",
    "\n",
    "plt.rcParams['figure.figsize'] = [6, 5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "key = tfp.util.SeedStream(8, salt=\"ch-08\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 8.1 Building an interaction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 8.1.1 Making a rugged model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.1\n",
    "\n",
    "\n",
    "Every inference is conditional on something whether we notice it or not !\n",
    "\n",
    "So far in the book, it was assumed that the predictor variable has an independent association with the mean of the outcome. What if want to allow the association to be conditional ?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "d = pd.read_csv(\"./data/rugged.csv\", sep=\";\")\n",
    "\n",
    "# make log version of outcome\n",
    "d[\"log_gdp\"] = d[\"rgdppc_2000\"].pipe(np.log)\n",
    "\n",
    "# extract countries with GDP data\n",
    "dd = d[d[\"rgdppc_2000\"].notnull()].copy()\n",
    "\n",
    "# rescale variables\n",
    "dd[\"log_gdp_std\"] = dd.log_gdp / dd.log_gdp.mean()\n",
    "dd[\"rugged_std\"] = dd.rugged / dd.rugged.max()\n",
    "\n",
    "# split countries into Africa and not-Africa\n",
    "d_A1 = dd[dd[\"cont_africa\"] == 1]  # Africa\n",
    "d_A0 = dd[dd[\"cont_africa\"] == 0]  # not Africa"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note - terrain ruggedness is divided by the maximum value observed"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tfpx version\n",
    "def model_81n(rugged_std, scale_a, scale_b):\n",
    "    def _generator():\n",
    "        a = yield tfdx.Normal(loc=1.0, scale=scale_a, name=\"a\")\n",
    "        b = yield tfdx.Normal(loc=0.0, scale=scale_b, name=\"b\")\n",
    "        sigma = yield tfdx.Exponential(rate=1.0, name=\"sigma\")\n",
    "        mu = a + b * (rugged_std - 0.215)\n",
    "        yield tfdx.Normal(loc=mu, scale=sigma, name=\"log_gdp_std\")\n",
    "\n",
    "    return tfdx.JointDistributionCoroutineAutoBatched(_generator, validate_args=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "rugged_seq = np.linspace(-0.1, 1.1, num=30)\n",
    "jd_81n_test = model_81n(rugged_std=rugged_seq, scale_a=1.0, scale_b=1.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_81, ss_81 = jd_81n_test.sample_distributions(50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mu = ds_81[-1].loc\n",
    "\n",
    "plt.subplot(xlim=(0, 1), ylim=(0.5, 1.5), xlabel=\"ruggedness\", ylabel=\"log GDP\")\n",
    "plt.gca().axhline(dd.log_gdp_std.min(), ls=\"--\")\n",
    "plt.gca().axhline(dd.log_gdp_std.max(), ls=\"--\")\n",
    "\n",
    "for i in range(50):\n",
    "    plt.plot(rugged_seq, mu[i], \"k\", alpha=0.3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.36"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "idx_81 = ss_81._field_name_to_index\n",
    "np.sum(np.abs(ss_81[idx_81['b']]) > 0.6) / ss_81[idx_81['b']].shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "rugged_seq = np.linspace(-0.1, 1.1, num=30)\n",
    "jd_81n_test = model_81n(rugged_std=rugged_seq, scale_a=0.1, scale_b=0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_81, ss_81 = jd_81n_test.sample_distributions(50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mu = ds_81[-1].loc\n",
    "\n",
    "plt.subplot(xlim=(0, 1), ylim=(0.5, 1.5), xlabel=\"ruggedness\", ylabel=\"log GDP\")\n",
    "plt.gca().axhline(dd.log_gdp_std.min(), ls=\"--\")\n",
    "plt.gca().axhline(dd.log_gdp_std.max(), ls=\"--\")\n",
    "\n",
    "for i in range(50):\n",
    "    plt.plot(rugged_seq, mu[i], \"k\", alpha=0.3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.6"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_81(rugged_std, scale_a, scale_b):\n",
    "    def _generator():\n",
    "        a = yield tfd.Normal(loc=1.0, scale=scale_a, name=\"a\")\n",
    "        b = yield tfd.Normal(loc=0.0, scale=scale_b, name=\"b\")\n",
    "        sigma = yield tfd.Exponential(rate=1.0, name=\"sigma\")\n",
    "        mu = a + b * (rugged_std - 0.215)\n",
    "        yield tfd.Normal(loc=mu, scale=sigma, name=\"log_gdp_std\")\n",
    "\n",
    "    return tfd.JointDistributionCoroutineAutoBatched(_generator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "jd_81 = model_81(dd.rugged_std.values, scale_a=0.1, scale_b=0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "jd_81_pd = jd_81.experimental_pin(log_gdp_std=dd.log_gdp_std.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "def target_log_prob_81(*states):\n",
    "    return jd_81_pd.unnormalized_log_prob(states)\n",
    "\n",
    "def sample_post_81(init_state, seed):\n",
    "    kernel = tfp.mcmc.NoUTurnSampler(target_log_prob_81, 1e-3)\n",
    "  \n",
    "    return tfp.mcmc.sample_chain(\n",
    "        num_results=1000,\n",
    "        current_state=init_state,\n",
    "        kernel=kernel,\n",
    "        num_burnin_steps=1500,\n",
    "        trace_fn=None, seed=seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_chains = 4\n",
    "init_state_81 = jd_81_pd.sample_unpinned(num_chains, seed=key())\n",
    "post_81 = sample_post_81(init_state_81, seed=key())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "post_81_dict = {k: jnp.moveaxis(v, 0, 1) for k, v in post_81._asdict().items()}\n",
    "dict_81 = dict(posterior=post_81_dict)\n",
    "idata_81 = az.from_dict(**dict_81)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>sd</th>\n",
       "      <th>hdi_3%</th>\n",
       "      <th>hdi_97%</th>\n",
       "      <th>mcse_mean</th>\n",
       "      <th>mcse_sd</th>\n",
       "      <th>ess_bulk</th>\n",
       "      <th>ess_tail</th>\n",
       "      <th>r_hat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>1.000</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.980</td>\n",
       "      <td>1.019</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>3185.0</td>\n",
       "      <td>2563.0</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>0.005</td>\n",
       "      <td>0.056</td>\n",
       "      <td>-0.096</td>\n",
       "      <td>0.113</td>\n",
       "      <td>0.002</td>\n",
       "      <td>0.002</td>\n",
       "      <td>552.0</td>\n",
       "      <td>737.0</td>\n",
       "      <td>1.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sigma</th>\n",
       "      <td>0.138</td>\n",
       "      <td>0.008</td>\n",
       "      <td>0.124</td>\n",
       "      <td>0.153</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>3437.0</td>\n",
       "      <td>2206.0</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        mean     sd  hdi_3%  hdi_97%  mcse_mean  mcse_sd  ess_bulk  ess_tail  \\\n",
       "a      1.000  0.011   0.980    1.019      0.000    0.000    3185.0    2563.0   \n",
       "b      0.005  0.056  -0.096    0.113      0.002    0.002     552.0     737.0   \n",
       "sigma  0.138  0.008   0.124    0.153      0.000    0.000    3437.0    2206.0   \n",
       "\n",
       "       r_hat  \n",
       "a       1.00  \n",
       "b       1.01  \n",
       "sigma   1.00  "
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.summary(idata_81)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 8.1.2 Adding an indicator variable isn't enough"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.7"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "# make variable to index Africa (0) or not (1)\n",
    "dd[\"cid\"] = np.where(dd.cont_africa.values == 1, 0, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_82(rugged_std, cid):\n",
    "    def _generator():\n",
    "        a = yield tfd.Normal(loc=[1.0, 1.0], scale=0.1, name=\"a\")\n",
    "        b = yield tfd.Normal(loc=0.0, scale=0.3, name=\"b\")\n",
    "        sigma = yield tfd.Exponential(rate=1.0, name=\"sigma\")\n",
    "        mu = jnp.take(a, cid) + b * (rugged_std - 0.215)\n",
    "        yield tfd.Normal(loc=mu, scale=sigma, name=\"log_gdp_std\")\n",
    "\n",
    "    return tfd.JointDistributionCoroutineAutoBatched(_generator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "jd_82 = model_82(dd.rugged_std.values, dd.cid.values)\n",
    "jd_82_pd = jd_82.experimental_pin(log_gdp_std=dd.log_gdp_std.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "def target_log_prob_82(*states):\n",
    "    return jd_82_pd.unnormalized_log_prob(states)\n",
    "\n",
    "def sample_post_82(init_state, seed):\n",
    "    kernel = tfp.mcmc.NoUTurnSampler(target_log_prob_82, 1e-3)\n",
    "  \n",
    "    return tfp.mcmc.sample_chain(\n",
    "        num_results=1000,\n",
    "        current_state=init_state,\n",
    "        kernel=kernel,\n",
    "        num_burnin_steps=1500,\n",
    "        trace_fn=None, seed=seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_chains = 4\n",
    "init_state_82 = jd_82_pd.sample_unpinned(num_chains, seed=key())\n",
    "post_82 = sample_post_82(init_state_82, seed=key())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "post_82_dict = {k: jnp.moveaxis(v, 0, 1) for k, v in post_82._asdict().items()}\n",
    "dict_82 = dict(posterior=post_82_dict)\n",
    "idata_82 = az.from_dict(**dict_82)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>sd</th>\n",
       "      <th>hdi_3%</th>\n",
       "      <th>hdi_97%</th>\n",
       "      <th>mcse_mean</th>\n",
       "      <th>mcse_sd</th>\n",
       "      <th>ess_bulk</th>\n",
       "      <th>ess_tail</th>\n",
       "      <th>r_hat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a[0]</th>\n",
       "      <td>0.880</td>\n",
       "      <td>0.016</td>\n",
       "      <td>0.850</td>\n",
       "      <td>0.909</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1980.0</td>\n",
       "      <td>2118.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>a[1]</th>\n",
       "      <td>1.049</td>\n",
       "      <td>0.010</td>\n",
       "      <td>1.031</td>\n",
       "      <td>1.069</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>3553.0</td>\n",
       "      <td>3016.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>-0.048</td>\n",
       "      <td>0.045</td>\n",
       "      <td>-0.132</td>\n",
       "      <td>0.032</td>\n",
       "      <td>0.002</td>\n",
       "      <td>0.001</td>\n",
       "      <td>651.0</td>\n",
       "      <td>933.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sigma</th>\n",
       "      <td>0.114</td>\n",
       "      <td>0.006</td>\n",
       "      <td>0.102</td>\n",
       "      <td>0.126</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>3992.0</td>\n",
       "      <td>2389.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        mean     sd  hdi_3%  hdi_97%  mcse_mean  mcse_sd  ess_bulk  ess_tail  \\\n",
       "a[0]   0.880  0.016   0.850    0.909      0.000    0.000    1980.0    2118.0   \n",
       "a[1]   1.049  0.010   1.031    1.069      0.000    0.000    3553.0    3016.0   \n",
       "b     -0.048  0.045  -0.132    0.032      0.002    0.001     651.0     933.0   \n",
       "sigma  0.114  0.006   0.102    0.126      0.000    0.000    3992.0    2389.0   \n",
       "\n",
       "       r_hat  \n",
       "a[0]     1.0  \n",
       "a[1]     1.0  \n",
       "b        1.0  \n",
       "sigma    1.0  "
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.summary(idata_82)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:5 out of the last 5 calls to <function pfor.<locals>.f at 0x356874720> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n"
     ]
    }
   ],
   "source": [
    "jd_81n = model_81n(rugged_std=dd.rugged_std.values, scale_a=0.1, scale_b=0.3)\n",
    "ds_81, _ = jd_81n.sample_distributions(value=post_81)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'a': 0, 'b': 1, 'sigma': 2, 'log_gdp_std': 3}"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ds_81._field_name_to_index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "ll_81 = ds_81[3].log_prob(dd.log_gdp_std.values)\n",
    "ll_81 = np.moveaxis(ll_81, 0, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "dict_81.update(log_likelihood={'ll': ll_81},\n",
    "               dims={\"ll\": [\"obs\"]},\n",
    "               coords={\"obs\": np.arange(len(dd.log_gdp_std.values))})\n",
    "idata_81 = az.from_dict(**dict_81)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Computed from 4000 posterior samples and 170 observations log-likelihood matrix.\n",
       "\n",
       "              Estimate       SE\n",
       "deviance_waic  -188.86    12.96\n",
       "p_waic            2.58        -"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.waic(idata_81, scale='deviance')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_82n(rugged_std, cid):\n",
    "    def _generator():\n",
    "        a = yield tfdx.Normal(loc=[1.0, 1.0], scale=0.1, name=\"a\")\n",
    "        b = yield tfdx.Normal(loc=0.0, scale=0.3, name=\"b\")\n",
    "        sigma = yield tfdx.Exponential(rate=1.0, name=\"sigma\")\n",
    "        # np.take, jnp.take not working\n",
    "        mu = tf.gather(a, cid) + b * (rugged_std - 0.215)\n",
    "        yield tfdx.Normal(loc=mu, scale=sigma, name=\"log_gdp_std\")\n",
    "\n",
    "    return tfdx.JointDistributionCoroutineAutoBatched(_generator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tf functions work only with tf tensors\n",
    "rugged_std_tensor = tf.convert_to_tensor(dd.rugged_std.values, dtype=tf.float32)\n",
    "cid_tensor = tf.convert_to_tensor(dd.cid.values, dtype=tf.int32)\n",
    "\n",
    "jd_82n = model_82n(rugged_std_tensor, cid_tensor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_82, _ = jd_82n.sample_distributions(value=post_82)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'a': 0, 'b': 1, 'sigma': 2, 'log_gdp_std': 3}"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ds_82._field_name_to_index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [],
   "source": [
    "ll_82 = ds_82[3].log_prob(dd.log_gdp_std.values)\n",
    "ll_82 = np.moveaxis(ll_82, 0, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [],
   "source": [
    "dict_82.update(log_likelihood={'ll': ll_82},\n",
    "               dims={\"ll\": [\"obs\"]},\n",
    "               coords={\"obs\": np.arange(len(dd.log_gdp_std.values))})\n",
    "idata_82 = az.from_dict(**dict_82)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/jun/miniforge3/envs/dl/lib/python3.12/site-packages/arviz/stats/stats.py:1632: UserWarning: For one or more samples the posterior variance of the log predictive densities exceeds 0.4. This could be indication of WAIC starting to fail. \n",
      "See http://arxiv.org/abs/1507.04544 for details\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Computed from 4000 posterior samples and 170 observations log-likelihood matrix.\n",
       "\n",
       "              Estimate       SE\n",
       "deviance_waic  -252.57    14.80\n",
       "p_waic            3.99        -\n",
       "\n",
       "There has been a warning during the calculation. Please check the results."
      ]
     },
     "execution_count": 155,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.waic(idata_82, scale='deviance')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/jun/miniforge3/envs/dl/lib/python3.12/site-packages/arviz/stats/stats.py:1632: UserWarning: For one or more samples the posterior variance of the log predictive densities exceeds 0.4. This could be indication of WAIC starting to fail. \n",
      "See http://arxiv.org/abs/1507.04544 for details\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>rank</th>\n",
       "      <th>elpd_waic</th>\n",
       "      <th>p_waic</th>\n",
       "      <th>elpd_diff</th>\n",
       "      <th>weight</th>\n",
       "      <th>se</th>\n",
       "      <th>dse</th>\n",
       "      <th>warning</th>\n",
       "      <th>scale</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>m8.2</th>\n",
       "      <td>0</td>\n",
       "      <td>-252.56914</td>\n",
       "      <td>3.986704</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.970202</td>\n",
       "      <td>14.802634</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>True</td>\n",
       "      <td>deviance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>m8.1</th>\n",
       "      <td>1</td>\n",
       "      <td>-188.85666</td>\n",
       "      <td>2.578659</td>\n",
       "      <td>63.71248</td>\n",
       "      <td>0.029798</td>\n",
       "      <td>12.956697</td>\n",
       "      <td>14.620128</td>\n",
       "      <td>False</td>\n",
       "      <td>deviance</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      rank  elpd_waic    p_waic  elpd_diff    weight         se        dse  \\\n",
       "m8.2     0 -252.56914  3.986704    0.00000  0.970202  14.802634   0.000000   \n",
       "m8.1     1 -188.85666  2.578659   63.71248  0.029798  12.956697  14.620128   \n",
       "\n",
       "      warning     scale  \n",
       "m8.2     True  deviance  \n",
       "m8.1    False  deviance  "
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.compare({\"m8.1\": idata_81, \"m8.2\": idata_82}, ic=\"waic\", scale=\"deviance\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.11"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>sd</th>\n",
       "      <th>hdi_6%</th>\n",
       "      <th>hdi_94%</th>\n",
       "      <th>mcse_mean</th>\n",
       "      <th>mcse_sd</th>\n",
       "      <th>ess_bulk</th>\n",
       "      <th>ess_tail</th>\n",
       "      <th>r_hat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>-0.169</td>\n",
       "      <td>0.019</td>\n",
       "      <td>-0.199</td>\n",
       "      <td>-0.14</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2548.0</td>\n",
       "      <td>1955.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    mean     sd  hdi_6%  hdi_94%  mcse_mean  mcse_sd  ess_bulk  ess_tail  \\\n",
       "a -0.169  0.019  -0.199    -0.14        0.0      0.0    2548.0    1955.0   \n",
       "\n",
       "   r_hat  \n",
       "a    1.0  "
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.summary(idata_82.posterior['a'][..., 0] - idata_82.posterior['a'][..., 1], hdi_prob=0.88)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.13"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_83(rugged_std, cid):\n",
    "    def _generator():\n",
    "        a = yield tfd.Normal(loc=[1.0, 1.0], scale=0.1, name=\"a\")\n",
    "        b = yield tfd.Normal(loc=[0.0, 0.0], scale=0.3, name=\"b\")\n",
    "        sigma = yield tfd.Exponential(rate=1.0, name=\"sigma\")\n",
    "        mu = jnp.take(a, cid) + jnp.take(b, cid) * (rugged_std - 0.215)\n",
    "        yield tfd.Normal(loc=mu, scale=sigma, name=\"log_gdp_std\")\n",
    "\n",
    "    return tfd.JointDistributionCoroutineAutoBatched(_generator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [],
   "source": [
    "jd_83 = model_83(dd.rugged_std.values, dd.cid.values)\n",
    "jd_83_pd = jd_83.experimental_pin(log_gdp_std=dd.log_gdp_std.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [],
   "source": [
    "def target_log_prob_83(*states):\n",
    "    return jd_83_pd.unnormalized_log_prob(states)\n",
    "\n",
    "def sample_post_83(init_state, seed):\n",
    "    kernel = tfp.mcmc.NoUTurnSampler(target_log_prob_83, 1e-3)\n",
    "  \n",
    "    return tfp.mcmc.sample_chain(\n",
    "        num_results=1000,\n",
    "        current_state=init_state,\n",
    "        kernel=kernel,\n",
    "        num_burnin_steps=1500,\n",
    "        trace_fn=None, seed=seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_chains = 4\n",
    "init_state_83 = jd_83_pd.sample_unpinned(num_chains, seed=key())\n",
    "post_83 = sample_post_83(init_state_83, seed=key())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [],
   "source": [
    "post_83_dict = {k: jnp.moveaxis(v, 0, 1) for k, v in post_83._asdict().items()}\n",
    "dict_83 = dict(posterior=post_83_dict)\n",
    "idata_83 = az.from_dict(**dict_83)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>sd</th>\n",
       "      <th>hdi_3%</th>\n",
       "      <th>hdi_97%</th>\n",
       "      <th>mcse_mean</th>\n",
       "      <th>mcse_sd</th>\n",
       "      <th>ess_bulk</th>\n",
       "      <th>ess_tail</th>\n",
       "      <th>r_hat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a[0]</th>\n",
       "      <td>0.887</td>\n",
       "      <td>0.016</td>\n",
       "      <td>0.858</td>\n",
       "      <td>0.918</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>3381.0</td>\n",
       "      <td>2625.0</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>a[1]</th>\n",
       "      <td>1.050</td>\n",
       "      <td>0.010</td>\n",
       "      <td>1.032</td>\n",
       "      <td>1.071</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>3729.0</td>\n",
       "      <td>2410.0</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b[0]</th>\n",
       "      <td>0.129</td>\n",
       "      <td>0.075</td>\n",
       "      <td>-0.008</td>\n",
       "      <td>0.273</td>\n",
       "      <td>0.002</td>\n",
       "      <td>0.002</td>\n",
       "      <td>1094.0</td>\n",
       "      <td>1522.0</td>\n",
       "      <td>1.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b[1]</th>\n",
       "      <td>-0.141</td>\n",
       "      <td>0.056</td>\n",
       "      <td>-0.241</td>\n",
       "      <td>-0.033</td>\n",
       "      <td>0.002</td>\n",
       "      <td>0.001</td>\n",
       "      <td>1228.0</td>\n",
       "      <td>891.0</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sigma</th>\n",
       "      <td>0.112</td>\n",
       "      <td>0.006</td>\n",
       "      <td>0.100</td>\n",
       "      <td>0.123</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>4184.0</td>\n",
       "      <td>2817.0</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        mean     sd  hdi_3%  hdi_97%  mcse_mean  mcse_sd  ess_bulk  ess_tail  \\\n",
       "a[0]   0.887  0.016   0.858    0.918      0.000    0.000    3381.0    2625.0   \n",
       "a[1]   1.050  0.010   1.032    1.071      0.000    0.000    3729.0    2410.0   \n",
       "b[0]   0.129  0.075  -0.008    0.273      0.002    0.002    1094.0    1522.0   \n",
       "b[1]  -0.141  0.056  -0.241   -0.033      0.002    0.001    1228.0     891.0   \n",
       "sigma  0.112  0.006   0.100    0.123      0.000    0.000    4184.0    2817.0   \n",
       "\n",
       "       r_hat  \n",
       "a[0]    1.00  \n",
       "a[1]    1.00  \n",
       "b[0]    1.01  \n",
       "b[1]    1.00  \n",
       "sigma   1.00  "
      ]
     },
     "execution_count": 186,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.summary(idata_83)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.15"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_83n(rugged_std, cid):\n",
    "    def _generator():\n",
    "        a = yield tfdx.Normal(loc=[1.0, 1.0], scale=0.1, name=\"a\")\n",
    "        b = yield tfdx.Normal(loc=[0.0, 0.0], scale=0.3, name=\"b\")\n",
    "        sigma = yield tfdx.Exponential(rate=1.0, name=\"sigma\")\n",
    "        mu = tf.gather(a, cid) + tf.gather(b, cid) * (rugged_std - 0.215)\n",
    "        yield tfdx.Normal(loc=mu, scale=sigma, name=\"log_gdp_std\")\n",
    "\n",
    "    return tfdx.JointDistributionCoroutineAutoBatched(_generator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {},
   "outputs": [],
   "source": [
    "rugged_std_tensor = tf.convert_to_tensor(dd.rugged_std.values, dtype=tf.float32)\n",
    "cid_tensor = tf.convert_to_tensor(dd.cid.values, dtype=tf.int32)\n",
    "\n",
    "jd_83n = model_83n(rugged_std_tensor, cid_tensor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_83, _ = jd_83n.sample_distributions(value=post_83)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'a': 0, 'b': 1, 'sigma': 2, 'log_gdp_std': 3}"
      ]
     },
     "execution_count": 176,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ds_83._field_name_to_index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "metadata": {},
   "outputs": [],
   "source": [
    "ll_83 = ds_82[3].log_prob(dd.log_gdp_std.values)\n",
    "ll_83 = np.moveaxis(ll_83, 0, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {},
   "outputs": [],
   "source": [
    "dict_83.update(log_likelihood={'ll': ll_83},\n",
    "               dims={\"ll\": [\"obs\"]},\n",
    "               coords={\"obs\": np.arange(len(dd.log_gdp_std.values))})\n",
    "idata_83 = az.from_dict(**dict_83)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>rank</th>\n",
       "      <th>elpd_loo</th>\n",
       "      <th>p_loo</th>\n",
       "      <th>elpd_diff</th>\n",
       "      <th>weight</th>\n",
       "      <th>se</th>\n",
       "      <th>dse</th>\n",
       "      <th>warning</th>\n",
       "      <th>scale</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>m8.2</th>\n",
       "      <td>0</td>\n",
       "      <td>-252.532418</td>\n",
       "      <td>4.005064</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.485076</td>\n",
       "      <td>14.810944</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>False</td>\n",
       "      <td>deviance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>m8.3</th>\n",
       "      <td>1</td>\n",
       "      <td>-252.532291</td>\n",
       "      <td>4.005128</td>\n",
       "      <td>0.000127</td>\n",
       "      <td>0.484847</td>\n",
       "      <td>14.810975</td>\n",
       "      <td>0.000078</td>\n",
       "      <td>False</td>\n",
       "      <td>deviance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>m8.1</th>\n",
       "      <td>2</td>\n",
       "      <td>-188.846403</td>\n",
       "      <td>2.583787</td>\n",
       "      <td>63.686015</td>\n",
       "      <td>0.030076</td>\n",
       "      <td>12.957343</td>\n",
       "      <td>14.626487</td>\n",
       "      <td>False</td>\n",
       "      <td>deviance</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      rank    elpd_loo     p_loo  elpd_diff    weight         se        dse  \\\n",
       "m8.2     0 -252.532418  4.005064   0.000000  0.485076  14.810944   0.000000   \n",
       "m8.3     1 -252.532291  4.005128   0.000127  0.484847  14.810975   0.000078   \n",
       "m8.1     2 -188.846403  2.583787  63.686015  0.030076  12.957343  14.626487   \n",
       "\n",
       "      warning     scale  \n",
       "m8.2    False  deviance  \n",
       "m8.3    False  deviance  \n",
       "m8.1    False  deviance  "
      ]
     },
     "execution_count": 190,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.compare({\"m8.1\": idata_81, \"m8.2\": idata_82, \"m8.3\": idata_83}, ic='loo', scale=\"deviance\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Computed from 4000 posterior samples and 170 observations log-likelihood matrix.\n",
       "\n",
       "         Estimate       SE\n",
       "elpd_loo   126.27     7.41\n",
       "p_loo        4.01        -\n",
       "------\n",
       "\n",
       "Pareto k diagnostic values:\n",
       "                         Count   Pct.\n",
       "(-Inf, 0.5]   (good)      170  100.0%\n",
       " (0.5, 0.7]   (ok)          0    0.0%\n",
       "   (0.7, 1]   (bad)         0    0.0%\n",
       "   (1, Inf)   (very bad)    0    0.0%"
      ]
     },
     "execution_count": 193,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.loo(idata_83)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 8.1.4 Plotting the interaction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.17"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "metadata": {},
   "outputs": [],
   "source": [
    "rugged_seq = np.linspace(start=-1, stop=1.1, num=30)\n",
    "\n",
    "# Africa, cid=0\n",
    "jd_83_cid2 = model_83n(rugged_std=rugged_seq, cid=np.repeat(0, 30))\n",
    "ds_cid2, _ = jd_83_cid2.sample_distributions(value=post_83)\n",
    "Africa_mu = np.mean(ds_cid2[-1].loc, axis=(0, 1))\n",
    "Africa_ci = np.percentile(ds_cid2[-1].loc, q=(1.5, 98.5), axis=(0, 1))\n",
    "\n",
    "# NotAfrica, cid=1\n",
    "jd_83_cid1 = model_83n(rugged_std=rugged_seq, cid=np.repeat(1, 30))\n",
    "ds_cid1, _ = jd_83_cid1.sample_distributions(value=post_83)\n",
    "notAfrica_mu = np.mean(ds_cid1[-1].loc, axis=(0, 1))\n",
    "notAfrica_ci = np.percentile(ds_cid1[-1].loc, q=(1.5, 98.5), axis=(0, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Africa_idx = dd['cid'] == 0\n",
    "notAfrica_idx = dd['cid'] == 1\n",
    "\n",
    "fig, ax = plt.subplots(1, 2, sharex=True, sharey=True, figsize=(12, 6))\n",
    "\n",
    "ax[0].scatter(dd['rugged_std'][Africa_idx], dd['log_gdp_std'][Africa_idx])\n",
    "ax[0].plot(rugged_seq, Africa_mu)\n",
    "ax[0].fill_between(rugged_seq, Africa_ci[0], Africa_ci[1], alpha=0.2)\n",
    "ax[0].set_xlim(0.0, 1.0)\n",
    "\n",
    "ax[1].scatter(dd['rugged_std'][notAfrica_idx], dd['log_gdp_std'][notAfrica_idx])\n",
    "ax[1].plot(rugged_seq, notAfrica_mu)\n",
    "ax[1].fill_between(rugged_seq, notAfrica_ci[0], notAfrica_ci[1], alpha=0.2);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 8.3 Continuous interactions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 8.3.1 A winter flower"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.19"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 27 entries, 0 to 26\n",
      "Data columns (total 4 columns):\n",
      " #   Column  Non-Null Count  Dtype  \n",
      "---  ------  --------------  -----  \n",
      " 0   bed     27 non-null     object \n",
      " 1   water   27 non-null     int64  \n",
      " 2   shade   27 non-null     int64  \n",
      " 3   blooms  27 non-null     float64\n",
      "dtypes: float64(1), int64(2), object(1)\n",
      "memory usage: 996.0+ bytes\n"
     ]
    }
   ],
   "source": [
    "d = pd.read_csv(\"./data/tulips.csv\", sep=\";\")\n",
    "d.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>bed</th>\n",
       "      <th>water</th>\n",
       "      <th>shade</th>\n",
       "      <th>blooms</th>\n",
       "      <th>blooms_std</th>\n",
       "      <th>water_cent</th>\n",
       "      <th>shade_cent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>111.04</td>\n",
       "      <td>0.307029</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>183.47</td>\n",
       "      <td>0.507300</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>59.16</td>\n",
       "      <td>0.163579</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  bed  water  shade  blooms  blooms_std  water_cent  shade_cent\n",
       "0   a      1      1    0.00    0.000000        -1.0        -1.0\n",
       "1   a      1      2    0.00    0.000000        -1.0         0.0\n",
       "2   a      1      3  111.04    0.307029        -1.0         1.0\n",
       "3   a      2      1  183.47    0.507300         0.0        -1.0\n",
       "4   a      2      2   59.16    0.163579         0.0         0.0"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d[\"blooms_std\"] = d.blooms / d.blooms.max()\n",
    "d[\"water_cent\"] = d.water - d.water.mean()\n",
    "d[\"shade_cent\"] = d.shade - d.shade.mean()\n",
    "d.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.21"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "float"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(1e4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Array(0.6163, dtype=float32)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a = tfd.Normal(loc=0.5, scale=1.0).sample(10000, seed=key())\n",
    "np.sum((a < 0) | (a > 1)) / a.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.22"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Array(0.0455, dtype=float32)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a = tfd.Normal(loc=0.5, scale=0.25).sample(10000, seed=key())\n",
    "np.sum((a < 0) | (a > 1)) / a.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.23"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_84(water_cent, shade_cent):\n",
    "    def _generator():\n",
    "        a = yield tfd.Normal(loc=0.5, scale=0.25, name=\"a\")\n",
    "        bw = yield tfd.Normal(loc=0.0, scale=0.25, name=\"bw\")\n",
    "        bs = yield tfd.Normal(loc=0.0, scale=0.25, name=\"bs\")\n",
    "        sigma = yield tfd.Exponential(rate=1.0, name=\"sigma\")\n",
    "        mu = a + bw * water_cent + bs * shade_cent\n",
    "        yield tfd.Normal(loc=mu, scale=sigma, name=\"blooms_std\")\n",
    "\n",
    "    return tfd.JointDistributionCoroutineAutoBatched(_generator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "jd_84 = model_84(d.water_cent.values, d.shade_cent.values)\n",
    "jd_84_pd = jd_84.experimental_pin(blooms_std=d.blooms_std.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def target_log_prob_84(*states):\n",
    "    return jd_84_pd.unnormalized_log_prob(states)\n",
    "\n",
    "def sample_post_84(init_state, seed):\n",
    "    kernel = tfp.mcmc.NoUTurnSampler(target_log_prob_84, 1e-3)\n",
    "  \n",
    "    return tfp.mcmc.sample_chain(\n",
    "        num_results=1000,\n",
    "        current_state=init_state,\n",
    "        kernel=kernel,\n",
    "        num_burnin_steps=1500,\n",
    "        trace_fn=None, seed=seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_chains = 4\n",
    "init_state_84 = jd_84_pd.sample_unpinned(num_chains, seed=key())\n",
    "post_84 = sample_post_84(init_state_84, seed=key())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "post_84_dict = {k: jnp.moveaxis(v, 0, 1) for k, v in post_84._asdict().items()}\n",
    "dict_84 = dict(posterior=post_84_dict)\n",
    "idata_84 = az.from_dict(**dict_84)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>sd</th>\n",
       "      <th>hdi_3%</th>\n",
       "      <th>hdi_97%</th>\n",
       "      <th>mcse_mean</th>\n",
       "      <th>mcse_sd</th>\n",
       "      <th>ess_bulk</th>\n",
       "      <th>ess_tail</th>\n",
       "      <th>r_hat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>0.359</td>\n",
       "      <td>0.034</td>\n",
       "      <td>0.296</td>\n",
       "      <td>0.426</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.000</td>\n",
       "      <td>3233.0</td>\n",
       "      <td>2845.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bs</th>\n",
       "      <td>-0.113</td>\n",
       "      <td>0.041</td>\n",
       "      <td>-0.187</td>\n",
       "      <td>-0.034</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>2409.0</td>\n",
       "      <td>2583.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bw</th>\n",
       "      <td>0.204</td>\n",
       "      <td>0.042</td>\n",
       "      <td>0.125</td>\n",
       "      <td>0.282</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>2248.0</td>\n",
       "      <td>2085.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sigma</th>\n",
       "      <td>0.176</td>\n",
       "      <td>0.027</td>\n",
       "      <td>0.129</td>\n",
       "      <td>0.229</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.000</td>\n",
       "      <td>2895.0</td>\n",
       "      <td>2515.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        mean     sd  hdi_3%  hdi_97%  mcse_mean  mcse_sd  ess_bulk  ess_tail  \\\n",
       "a      0.359  0.034   0.296    0.426      0.001    0.000    3233.0    2845.0   \n",
       "bs    -0.113  0.041  -0.187   -0.034      0.001    0.001    2409.0    2583.0   \n",
       "bw     0.204  0.042   0.125    0.282      0.001    0.001    2248.0    2085.0   \n",
       "sigma  0.176  0.027   0.129    0.229      0.001    0.000    2895.0    2515.0   \n",
       "\n",
       "       r_hat  \n",
       "a        1.0  \n",
       "bs       1.0  \n",
       "bw       1.0  \n",
       "sigma    1.0  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.summary(idata_84)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.24"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_85(water_cent, shade_cent):\n",
    "    def _generator():\n",
    "        a = yield tfd.Normal(loc=0.5, scale=0.25, name=\"a\")\n",
    "        bw = yield tfd.Normal(loc=0.0, scale=0.25, name=\"bw\")\n",
    "        bs = yield tfd.Normal(loc=0.0, scale=0.25, name=\"bs\")\n",
    "        bws = yield tfd.Normal(loc=0.0, scale=0.25, name=\"bws\")\n",
    "        sigma = yield tfd.Exponential(rate=1.0, name=\"sigma\")\n",
    "        mu = a + bw * water_cent + bs * shade_cent + bws * water_cent*shade_cent\n",
    "        yield tfd.Normal(loc=mu, scale=sigma, name=\"blooms_std\")\n",
    "\n",
    "    return tfd.JointDistributionCoroutineAutoBatched(_generator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "jd_85 = model_85(d.water_cent.values, d.shade_cent.values)\n",
    "jd_85_pd = jd_85.experimental_pin(blooms_std=d.blooms_std.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def target_log_prob_85(*states):\n",
    "    return jd_85_pd.unnormalized_log_prob(states)\n",
    "\n",
    "def sample_post_85(init_state, seed):\n",
    "    kernel = tfp.mcmc.NoUTurnSampler(target_log_prob_85, 1e-3)\n",
    "  \n",
    "    return tfp.mcmc.sample_chain(\n",
    "        num_results=1000,\n",
    "        current_state=init_state,\n",
    "        kernel=kernel,\n",
    "        num_burnin_steps=1500,\n",
    "        trace_fn=None, seed=seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_chains = 4\n",
    "init_state_85 = jd_85_pd.sample_unpinned(num_chains, seed=key())\n",
    "post_85 = sample_post_85(init_state_85, seed=key())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "post_85_dict = {k: jnp.moveaxis(v, 0, 1) for k, v in post_85._asdict().items()}\n",
    "dict_85 = dict(posterior=post_85_dict)\n",
    "idata_85 = az.from_dict(**dict_85)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>sd</th>\n",
       "      <th>hdi_3%</th>\n",
       "      <th>hdi_97%</th>\n",
       "      <th>mcse_mean</th>\n",
       "      <th>mcse_sd</th>\n",
       "      <th>ess_bulk</th>\n",
       "      <th>ess_tail</th>\n",
       "      <th>r_hat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>0.359</td>\n",
       "      <td>0.027</td>\n",
       "      <td>0.308</td>\n",
       "      <td>0.411</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>3530.0</td>\n",
       "      <td>2207.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bs</th>\n",
       "      <td>-0.113</td>\n",
       "      <td>0.033</td>\n",
       "      <td>-0.172</td>\n",
       "      <td>-0.049</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.000</td>\n",
       "      <td>3103.0</td>\n",
       "      <td>2597.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bw</th>\n",
       "      <td>0.206</td>\n",
       "      <td>0.034</td>\n",
       "      <td>0.145</td>\n",
       "      <td>0.272</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.000</td>\n",
       "      <td>2970.0</td>\n",
       "      <td>2133.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bws</th>\n",
       "      <td>-0.142</td>\n",
       "      <td>0.042</td>\n",
       "      <td>-0.220</td>\n",
       "      <td>-0.064</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>2376.0</td>\n",
       "      <td>2315.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sigma</th>\n",
       "      <td>0.142</td>\n",
       "      <td>0.022</td>\n",
       "      <td>0.103</td>\n",
       "      <td>0.182</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>3508.0</td>\n",
       "      <td>2665.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        mean     sd  hdi_3%  hdi_97%  mcse_mean  mcse_sd  ess_bulk  ess_tail  \\\n",
       "a      0.359  0.027   0.308    0.411      0.000    0.000    3530.0    2207.0   \n",
       "bs    -0.113  0.033  -0.172   -0.049      0.001    0.000    3103.0    2597.0   \n",
       "bw     0.206  0.034   0.145    0.272      0.001    0.000    2970.0    2133.0   \n",
       "bws   -0.142  0.042  -0.220   -0.064      0.001    0.001    2376.0    2315.0   \n",
       "sigma  0.142  0.022   0.103    0.182      0.000    0.000    3508.0    2665.0   \n",
       "\n",
       "       r_hat  \n",
       "a        1.0  \n",
       "bs       1.0  \n",
       "bw       1.0  \n",
       "bws      1.0  \n",
       "sigma    1.0  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.summary(idata_85)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 8.3.3 Plotting posterior predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Code 8.25"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tfpx version\n",
    "def model_84x(water_cent, shade_cent):\n",
    "    def _generator():\n",
    "        a = yield tfdx.Normal(loc=0.5, scale=0.25, name=\"a\")\n",
    "        bw = yield tfdx.Normal(loc=0.0, scale=0.25, name=\"bw\")\n",
    "        bs = yield tfdx.Normal(loc=0.0, scale=0.25, name=\"bs\")\n",
    "        sigma = yield tfdx.Exponential(rate=1.0, name=\"sigma\")\n",
    "        mu = a + bw * water_cent + bs * shade_cent\n",
    "        yield tfdx.Normal(loc=mu, scale=sigma, name=\"blooms_std\")\n",
    "\n",
    "    return tfdx.JointDistributionCoroutineAutoBatched(_generator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x300 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_, axes = plt.subplots(1, 3, figsize=(9, 3), sharey=True)\n",
    "for ax, s in zip(axes, np.array([-1., 0., 1.])):\n",
    "    idx = d.shade_cent.values == s\n",
    "    ax.scatter(d.water_cent.values[idx], d.blooms_std.values[idx])\n",
    "    ax.set(xlim=(-1.1, 1.1), ylim=(-0.1, 1.1), xlabel=\"water\", ylabel=\"blooms\")\n",
    "\n",
    "    shade_cent = s\n",
    "    water_cent = np.array([-1., 0., 1.])\n",
    "\n",
    "    jd_84x = model_84x(water_cent, shade_cent)\n",
    "    ds, _ = jd_84x.sample_distributions(value=post_84)\n",
    "    # ds[-1].loc: (draw, chain, obs) --> (1000, 4, 3)\n",
    "    # mean through chains\n",
    "    mu = ds[-1].loc.numpy().mean(axis=1) \n",
    "\n",
    "    for i in range(20):\n",
    "        ax.plot(range(-1, 2), mu[i], \"k\", alpha=0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_85x(water_cent, shade_cent):\n",
    "    def _generator():\n",
    "        a = yield tfdx.Normal(loc=0.5, scale=0.25, name=\"a\")\n",
    "        bw = yield tfdx.Normal(loc=0.0, scale=0.25, name=\"bw\")\n",
    "        bs = yield tfdx.Normal(loc=0.0, scale=0.25, name=\"bs\")\n",
    "        bws = yield tfdx.Normal(loc=0.0, scale=0.25, name=\"bws\")\n",
    "        sigma = yield tfdx.Exponential(rate=1.0, name=\"sigma\")\n",
    "        mu = a + bw * water_cent + bs * shade_cent + bws * water_cent*shade_cent\n",
    "        yield tfdx.Normal(loc=mu, scale=sigma, name=\"blooms_std\")\n",
    "\n",
    "    return tfdx.JointDistributionCoroutineAutoBatched(_generator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x300 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "_, axes = plt.subplots(1, 3, figsize=(9, 3), sharey=True)\n",
    "for ax, s in zip(axes, np.array([-1., 0., 1.])):\n",
    "    idx = d.shade_cent.values == s\n",
    "    ax.scatter(d.water_cent.values[idx], d.blooms_std.values[idx])\n",
    "    ax.set(xlim=(-1.1, 1.1), ylim=(-0.1, 1.1), xlabel=\"water\", ylabel=\"blooms\")\n",
    "\n",
    "    shade_cent = s\n",
    "    water_cent = np.array([-1., 0., 1.])\n",
    "\n",
    "    jd_85x = model_85x(water_cent, shade_cent)\n",
    "    ds, _ = jd_85x.sample_distributions(value=post_85)\n",
    "    # ds[-1].loc: (draw, chain, obs) --> (1000, 4, 3)\n",
    "    # mean through chains --> (1000, 3)\n",
    "    mu = ds[-1].loc.numpy().mean(axis=1) \n",
    "\n",
    "    for i in range(20):\n",
    "        ax.plot(range(-1, 2), mu[i], \"k\", alpha=0.3)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.4"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}