aflaxman/2020_11_11a_pymc_spline_example.ipynb

## 2020_11_11a_pymc_spline_example.ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wed Nov 11 13:45:27 PST 2020\r\n"
     ]
    }
   ],
   "source": [
    "import numpy as np, matplotlib.pyplot as plt, pandas as pd\n",
    "pd.set_option('display.max_rows', 8)\n",
    "!date\n",
    "\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Splines in PyMC3\n",
    "\n",
    "A [question on stack exchange](https://stats.stackexchange.com/questions/495745/spline-regression-via-pymc3) asks how to do splines in PyMC3 the way they are done in [a Stan case study](https://mc-stan.org/users/documentation/case-studies/splines_in_stan.html).\n",
    "\n",
    "I don't think I'll have time to do all of it, but I'll do a little:\n",
    "\n",
    "    library(\"splines\")\n",
    "    library(\"rstan\")\n",
    "    X <- seq(from=-5, to=5, by=.1) # generating inputs\n",
    "    B <- t(bs(X, knots=seq(-5,5,1), degree=3, intercept = TRUE)) # creating the B-splines\n",
    "    num_data <- length(X); num_basis <- nrow(B)\n",
    "    a0 <- 0.2 # intercept\n",
    "    a <- rnorm(num_basis, 0, 1) # coefficients of B-splines\n",
    "    Y_true <- as.vector(a0*X + a%*%B) # generating the output\n",
    "    Y <- Y_true + rnorm(length(X),0,.2) # adding noise\n",
    "    rstan_options(auto_write = TRUE)\n",
    "    options(mc.cores = parallel::detectCores())\n",
    "    sm<-stan_model(\"fit_basis.stan\")\n",
    "    fit<-sampling(sm,iter=500,control = list(adapt_delta=0.95))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING (theano.tensor.blas): Using NumPy C-API based implementation for BLAS functions.\n"
     ]
    }
   ],
   "source": [
    "import pymc3 as pm, theano.tensor as tt, patsy\n",
    "np.random.seed(12345) # set random seed for reproducibility"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Intercept</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[0]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[1]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[2]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[3]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[4]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[5]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[6]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[7]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[8]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[9]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[10]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[11]</th>\n",
       "      <th>bs(X, knots=np.arange(-5, 5, 1), degree=3)[12]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.729</td>\n",
       "      <td>0.25675</td>\n",
       "      <td>0.014083</td>\n",
       "      <td>0.000167</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.512</td>\n",
       "      <td>0.43400</td>\n",
       "      <td>0.052667</td>\n",
       "      <td>0.001333</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.343</td>\n",
       "      <td>0.54225</td>\n",
       "      <td>0.110250</td>\n",
       "      <td>0.004500</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.005445</td>\n",
       "      <td>0.126596</td>\n",
       "      <td>0.571663</td>\n",
       "      <td>0.296296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001613</td>\n",
       "      <td>0.060902</td>\n",
       "      <td>0.466977</td>\n",
       "      <td>0.470508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000202</td>\n",
       "      <td>0.016385</td>\n",
       "      <td>0.281082</td>\n",
       "      <td>0.702332</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Intercept  bs(X, knots=np.arange(-5, 5, 1), degree=3)[0]  \\\n",
       "0         1.0                                          1.000   \n",
       "1         1.0                                          0.729   \n",
       "2         1.0                                          0.512   \n",
       "3         1.0                                          0.343   \n",
       "..        ...                                            ...   \n",
       "96        1.0                                          0.000   \n",
       "97        1.0                                          0.000   \n",
       "98        1.0                                          0.000   \n",
       "99        1.0                                          0.000   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[1]  \\\n",
       "0                                         0.00000   \n",
       "1                                         0.25675   \n",
       "2                                         0.43400   \n",
       "3                                         0.54225   \n",
       "..                                            ...   \n",
       "96                                        0.00000   \n",
       "97                                        0.00000   \n",
       "98                                        0.00000   \n",
       "99                                        0.00000   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[2]  \\\n",
       "0                                        0.000000   \n",
       "1                                        0.014083   \n",
       "2                                        0.052667   \n",
       "3                                        0.110250   \n",
       "..                                            ...   \n",
       "96                                       0.000000   \n",
       "97                                       0.000000   \n",
       "98                                       0.000000   \n",
       "99                                       0.000000   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[3]  \\\n",
       "0                                        0.000000   \n",
       "1                                        0.000167   \n",
       "2                                        0.001333   \n",
       "3                                        0.004500   \n",
       "..                                            ...   \n",
       "96                                       0.000000   \n",
       "97                                       0.000000   \n",
       "98                                       0.000000   \n",
       "99                                       0.000000   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[4]  \\\n",
       "0                                             0.0   \n",
       "1                                             0.0   \n",
       "2                                             0.0   \n",
       "3                                             0.0   \n",
       "..                                            ...   \n",
       "96                                            0.0   \n",
       "97                                            0.0   \n",
       "98                                            0.0   \n",
       "99                                            0.0   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[5]  \\\n",
       "0                                             0.0   \n",
       "1                                             0.0   \n",
       "2                                             0.0   \n",
       "3                                             0.0   \n",
       "..                                            ...   \n",
       "96                                            0.0   \n",
       "97                                            0.0   \n",
       "98                                            0.0   \n",
       "99                                            0.0   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[6]  \\\n",
       "0                                             0.0   \n",
       "1                                             0.0   \n",
       "2                                             0.0   \n",
       "3                                             0.0   \n",
       "..                                            ...   \n",
       "96                                            0.0   \n",
       "97                                            0.0   \n",
       "98                                            0.0   \n",
       "99                                            0.0   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[7]  \\\n",
       "0                                             0.0   \n",
       "1                                             0.0   \n",
       "2                                             0.0   \n",
       "3                                             0.0   \n",
       "..                                            ...   \n",
       "96                                            0.0   \n",
       "97                                            0.0   \n",
       "98                                            0.0   \n",
       "99                                            0.0   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[8]  \\\n",
       "0                                             0.0   \n",
       "1                                             0.0   \n",
       "2                                             0.0   \n",
       "3                                             0.0   \n",
       "..                                            ...   \n",
       "96                                            0.0   \n",
       "97                                            0.0   \n",
       "98                                            0.0   \n",
       "99                                            0.0   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[9]  \\\n",
       "0                                        0.000000   \n",
       "1                                        0.000000   \n",
       "2                                        0.000000   \n",
       "3                                        0.000000   \n",
       "..                                            ...   \n",
       "96                                       0.005445   \n",
       "97                                       0.001613   \n",
       "98                                       0.000202   \n",
       "99                                       0.000000   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[10]  \\\n",
       "0                                         0.000000   \n",
       "1                                         0.000000   \n",
       "2                                         0.000000   \n",
       "3                                         0.000000   \n",
       "..                                             ...   \n",
       "96                                        0.126596   \n",
       "97                                        0.060902   \n",
       "98                                        0.016385   \n",
       "99                                        0.000000   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[11]  \\\n",
       "0                                         0.000000   \n",
       "1                                         0.000000   \n",
       "2                                         0.000000   \n",
       "3                                         0.000000   \n",
       "..                                             ...   \n",
       "96                                        0.571663   \n",
       "97                                        0.466977   \n",
       "98                                        0.281082   \n",
       "99                                        0.000000   \n",
       "\n",
       "    bs(X, knots=np.arange(-5, 5, 1), degree=3)[12]  \n",
       "0                                         0.000000  \n",
       "1                                         0.000000  \n",
       "2                                         0.000000  \n",
       "3                                         0.000000  \n",
       "..                                             ...  \n",
       "96                                        0.296296  \n",
       "97                                        0.470508  \n",
       "98                                        0.702332  \n",
       "99                                        1.000000  \n",
       "\n",
       "[100 rows x 14 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame({'X':np.arange(-5, 5, .1)})\n",
    "B = patsy.dmatrix('bs(X, knots=np.arange(-5,5,1), degree=3)', df, \n",
    "                  return_type='dataframe')\n",
    "B"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_data, num_basis = B.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "a0 = 0.2 # intercept\n",
    "a = np.random.normal(0, 1, size=num_basis) # coefficients of B-splines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_true = a0 + np.dot(B, a) # generating the output\n",
    "Y = Y_true + np.random.normal(0, .2, size=num_data) # adding noise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# this is supposed to make things easy, but I think I'm missing something\n",
    "\n",
    "# df['Y'] = Y\n",
    "\n",
    "# with pm.Model() as model:\n",
    "#     # specify glm and pass in data. The resulting linear model, its likelihood and\n",
    "#     # and all its parameters are automatically added to our model.\n",
    "#     glm = pm.glm.GLM.from_formula('Y ~ bs(X, knots=np.arange(-5,5,1), degree=3)', df,\n",
    "#                         priors={'Intercept': pm.Flat.dist(),\n",
    "#                                 'Regressor': pm.Normal.dist(mu=0, tau=1),\n",
    "#                                }\n",
    "#                         )\n",
    "#     y_est = pm.Deterministic('y_est', glm.y_est) # make predictions \n",
    "#     trace = pm.sample(500, cores=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Auto-assigning NUTS sampler...\n",
      "Initializing NUTS using jitter+adapt_diag...\n",
      "Multiprocess sampling (2 chains in 2 jobs)\n",
      "NUTS: [sigma, tau, a_raw]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "        <style>\n",
       "            /* Turns off some styling */\n",
       "            progress {\n",
       "                /* gets rid of default border in Firefox and Opera. */\n",
       "                border: none;\n",
       "                /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "                background-size: auto;\n",
       "            }\n",
       "            .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "                background: #F44336;\n",
       "            }\n",
       "        </style>\n",
       "      <progress value='3000' class='' max='3000' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      100.00% [3000/3000 00:13<00:00 Sampling 2 chains, 94 divergences]\n",
       "    </div>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 1_000 tune and 500 draw iterations (2_000 + 1_000 draws total) took 14 seconds.\n",
      "There were 6 divergences after tuning. Increase `target_accept` or reparameterize.\n",
      "The acceptance probability does not match the target. It is 0.8813382039916366, but should be close to 0.8. Try to increase the number of tuning steps.\n",
      "There were 88 divergences after tuning. Increase `target_accept` or reparameterize.\n",
      "The acceptance probability does not match the target. It is 0.7106573820286728, but should be close to 0.8. Try to increase the number of tuning steps.\n",
      "The estimated number of effective samples is smaller than 200 for some parameters.\n"
     ]
    }
   ],
   "source": [
    "df = pd.DataFrame({'X':np.arange(-5, 5, .1),\n",
    "                   'Y':Y  # add your data here\n",
    "                  })\n",
    "B = patsy.dmatrix('bs(X, knots=np.arange(-5,5,1), degree=3)', df, \n",
    "                  return_type='dataframe')\n",
    "\n",
    "with pm.Model() as model:\n",
    "    a_raw = pm.Normal('a_raw', mu=0, sd=1, shape=num_basis)\n",
    "    tau = pm.Cauchy('tau', alpha=1, beta=1)\n",
    "    sigma = pm.Cauchy('sigma', alpha=1, beta=1)\n",
    "    y_hat = pm.Deterministic('y_hat', tt.dot(B, a_raw*tau))\n",
    "    y = pm.Normal('y', mu=y_hat, sd=sigma, observed=df.Y)\n",
    "    trace = pm.sample(500, cores=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(trace.y_hat);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(df.X, df.Y, 'o', mec='k', mew=1, color='none', label='data')\n",
    "plt.plot(df.X, Y_true, '-', color='blue', label='truth')\n",
    "plt.plot(df.X, trace.y_hat.mean(axis=0), '-', color='red', label='prediction')\n",
    "plt.legend(loc=(1.01, .01));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "dismod_mr",
   "language": "python",
   "name": "dismod_mr"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}