diff --git a/examples/causal_inference/counterfactuals_do_operator.ipynb b/examples/causal_inference/counterfactuals_do_operator.ipynb new file mode 100644 index 000000000..8b33bee52 --- /dev/null +++ b/examples/causal_inference/counterfactuals_do_operator.ipynb @@ -0,0 +1,3035 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "domestic-remove", + "metadata": {}, + "source": [ + "(counterfactuals_do_operator)=\n", + "# Counterfactual generation using pymc do-operator\n", + "\n", + ":::{post} August, 2023\n", + ":tags: causality, causal inference, do-operator, counterfactuals\n", + ":category: beginner, reference\n", + ":author: Shekhar Khandelwal\n", + ":::" + ] + }, + { + "cell_type": "markdown", + "id": "72588976-efc3-4adc-bec2-bc5b6ac4b7e1", + "metadata": {}, + "source": [ + "This is some introductory text. Consult the [style guide](https://docs.pymc.io/en/latest/contributing/jupyter_style.html)." + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "elect-softball", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "import arviz as az\n", + "import matplotlib.pyplot as plt\n", + "import numpy as np\n", + "import pandas as pd\n", + "import pymc as pm\n", + "import pymc_experimental as pmx\n", + "from packaging import version\n", + "# import the new functionality\n", + "from pymc_experimental.model_transform.conditioning import do, observe\n", + "import warnings\n", + "warnings.filterwarnings('ignore')" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "level-balance", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "%config InlineBackend.figure_format = 'retina' # high resolution figures\n", + "az.style.use(\"arviz-darkgrid\")\n", + "rng = np.random.default_rng(42)\n", + "SEED = 8927" + ] + }, + { + "cell_type": "markdown", + "id": "sapphire-yellow", + "metadata": {}, + "source": [ + "# Introduction\n", + "\n", + "In the realm of data science and analytics, understanding the causal relationships between variables is paramount. While traditional statistical methods have provided insights into these relationships, the advent of probabilistic programming has ushered in a new era of causal analysis. In this article, we will explore the power of counterfactuals in causal analysis using the PyMC framework, with a special focus on the “do-operator.”\n", + "Counterfactuals are essentially “what-if” scenarios that allow us to understand the potential outcomes had a different action been taken or a different condition been present. By leveraging the PyMC framework and its “do-operator,” we can programmatically simulate these scenarios, giving us a deeper understanding of the relationships between predictors and target variables.\n", + "\n", + "Through a step-by-step guide, we will delve into the process of building a PyMC model skeleton, generating data using the do-operator, and validating the relationships captured by the model. Furthermore, we will explore the magic of the do-operator in simulating different ‘what-if’ scenarios, akin to programmatic A/B testing.\n", + "\n", + "- Step 1. Build a pymc model skeleton\n", + "- Step 2. Use model skeleton and generate data using do-operator to infuse relationship between predictors and target variable (ssshhh, that’s a hidden superhero feature of do-operator ;) )\n", + "- Step 3. Use observe-operator to assign generated data on the model skeleton\n", + "- Step 4. Create samples and validate that the infused relationship between predictors and target variable are captured by the model samples (isn’t that what we expect a predictive model to do ;) )\n", + "- Step 5. Use do-operator to time travel, and generate target variable with different ‘what-if’ scenarios (basically mimic A/B testing…programatically)\n" + ] + }, + { + "cell_type": "markdown", + "id": "d3756a0d-447e-4ffd-8305-e0f2329dbc3a", + "metadata": {}, + "source": [ + "### Step 1. Build a pymc model skeleton\n", + "\n", + "For this demo, we are building a very simple Linear Regression model.\n", + "- Predictor — ‘a’, ‘b’, ‘c’\n", + "- Target Variable — ‘y’\n", + "- Coefficients —\n", + ">- ‘beta_ay’ -> coefficient of |a|\n", + ">- ‘beta_by’ -> coefficient of |b|\n", + ">- ‘beta_cy’ -> coefficient of |c|" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "21e66b38", + "metadata": {}, + "outputs": [ + { + "data": { + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "clusteri (1)\n", + "\n", + "i (1)\n", + "\n", + "\n", + "\n", + "beta_by\n", + "\n", + "beta_by\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "y_mu\n", + "\n", + "y_mu\n", + "~\n", + "Deterministic\n", + "\n", + "\n", + "\n", + "beta_by->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "beta_ay\n", + "\n", + "beta_ay\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "beta_ay->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "beta_y0\n", + "\n", + "beta_y0\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "beta_y0->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "sigma_y\n", + "\n", + "sigma_y\n", + "~\n", + "HalfNormal\n", + "\n", + "\n", + "\n", + "y\n", + "\n", + "y\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "sigma_y->y\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "beta_cy\n", + "\n", + "beta_cy\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "beta_cy->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "c\n", + "\n", + "c\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "c->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "b\n", + "\n", + "b\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "b->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "y_mu->y\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "a\n", + "\n", + "a\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "a->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "execution_count": 5, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "with pm.Model(coords_mutable={\"i\": [0]}) as model_generative:\n", + " # priors\n", + " beta_y0 = pm.Normal(\"beta_y0\")\n", + " beta_ay = pm.Normal(\"beta_ay\")\n", + " beta_by = pm.Normal(\"beta_by\")\n", + " beta_cy = pm.Normal(\"beta_cy\")\n", + " # observation noise on Y\n", + " sigma_y = pm.HalfNormal(\"sigma_y\")\n", + " # core nodes and causal relationships\n", + " a = pm.Normal(\"a\", mu=0, sigma=1, dims=\"i\")\n", + " b = pm.Normal(\"b\", mu=0, sigma=1, dims=\"i\")\n", + " c = pm.Normal(\"c\", mu=0, sigma=1, dims=\"i\")\n", + " y_mu = pm.Deterministic(\"y_mu\", beta_y0 + (beta_ay * a) + (beta_by * b) + (beta_cy * c), dims=\"i\")\n", + " y = pm.Normal(\"y\", mu=y_mu, sigma=sigma_y, dims=\"i\")\n", + "\n", + "\n", + "pm.model_to_graphviz(model_generative)" + ] + }, + { + "cell_type": "markdown", + "id": "e2320755-54f5-4051-b73e-feb60de8b783", + "metadata": {}, + "source": [ + "### Step 2. Use model skeleton and generate data using do-operator to infuse relationship between predictors and target variable. We will use this generated data for modelling later.\n", + "\n", + "Let’s first define the predictors relationship with target variable." + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "62d01fc3-9a12-4dcd-b2f1-d7a09116a3c6", + "metadata": {}, + "outputs": [], + "source": [ + "true_values = {\n", + " \"beta_ay\": 1.5,\n", + " \"beta_by\": 0.7,\n", + " \"beta_cy\": 0.3,\n", + " \"sigma_y\": 0.2,\n", + " \"beta_y0\": 0.0\n", + "}" + ] + }, + { + "cell_type": "markdown", + "id": "4bad6441-6921-446b-82a1-6a995e74faff", + "metadata": {}, + "source": [ + "Basically what we are saying here is, we are intentionally defining the coefficient values, which we expect predictive model to predict later on.\n", + "\n", + "Now the magic begins. We will use do-operator to use this dictionary and sample data variables. How do we do this ? Simple by passing two arguments to pymc do-operator. First, the model skeleton object. And second, the coefficient dictionary." + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "fbf04aa4-e68f-43fd-ba70-91a287b6b12d", + "metadata": {}, + "outputs": [], + "source": [ + "model_simulate = do(model_generative, true_values)" + ] + }, + { + "cell_type": "markdown", + "id": "a149f565-ccb3-4118-ac5a-da67733e3e5a", + "metadata": {}, + "source": [ + "This will create a new model object with the coefficent variables values infused. " + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "113da0d7-b9d7-4cd2-98fa-7fa794169b94", + "metadata": {}, + "outputs": [ + { + "data": { + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "clusteri (1)\n", + "\n", + "i (1)\n", + "\n", + "\n", + "\n", + "beta_by\n", + "\n", + "beta_by\n", + "~\n", + "ConstantData\n", + "\n", + "\n", + "\n", + "y_mu\n", + "\n", + "y_mu\n", + "~\n", + "Deterministic\n", + "\n", + "\n", + "\n", + "beta_by->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "beta_ay\n", + "\n", + "beta_ay\n", + "~\n", + "ConstantData\n", + "\n", + "\n", + "\n", + "beta_ay->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "beta_y0\n", + "\n", + "beta_y0\n", + "~\n", + "ConstantData\n", + "\n", + "\n", + "\n", + "beta_y0->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "sigma_y\n", + "\n", + "sigma_y\n", + "~\n", + "ConstantData\n", + "\n", + "\n", + "\n", + "y\n", + "\n", + "y\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "sigma_y->y\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "beta_cy\n", + "\n", + "beta_cy\n", + "~\n", + "ConstantData\n", + "\n", + "\n", + "\n", + "beta_cy->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "c\n", + "\n", + "c\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "c->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "b\n", + "\n", + "b\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "b->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "y_mu->y\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "a\n", + "\n", + "a\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "a->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "pm.model_to_graphviz(model_simulate)" + ] + }, + { + "cell_type": "markdown", + "id": "af0e13da-29e6-44a6-852e-c3e965d7c462", + "metadata": {}, + "source": [ + "The gray shades on the coefficient variables depicts the tale. Check the previous model graph, it was all white.\n", + "\n", + "Now, all we have to do is generate samples, the known pymc way.\n", + "\n", + "Lets generate 100 samples." + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "2c651c0a-29f8-4669-baf7-986687f59317", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Sampling: [a, b, c, y]\n" + ] + } + ], + "source": [ + "N = 100\n", + "\n", + "with model_simulate:\n", + " simulate = pm.sample_prior_predictive(samples=N)" + ] + }, + { + "cell_type": "markdown", + "id": "9ecd7313-e68c-4439-803c-576a5c474e4b", + "metadata": {}, + "source": [ + "We know that this generates an Arviz object, and since we have called sample_prior_predictive, hence the object will only contain priors." + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "99a3fede-e773-4823-bb5e-ece89805bcc6", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "\n", + "
\n", + "
\n", + "
arviz.InferenceData
\n", + "
\n", + "
    \n", + " \n", + "
  • \n", + " \n", + " \n", + "
    \n", + "
    \n", + "
      \n", + "
      \n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "
      <xarray.Dataset>\n",
+       "Dimensions:  (chain: 1, draw: 100, i: 1)\n",
+       "Coordinates:\n",
+       "  * chain    (chain) int64 0\n",
+       "  * draw     (draw) int64 0 1 2 3 4 5 6 7 8 9 ... 90 91 92 93 94 95 96 97 98 99\n",
+       "  * i        (i) int64 0\n",
+       "Data variables:\n",
+       "    c        (chain, draw, i) float64 0.2479 1.523 -0.1481 ... 0.3244 -1.279\n",
+       "    b        (chain, draw, i) float64 -0.03438 1.362 -1.12 ... 0.2729 -0.9092\n",
+       "    y        (chain, draw, i) float64 -0.7672 2.557 -1.755 ... 0.0005881 -1.189\n",
+       "    y_mu     (chain, draw, i) float64 -0.7951 2.484 -1.809 ... -0.144 -0.6784\n",
+       "    a        (chain, draw, i) float64 -0.5636 0.7158 -0.6534 ... -0.2882 0.2278\n",
+       "Attributes:\n",
+       "    created_at:                 2023-08-25T09:08:58.839382\n",
+       "    arviz_version:              0.15.1\n",
+       "    inference_library:          pymc\n",
+       "    inference_library_version:  5.6.0

      \n", + "
    \n", + "
    \n", + "
    • \n", + " \n", + "
  • \n", + " \n", + " \n", + "
    \n", + "
    \n", + "
      \n", + "
      \n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "
      <xarray.Dataset>\n",
+       "Dimensions:  ()\n",
+       "Data variables:\n",
+       "    beta_y0  float64 0.0\n",
+       "    beta_ay  float64 1.5\n",
+       "    beta_by  float64 0.7\n",
+       "    beta_cy  float64 0.3\n",
+       "    sigma_y  float64 0.2\n",
+       "Attributes:\n",
+       "    created_at:                 2023-08-25T09:08:58.842535\n",
+       "    arviz_version:              0.15.1\n",
+       "    inference_library:          pymc\n",
+       "    inference_library_version:  5.6.0

      \n", + "
    \n", + "
    \n", + "
    • \n", + " \n", + "
\n", + "
\n", + " " + ], + "text/plain": [ + "Inference data with groups:\n", + "\t> prior\n", + "\t> constant_data" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "simulate" + ] + }, + { + "cell_type": "markdown", + "id": "9cc7caf3-61d7-47ac-a207-fa22778a9f2a", + "metadata": {}, + "source": [ + "Extract the sampled prior data into a pandas dataframe." + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "86e38344-28ad-4e0d-a987-4385ed320571", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "(100, 4)\n" + ] + }, + { + "data": { + "text/html": [ + "
See Full Dataframe in Mito
\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
abcy
0-0.563618-0.0343780.247925-0.767228
10.7158451.3618171.5232502.557340
2-0.653367-1.120212-0.148130-1.755221
30.0837410.091703-0.3003490.292252
40.444869-1.2895641.3353200.535065
" + ], + "text/plain": [ + " a b c y\n", + "0 -0.563618 -0.034378 0.247925 -0.767228\n", + "1 0.715845 1.361817 1.523250 2.557340\n", + "2 -0.653367 -1.120212 -0.148130 -1.755221\n", + "3 0.083741 0.091703 -0.300349 0.292252\n", + "4 0.444869 -1.289564 1.335320 0.535065" + ] + }, + "execution_count": 12, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "observed = {\n", + " \"a\": simulate.prior[\"a\"].values.flatten(),\n", + " \"b\": simulate.prior[\"b\"].values.flatten(),\n", + " \"c\": simulate.prior[\"c\"].values.flatten(),\n", + " \"y\": simulate.prior[\"y\"].values.flatten()\n", + "}\n", + "\n", + "df = pd.DataFrame(observed)\n", + "print(df.shape)\n", + "df.head(5)" + ] + }, + { + "cell_type": "markdown", + "id": "410b2941-ee10-444b-ab5b-36f229a6dba7", + "metadata": {}, + "source": [ + "Ok, so now we are all set with a sample data.\n", + "\n", + "Before we move to Step 3, just for fun, lets see if a simple Linear Regression model can extract these coefficients." + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "352d549c-f6dc-4a42-9387-df6a201e9bb2", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "array([1.52519075, 0.70393163, 0.30104623])" + ] + }, + "execution_count": 13, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from sklearn.linear_model import LinearRegression\n", + "# Splitting data into predictors and target variable\n", + "X = df[['a', 'b', 'c']]\n", + "y = df['y']\n", + "\n", + "# Building the linear regression model\n", + "model = LinearRegression()\n", + "model.fit(X, y)\n", + "\n", + "# Getting the coefficients for each predictor\n", + "coefficients = model.coef_\n", + "coefficients" + ] + }, + { + "cell_type": "markdown", + "id": "ae564cd5-23e4-4225-9ee8-e642ae47eeb7", + "metadata": {}, + "source": [ + "Close enough ! Okay, lets not digress from the original topic. The pymc magic !\n", + "\n", + "### Step 3. Use observe-operator to assign generated data on the model skeleton\n", + "\n", + "Now, this is another cool feature of pymc newly introduced observe method. Observe method, takes in a model skeleton and the dictionary with the data for the variables we want to infuse into the model." + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "17860fac-d25b-46bf-b4d7-8a920610a853", + "metadata": {}, + "outputs": [ + { + "data": { + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "clusteri (100)\n", + "\n", + "i (100)\n", + "\n", + "\n", + "\n", + "beta_by\n", + "\n", + "beta_by\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "y_mu\n", + "\n", + "y_mu\n", + "~\n", + "Deterministic\n", + "\n", + "\n", + "\n", + "beta_by->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "beta_ay\n", + "\n", + "beta_ay\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "beta_ay->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "beta_y0\n", + "\n", + "beta_y0\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "beta_y0->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "sigma_y\n", + "\n", + "sigma_y\n", + "~\n", + "HalfNormal\n", + "\n", + "\n", + "\n", + "y\n", + "\n", + "y\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "sigma_y->y\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "beta_cy\n", + "\n", + "beta_cy\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "beta_cy->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "c\n", + "\n", + "c\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "c->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "b\n", + "\n", + "b\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "b->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "y_mu->y\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "a\n", + "\n", + "a\n", + "~\n", + "Normal\n", + "\n", + "\n", + "\n", + "a->y_mu\n", + "\n", + "\n", + "\n", + "\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "execution_count": 14, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data_dict={\"a\": df[\"a\"], \n", + " \"b\": df[\"b\"], \n", + " \"c\": df[\"c\"], \n", + " \"y\": df[\"y\"]}\n", + "model_inference = observe(model_generative, data_dict)\n", + "model_inference.set_dim(\"i\", N, coord_values=np.arange(N))\n", + "pm.model_to_graphviz(model_inference)" + ] + }, + { + "cell_type": "markdown", + "id": "5ad86c3a-ca67-406c-b114-1f5449b354a1", + "metadata": {}, + "source": [ + "See the gray matter again. This time we have observed data infused into the model, and we have to sample for the coefficient and other parameters.\n", + "\n", + "So, lets sample.\n", + "\n", + "### Step 4. Create samples and validate that the infused relationship between predictors and target variable are captured by the model samples" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "6ef56be2-06a9-49b8-8044-e789f1d254e9", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Auto-assigning NUTS sampler...\n", + "Initializing NUTS using jitter+adapt_diag...\n", + "Multiprocess sampling (4 chains in 4 jobs)\n", + "NUTS: [beta_y0, beta_ay, beta_by, beta_cy, sigma_y]\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " 100.00% [8000/8000 00:03<00:00 Sampling 4 chains, 0 divergences]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Sampling 4 chains for 1_000 tune and 1_000 draw iterations (4_000 + 4_000 draws total) took 21 seconds.\n" + ] + } + ], + "source": [ + "with model_inference:\n", + " idata = pm.sample(random_seed=SEED)" + ] + }, + { + "cell_type": "markdown", + "id": "375287a3-068a-4815-b86b-f472676a5416", + "metadata": {}, + "source": [ + "Now, lets validate if model captured the infused coefficient values in the data." + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "7eb06ae5-ca72-44c1-b83b-82e50180c9a7", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": { + "image/png": { + "height": 590, + "width": 1188 + } + }, + "output_type": "display_data" + } + ], + "source": [ + "az.plot_posterior(\n", + " idata,\n", + " var_names=list(true_values.keys()),\n", + " ref_val=list(true_values.values()),\n", + " figsize=(12, 6),\n", + ")\n", + "plt.tight_layout();" + ] + }, + { + "cell_type": "markdown", + "id": "c89e6a59-db38-499e-a6ac-73a53a4fca19", + "metadata": {}, + "source": [ + "BAM ! Pretty nice fit !\n", + "\n", + "Now, lets do what we are supposed to do ! Counterfactuals.\n", + "\n", + "Basically, this is about generating target variable values with different predictor values. Basically, answering what if questions !\n", + "\n", + "_What-if there was all ‘b’ values as 0 ?_\n", + "\n", + "_What-if all ‘b’ values were double ?_\n", + "\n", + "How to do this ? Here you go..\n", + "\n", + "### Step 5. Use do-operator to time travel, and generate target variable with different ‘what-if’ scenarios.\n", + "Since, we want to experiment with ‘b’, lets first assign observed values to ‘a’ and ‘c’. Not to ‘y’, because that’s what we want to sample. Correct !" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "c0643453-0bb6-4ceb-9fb1-3adf57f031b9", + "metadata": {}, + "outputs": [], + "source": [ + "model_counterfactual = do(model_inference, {\"a\": df[\"a\"], \"c\": df[\"c\"]})" + ] + }, + { + "cell_type": "markdown", + "id": "8e8067a3-6430-4cf9-b53e-a5bfbd8e4488", + "metadata": {}, + "source": [ + "Now, lets begin the fun part. Let’s generate counterfactuals.\n", + "\n", + "### _Scenario 1 :- What if all values for ‘b’ were 0 ?_" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "f7cbeba3-d047-447f-bfe2-077604fd2fd9", + "metadata": {}, + "outputs": [], + "source": [ + "model_b0 = do(model_counterfactual, {\"b\": np.zeros(N, dtype=\"int32\")}, prune_vars=True)\n", + "model_b1 = do(model_counterfactual, {\"b\": df[\"b\"]}, prune_vars=True)" + ] + }, + { + "cell_type": "markdown", + "id": "d3b23275-029d-40a2-8603-796c740bed29", + "metadata": {}, + "source": [ + "Just sample." + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "id": "edd9b175-5a4b-457a-b9d9-560251733600", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Sampling: []\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " 100.00% [4000/4000 00:00<00:00]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Sampling: []\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " 100.00% [4000/4000 00:00<00:00]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Sample when 'b' was 0: P(y | (a,c), do(b=0))\n", + "idata_b0 = pm.sample_posterior_predictive(\n", + " idata,\n", + " model=model_b0,\n", + " predictions=True,\n", + " var_names=[\"y_mu\"],\n", + " random_seed=SEED,\n", + ")\n", + "# Sample when 'b' was as observed: P(y | (a,c), do(b=observed))\n", + "idata_b1 = pm.sample_posterior_predictive(\n", + " idata,\n", + " model=model_b1,\n", + " predictions=True,\n", + " var_names=[\"y_mu\"],\n", + " random_seed=SEED,\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "6cabe3d1-ecab-4bb8-bcea-d45e91ca0b04", + "metadata": {}, + "source": [ + "Some basic python and here we have the counterfactuals." + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "4d853c28-4029-4372-a46d-132239918fe5", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
See Full Dataframe in Mito
\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
abcyb_scenario_1y_scenario_1
0-0.563618-0.0343780.247925-0.7672280-0.790475
10.7158451.3618171.5232502.55734001.543476
2-0.653367-1.120212-0.148130-1.7552210-1.046445
30.0837410.091703-0.3003490.29225200.031290
40.444869-1.2895641.3353200.53506501.073895
" + ], + "text/plain": [ + " a b c y b_scenario_1 y_scenario_1\n", + "0 -0.563618 -0.034378 0.247925 -0.767228 0 -0.790475\n", + "1 0.715845 1.361817 1.523250 2.557340 0 1.543476\n", + "2 -0.653367 -1.120212 -0.148130 -1.755221 0 -1.046445\n", + "3 0.083741 0.091703 -0.300349 0.292252 0 0.031290\n", + "4 0.444869 -1.289564 1.335320 0.535065 0 1.073895" + ] + }, + "execution_count": 22, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df[\"b_scenario_1\"]=0\n", + "df[\"y_scenario_1\"]=idata_b0.predictions.y_mu.mean((\"chain\", \"draw\")).values.reshape(1, -1).flatten()\n", + "df.head(5)" + ] + }, + { + "cell_type": "markdown", + "id": "3ad2d911-2948-4553-9322-121f064696e0", + "metadata": {}, + "source": [ + "### _Scenario 2: What if ‘b’ was 5 times as observed_" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "id": "1c22e18f-cb96-4ca9-93f4-e8323b1528b1", + "metadata": {}, + "outputs": [], + "source": [ + "model_b0 = do(model_counterfactual, {\"b\": 5*df[\"b\"]}, prune_vars=True)\n", + "model_b1 = do(model_counterfactual, {\"b\": df[\"b\"]}, prune_vars=True)" + ] + }, + { + "cell_type": "markdown", + "id": "5422848d-e6cd-4f3e-8641-640ed227ea78", + "metadata": {}, + "source": [ + "Sample." + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "id": "97990259-7cd9-4801-8af0-b8e3a46ffa7b", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Sampling: []\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " 100.00% [4000/4000 00:00<00:00]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Sampling: []\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " 100.00% [4000/4000 00:00<00:00]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
See Full Dataframe in Mito
\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
abcyb_scenario_1y_scenario_1b_scenario_2y_scenario_2
0-0.563618-0.0343780.247925-0.7672280-0.790475-0.171889-0.911420
10.7158451.3618171.5232502.55734001.5434766.8090876.334516
2-0.653367-1.120212-0.148130-1.7552210-1.046445-5.601060-4.987488
30.0837410.091703-0.3003490.29225200.0312900.4585170.353914
40.444869-1.2895641.3353200.53506501.073895-6.447820-3.462948
" + ], + "text/plain": [ + " a b c y b_scenario_1 y_scenario_1 \\\n", + "0 -0.563618 -0.034378 0.247925 -0.767228 0 -0.790475 \n", + "1 0.715845 1.361817 1.523250 2.557340 0 1.543476 \n", + "2 -0.653367 -1.120212 -0.148130 -1.755221 0 -1.046445 \n", + "3 0.083741 0.091703 -0.300349 0.292252 0 0.031290 \n", + "4 0.444869 -1.289564 1.335320 0.535065 0 1.073895 \n", + "\n", + " b_scenario_2 y_scenario_2 \n", + "0 -0.171889 -0.911420 \n", + "1 6.809087 6.334516 \n", + "2 -5.601060 -4.987488 \n", + "3 0.458517 0.353914 \n", + "4 -6.447820 -3.462948 " + ] + }, + "execution_count": 24, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Sample when 'b' was 5 times b: P(y | (a,c), do(b=5*b))\n", + "idata_b0 = pm.sample_posterior_predictive(\n", + " idata,\n", + " model=model_b0,\n", + " predictions=True,\n", + " var_names=[\"y_mu\"],\n", + " random_seed=SEED,\n", + ")\n", + "# Sample when 'b' was as observed: P(y | (a,c), do(b=observed))\n", + "idata_b1 = pm.sample_posterior_predictive(\n", + " idata,\n", + " model=model_b1,\n", + " predictions=True,\n", + " var_names=[\"y_mu\"],\n", + " random_seed=SEED,\n", + ")\n", + "\n", + "df[\"b_scenario_2\"]=5*df[\"b\"]\n", + "df[\"y_scenario_2\"]=idata_b0.predictions.y_mu.mean((\"chain\", \"draw\")).values.reshape(1, -1).flatten()\n", + "df.head(5)" + ] + }, + { + "cell_type": "markdown", + "id": "2005665d-300b-4a01-91e4-a0901dbbaa97", + "metadata": {}, + "source": [ + "Ok, so now you got the idea. It's an open playground. Go back in time, change whatever you want to change, and see how output changes.\n", + "\n", + "This opens the door for many more possibilities in various use cases. Especially, Causal Analytics !" + ] + }, + { + "cell_type": "markdown", + "id": "b743d58b-2678-4e17-9947-a8fe4ed03e21", + "metadata": {}, + "source": [ + "## Authors\n", + "- Authored by [Shekhar Khandelwal](https://github.com/shekharkhandelwal1983) in August 2023 " + ] + }, + { + "cell_type": "markdown", + "id": "closed-frank", + "metadata": {}, + "source": [ + "## References\n", + "\n", + "https://medium.com/@khandelwal-shekhar/counterfactuals-for-causal-analysis-via-pymc-do-operator-234ba04e4e80\n", + "\n", + "https://www.pymc-labs.io/blog-posts/causal-analysis-with-pymc-answering-what-if-with-the-new-do-operator/" + ] + }, + { + "cell_type": "markdown", + "id": "0717070c-04aa-4836-ab95-6b3eff0dcaaf", + "metadata": {}, + "source": [ + "## Watermark" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "id": "sound-calculation", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Last updated: Fri Aug 25 2023\n", + "\n", + "Python implementation: CPython\n", + "Python version : 3.10.12\n", + "IPython version : 8.14.0\n", + "\n", + "pytensor: 2.12.3\n", + "\n", + "json : 2.0.9\n", + "packaging : 23.0\n", + "matplotlib : 3.7.2\n", + "pandas : 1.5.3\n", + "mitosheet : 0.1.487\n", + "pymc_experimental: 0.0.8\n", + "pymc : 5.6.0\n", + "IPython : 8.14.0\n", + "numpy : 1.23.5\n", + "ctypes : 1.1.0\n", + "arviz : 0.15.1\n", + "\n", + "Watermark: 2.4.3\n", + "\n" + ] + } + ], + "source": [ + "%load_ext watermark\n", + "%watermark -n -u -v -iv -w -p pytensor" + ] + }, + { + "cell_type": "markdown", + "id": "1e4386fc-4de9-4535-a160-d929315633ef", + "metadata": {}, + "source": [ + ":::{include} ../page_footer.md\n", + ":::" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.12" + }, + "vscode": { + "interpreter": { + "hash": "d5f0cba85daacbebbd957da1105312a62c58952ca942f7218a10e4aa5f415a19" + } + } + }, + "nbformat": 4, + "nbformat_minor": 5 +}