{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Uplift Trees/Forests Visualization"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Introduction\n",
"This example notebooks illustrates how to visualize uplift trees for interpretation and diagnosis. \n",
"\n",
"#### Supported Models\n",
"These visualization functions work only for tree-based algorithms:\n",
"\n",
"- Uplift tree/random forests on KL divergence, Euclidean Distance, and Chi-Square\n",
"- Uplift tree/random forests on Contextual Treatment Selection\n",
"\n",
"Currently, they are NOT supporting Meta-learner algorithms\n",
"\n",
"- S-learner\n",
"- T-learner\n",
"- X-learner\n",
"- R-learner\n",
"\n",
"#### Supported Usage\n",
"This notebook will show how to use visualization for:\n",
"\n",
"- Uplift Tree and Uplift Random Forest\n",
" - Visualize a trained uplift classification tree model\n",
" - Visualize an uplift tree in a trained uplift random forests\n",
"\n",
"- Training and Validation Data\n",
" - Visualize the validation tree: fill the trained uplift classification tree with validation (or testing) data, and show the statistics for both training data and validation data\n",
" \n",
"- One Treatment Group and Multiple Treatment Groups\n",
" - Visualize the case where there are one control group and one treatment group\n",
" - Visualize the case where there are one control group and multiple treatment groups\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 1 Load Modules"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Load CausalML modules"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"ExecuteTime": {
"end_time": "2021-11-29T22:45:30.766847Z",
"start_time": "2021-11-29T22:45:29.045877Z"
}
},
"outputs": [],
"source": [
"from causalml.dataset import make_uplift_classification\n",
"from causalml.inference.tree import UpliftTreeClassifier, UpliftRandomForestClassifier\n",
"from causalml.inference.tree import uplift_tree_string, uplift_tree_plot"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Load standard modules"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"ExecuteTime": {
"end_time": "2021-11-29T22:45:30.772450Z",
"start_time": "2021-11-29T22:45:30.769222Z"
}
},
"outputs": [],
"source": [
"import numpy as np\n",
"import pandas as pd\n",
"from IPython.display import Image\n",
"from sklearn.model_selection import train_test_split"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## One Control + One Treatment for Uplift Classification Tree "
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"ExecuteTime": {
"end_time": "2021-11-29T22:45:30.906793Z",
"start_time": "2021-11-29T22:45:30.775793Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" mean | \n",
" size | \n",
"
\n",
" \n",
" | \n",
" conversion | \n",
" conversion | \n",
"
\n",
" \n",
" | treatment_group_key | \n",
" | \n",
" | \n",
"
\n",
" \n",
" \n",
" \n",
" | control | \n",
" 0.5110 | \n",
" 1000 | \n",
"
\n",
" \n",
" | treatment1 | \n",
" 0.5140 | \n",
" 1000 | \n",
"
\n",
" \n",
" | All | \n",
" 0.5125 | \n",
" 2000 | \n",
"
\n",
" \n",
"
\n",
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" mean | \n",
" size | \n",
"
\n",
" \n",
" | \n",
" conversion | \n",
" conversion | \n",
"
\n",
" \n",
" | treatment_group_key | \n",
" | \n",
" | \n",
"
\n",
" \n",
" \n",
" \n",
" | control | \n",
" 0.511 | \n",
" 1000 | \n",
"
\n",
" \n",
" | treatment1 | \n",
" 0.514 | \n",
" 1000 | \n",
"
\n",
" \n",
" | treatment2 | \n",
" 0.559 | \n",
" 1000 | \n",
"
\n",
" \n",
" | treatment3 | \n",
" 0.600 | \n",
" 1000 | \n",
"
\n",
" \n",
" | All | \n",
" 0.546 | \n",
" 4000 | \n",
"
\n",
" \n",
"
\n",
"