From 59f5a99d347c5149e5337452cbf6dbdb2ce23198 Mon Sep 17 00:00:00 2001
From: Chirag Nagpal <nagpalchirag1994@gmail.com>
Date: Sat, 12 Feb 2022 19:48:02 -0500
Subject: [PATCH] 	new file:   examples/CV Survival Regression on SUPPORT
 Dataset.ipynb 	new file:   examples/DCM on SUPPORT Dataset copy.ipynb 
 modified:   examples/DSM on SUPPORT Dataset.ipynb 	new file:  
 examples/Phenotyping Survival Data.ipynb

---
 ...rvival Regression on SUPPORT Dataset.ipynb | 1088 +++++++++++++++++
 examples/DCM on SUPPORT Dataset copy.ipynb    |  247 ++++
 examples/DSM on SUPPORT Dataset.ipynb         |  162 ++-
 examples/Phenotyping Survival Data.ipynb      |  152 +++
 4 files changed, 1632 insertions(+), 17 deletions(-)
 create mode 100644 examples/CV Survival Regression on SUPPORT Dataset.ipynb
 create mode 100644 examples/DCM on SUPPORT Dataset copy.ipynb
 create mode 100644 examples/Phenotyping Survival Data.ipynb

diff --git a/examples/CV Survival Regression on SUPPORT Dataset.ipynb b/examples/CV Survival Regression on SUPPORT Dataset.ipynb
new file mode 100644
index 0000000..afdf742
--- /dev/null
+++ b/examples/CV Survival Regression on SUPPORT Dataset.ipynb	
@@ -0,0 +1,1088 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# `auton-survival` Cross Validation Survival Regression\n",
+    "\n",
+    "`auton-survival` offers a simple to use API to train Survival Regression Models that performs cross validation model selection by minimizing integrated brier score. In this notebook we demonstrate the use of `auton-survival` to train survival models on the *SUPPORT* dataset in cross validation fashion."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import sys\n",
+    "\n",
+    "sys.path.append('../')\n",
+    "from auton_survival import datasets\n",
+    "outcomes, features = datasets.load_support()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from auton_survival.preprocessing import Preprocessor\n",
+    "\n",
+    "cat_feats = ['sex', 'dzgroup', 'dzclass', 'income', 'race', 'ca']\n",
+    "num_feats = ['age', 'num.co', 'meanbp', 'wblc', 'hrt', 'resp', \n",
+    "\t     'temp', 'pafi', 'alb', 'bili', 'crea', 'sod', 'ph', \n",
+    "             'glucose', 'bun', 'urine', 'adlp', 'adls']\n",
+    "\n",
+    "features = Preprocessor().fit_transform(features, cat_feats=cat_feats, num_feats=num_feats)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "horizons = [0.25, 0.5, 0.75]\n",
+    "times = np.quantile(outcomes.time[outcomes.event==1], horizons).tolist()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "  0%|          | 0/4 [00:00<?, ?it/s]\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1779/10000 [00:03<00:14, 554.25it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  7.14it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 14%|█▍        | 1418/10000 [00:02<00:17, 482.72it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  6.95it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1793/10000 [00:03<00:13, 588.09it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  6.14it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1770/10000 [00:03<00:16, 502.67it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  6.72it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1814/10000 [00:03<00:14, 556.46it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  7.86it/s]\n",
+      "100%|██████████| 5/5 [00:24<00:00,  4.85s/it]\n",
+      " 25%|██▌       | 1/4 [00:24<01:14, 25.00s/it]\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1779/10000 [00:04<00:19, 428.03it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  5.74it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 14%|█▍        | 1418/10000 [00:02<00:13, 618.01it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  7.09it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1793/10000 [00:03<00:14, 570.05it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  6.29it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1770/10000 [00:02<00:13, 596.33it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  7.31it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1814/10000 [00:03<00:15, 518.75it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  6.74it/s]\n",
+      "100%|██████████| 5/5 [00:24<00:00,  4.86s/it]\n",
+      " 50%|█████     | 2/4 [00:49<00:49, 25.00s/it]\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1779/10000 [00:02<00:13, 619.50it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  6.79it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 14%|█▍        | 1418/10000 [00:02<00:15, 540.40it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  5.69it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1793/10000 [00:03<00:14, 552.62it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  6.47it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1770/10000 [00:03<00:15, 520.40it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 90%|█████████ | 9/10 [00:01<00:00,  4.68it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1814/10000 [00:02<00:13, 619.36it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  6.73it/s]\n",
+      "100%|██████████| 5/5 [00:23<00:00,  4.74s/it]\n",
+      " 75%|███████▌  | 3/4 [01:14<00:24, 24.74s/it]\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1779/10000 [00:03<00:15, 528.24it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 80%|████████  | 8/10 [00:01<00:00,  5.58it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 14%|█▍        | 1418/10000 [00:02<00:14, 586.05it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 90%|█████████ | 9/10 [00:01<00:00,  5.60it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1793/10000 [00:02<00:13, 604.69it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  5.98it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1770/10000 [00:02<00:13, 592.69it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 90%|█████████ | 9/10 [00:01<00:00,  5.96it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      " 18%|█▊        | 1814/10000 [00:02<00:13, 624.91it/s]\n",
+      "\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "\u001b[A\n",
+      "100%|██████████| 10/10 [00:01<00:00,  6.77it/s]\n",
+      "100%|██████████| 5/5 [00:22<00:00,  4.53s/it]\n",
+      "100%|██████████| 4/4 [01:37<00:00, 24.45s/it]\n",
+      " 18%|█▊        | 1815/10000 [00:03<00:14, 575.52it/s]\n",
+      "100%|██████████| 10/10 [00:01<00:00,  5.19it/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "from auton_survival.experiments import SurvivalRegressionCV\n",
+    "\n",
+    "param_grid = {'k' : [3],\n",
+    "              'distribution' : ['Weibull'],\n",
+    "              'learning_rate' : [ 1e-4, 1e-3],\n",
+    "              'layers' : [[], [100]]}\n",
+    "\n",
+    "experiment = SurvivalRegressionCV(model='dsm', cv_folds=5, hyperparam_grid=param_grid, random_seed=0)\n",
+    "model = experiment.fit(features, outcomes)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([2, 0, 0, ..., 4, 4, 2])"
+      ]
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "experiment.folds"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "out_risk = model.predict_risk(features, times)\n",
+    "out_survival = model.predict_survival(features, times)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sksurv.metrics import concordance_index_ipcw, brier_score, cumulative_dynamic_auc"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0.11593564 0.18645067 0.20140866]\n",
+      "[0.12495799 0.18906541 0.20994536]\n",
+      "[0.1228231  0.18860809 0.20365317]\n",
+      "[0.13017307 0.19169085 0.20218919]\n",
+      "[0.12173914 0.18685545 0.20330223]\n"
+     ]
+    }
+   ],
+   "source": [
+    "from auton_survival.metrics import survival_regression_metric\n",
+    "\n",
+    "for fold in set(experiment.folds):\n",
+    "    print(survival_regression_metric('brs', out_survival, outcomes, times, experiment.folds, fold))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.7797837441187897 \n",
+      "0.7403985926375387 \n",
+      "0.7026575605080277 \n",
+      "0.7753023996424281 \n",
+      "0.720478809092244 \n",
+      "0.6852956869933615 \n",
+      "0.7794202898550725 \n",
+      "0.7353951125213819 \n",
+      "0.6962889282771891 \n",
+      "0.7676518744513308 \n",
+      "0.7298422836907454 \n",
+      "0.6952441350326791 \n",
+      "0.7993855420033212 \n",
+      "0.7372447791362597 \n",
+      "0.6976821314815531 \n"
+     ]
+    }
+   ],
+   "source": [
+    "from auton_survival.metrics import survival_regression_metric\n",
+    "\n",
+    "for fold in set(experiment.folds):\n",
+    "\tfor i, time in enumerate(times):\n",
+    "\t\tprint(survival_regression_metric('ctd', out_survival[:, i], outcomes, time, experiment.folds, fold), end=' ' )\n",
+    "\tprint()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 35,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "14.0\n",
+      "58.0\n",
+      "252.0\n",
+      "14.0\n",
+      "58.0\n",
+      "252.0\n",
+      "14.0\n",
+      "58.0\n",
+      "252.0\n",
+      "14.0\n",
+      "58.0\n",
+      "252.0\n",
+      "14.0\n",
+      "58.0\n",
+      "252.0\n"
+     ]
+    }
+   ],
+   "source": [
+    "for fold in set(experiment.folds):\n",
+    "\tfor time in times:\n",
+    "\t\tprint(time)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "interpreter": {
+   "hash": "c22fbbe4c37d04364aa4e785d7cd9729f94ca3cb878d5f955e35b0076c04a2d7"
+  },
+  "kernelspec": {
+   "display_name": "Python 3.8.12 ('base')",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.12"
+  },
+  "orig_nbformat": 4
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/examples/DCM on SUPPORT Dataset copy.ipynb b/examples/DCM on SUPPORT Dataset copy.ipynb
new file mode 100644
index 0000000..04455c5
--- /dev/null
+++ b/examples/DCM on SUPPORT Dataset copy.ipynb	
@@ -0,0 +1,247 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# DSM on SUPPORT Dataset"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The SUPPORT dataset comes from the Vanderbilt University study\n",
+    "to estimate survival for seriously ill hospitalized adults.\n",
+    "(Refer to http://biostat.mc.vanderbilt.edu/wiki/Main/SupportDesc.\n",
+    "for the original datasource.)\n",
+    "\n",
+    "In this notebook, we will apply Deep Survival Machines for survival prediction on the SUPPORT data."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Load the SUPPORT Dataset\n",
+    "\n",
+    "The package includes helper functions to load the dataset.\n",
+    "\n",
+    "X represents an np.array of features (covariates),\n",
+    "T is the event/censoring times and,\n",
+    "E is the censoring indicator."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from dsm import datasets\n",
+    "x, t, e = datasets.load_dataset('SUPPORT')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Compute horizons at which we evaluate the performance of DSM\n",
+    "\n",
+    "Survival predictions are issued at certain time horizons. Here we will evaluate the performance\n",
+    "of DSM to issue predictions at the 25th, 50th and 75th event time quantile as is standard practice in Survival Analysis."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "horizons = [0.25, 0.5, 0.75]\n",
+    "times = np.quantile(t[e==1], horizons).tolist()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Splitting the data into train, test and validation sets\n",
+    "\n",
+    "We will train DSM on 70% of the Data, use a Validation set of 10% for Model Selection and report performance on the remaining 20% held out test set."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "n = len(x)\n",
+    "\n",
+    "tr_size = int(n*0.70)\n",
+    "vl_size = int(n*0.10)\n",
+    "te_size = int(n*0.20)\n",
+    "\n",
+    "x_train, x_test, x_val = x[:tr_size], x[-te_size:], x[tr_size:tr_size+vl_size]\n",
+    "t_train, t_test, t_val = t[:tr_size], t[-te_size:], t[tr_size:tr_size+vl_size]\n",
+    "e_train, e_test, e_val = e[:tr_size], e[-te_size:], e[tr_size:tr_size+vl_size]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Setting the parameter grid\n",
+    "\n",
+    "Lets set up the parameter grid to tune hyper-parameters. We will tune the number of underlying survival distributions, \n",
+    "($K$), the distribution choices (Log-Normal or Weibull), the learning rate for the Adam optimizer between $1\\times10^{-3}$ and $1\\times10^{-4}$ and the number of hidden layers between $0, 1$ and $2$."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.model_selection import ParameterGrid"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "param_grid = {'k' : [3, 4, 6],\n",
+    "              'distribution' : ['LogNormal', 'Weibull'],\n",
+    "              'learning_rate' : [ 1e-4, 1e-3],\n",
+    "              'layers' : [ [], [100], [100, 100] ]\n",
+    "             }\n",
+    "params = ParameterGrid(param_grid)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Model Training and Selection"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from dsm import DeepSurvivalMachines"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [],
+   "source": [
+    "models = []\n",
+    "for param in params:\n",
+    "    model = DeepSurvivalMachines(k = param['k'],\n",
+    "                                 distribution = param['distribution'],\n",
+    "                                 layers = param['layers'])\n",
+    "    # The fit method is called to train the model\n",
+    "    model.fit(x_train, t_train, e_train, iters = 100, learning_rate = param['learning_rate'])\n",
+    "    models.append([[model.compute_nll(x_val, t_val, e_val), model]])\n",
+    "best_model = min(models)\n",
+    "model = best_model[0][1]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Inference"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "out_risk = model.predict_risk(x_test, times)\n",
+    "out_survival = model.predict_survival(x_test, times)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Evaluation\n",
+    "\n",
+    "We evaluate the performance of DSM in its discriminative ability (Time Dependent Concordance Index and Cumulative Dynamic AUC) as well as Brier Score."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sksurv.metrics import concordance_index_ipcw, brier_score, cumulative_dynamic_auc"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "cis = []\n",
+    "brs = []\n",
+    "\n",
+    "et_train = np.array([(e_train[i], t_train[i]) for i in range(len(e_train))],\n",
+    "                 dtype = [('e', bool), ('t', float)])\n",
+    "et_test = np.array([(e_test[i], t_test[i]) for i in range(len(e_test))],\n",
+    "                 dtype = [('e', bool), ('t', float)])\n",
+    "et_val = np.array([(e_val[i], t_val[i]) for i in range(len(e_val))],\n",
+    "                 dtype = [('e', bool), ('t', float)])\n",
+    "\n",
+    "for i, _ in enumerate(times):\n",
+    "    cis.append(concordance_index_ipcw(et_train, et_test, out_risk[:, i], times[i])[0])\n",
+    "brs.append(brier_score(et_train, et_test, out_survival, times)[1])\n",
+    "roc_auc = []\n",
+    "for i, _ in enumerate(times):\n",
+    "    roc_auc.append(cumulative_dynamic_auc(et_train, et_test, out_risk[:, i], times[i])[0])\n",
+    "for horizon in enumerate(horizons):\n",
+    "    print(f\"For {horizon[1]} quantile,\")\n",
+    "    print(\"TD Concordance Index:\", cis[horizon[0]])\n",
+    "    print(\"Brier Score:\", brs[0][horizon[0]])\n",
+    "    print(\"ROC AUC \", roc_auc[horizon[0]][0], \"\\n\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/examples/DSM on SUPPORT Dataset.ipynb b/examples/DSM on SUPPORT Dataset.ipynb
index 04455c5..5891699 100644
--- a/examples/DSM on SUPPORT Dataset.ipynb	
+++ b/examples/DSM on SUPPORT Dataset.ipynb	
@@ -34,12 +34,30 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import sys\n",
+    "\n",
+    "sys.path.append('../')\n",
+    "from auton_survival import datasets\n",
+    "outcomes, features = datasets.load_support()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
    "metadata": {},
    "outputs": [],
    "source": [
-    "from dsm import datasets\n",
-    "x, t, e = datasets.load_dataset('SUPPORT')"
+    "from auton_survival.preprocessing import Preprocessor\n",
+    "cat_feats = ['sex', 'dzgroup', 'dzclass', 'income', 'race', 'ca']\n",
+    "num_feats = ['age', 'num.co', 'meanbp', 'wblc', 'hrt', 'resp', \n",
+    "\t     'temp', 'pafi', 'alb', 'bili', 'crea', 'sod', 'ph', \n",
+    "             'glucose', 'bun', 'urine', 'adlp', 'adls']\n",
+    "\n",
+    "features = Preprocessor().fit_transform(features, cat_feats=cat_feats, num_feats=num_feats)"
    ]
   },
   {
@@ -54,13 +72,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [],
    "source": [
     "import numpy as np\n",
     "horizons = [0.25, 0.5, 0.75]\n",
-    "times = np.quantile(t[e==1], horizons).tolist()"
+    "times = np.quantile(outcomes.time[outcomes.event==1], horizons).tolist()"
    ]
   },
   {
@@ -74,10 +92,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [],
    "source": [
+    "x, t, e = features.values, outcomes.time.values, outcomes.event.values\n",
+    "\n",
     "n = len(x)\n",
     "\n",
     "tr_size = int(n*0.70)\n",
@@ -101,7 +121,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -110,7 +130,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 32,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -131,20 +151,99 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 33,
    "metadata": {},
    "outputs": [],
    "source": [
-    "from dsm import DeepSurvivalMachines"
+    "from auton_survival.models.dsm import DeepSurvivalMachines\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 34,
    "metadata": {
     "scrolled": true
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      " 12%|█▏        | 1242/10000 [00:02<00:15, 576.61it/s]\n",
+      "100%|██████████| 100/100 [00:13<00:00,  7.26it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:14, 589.45it/s]\n",
+      " 88%|████████▊ | 88/100 [00:11<00:01,  7.37it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:05<00:35, 247.08it/s]\n",
+      "100%|██████████| 100/100 [00:19<00:00,  5.25it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:17, 501.23it/s]\n",
+      " 13%|█▎        | 13/100 [00:02<00:14,  5.81it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:15, 574.96it/s]\n",
+      " 43%|████▎     | 43/100 [00:07<00:09,  5.83it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:14, 587.50it/s]\n",
+      "  9%|▉         | 9/100 [00:03<00:32,  2.80it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:03<00:21, 407.46it/s]\n",
+      "100%|██████████| 100/100 [00:20<00:00,  4.79it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:19, 453.10it/s]\n",
+      " 90%|█████████ | 90/100 [00:14<00:01,  6.13it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:01<00:13, 629.44it/s]\n",
+      " 96%|█████████▌| 96/100 [00:18<00:00,  5.22it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:15, 553.09it/s]\n",
+      " 14%|█▍        | 14/100 [00:02<00:17,  5.03it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:14, 603.60it/s]\n",
+      " 43%|████▎     | 43/100 [00:10<00:13,  4.19it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:20, 417.17it/s]\n",
+      "  8%|▊         | 8/100 [00:03<00:36,  2.49it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:04<00:28, 304.82it/s]\n",
+      "100%|██████████| 100/100 [00:27<00:00,  3.61it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:03<00:24, 351.58it/s]\n",
+      " 88%|████████▊ | 88/100 [00:20<00:02,  4.26it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:17, 514.35it/s]\n",
+      "100%|██████████| 100/100 [00:27<00:00,  3.60it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:17, 489.70it/s]\n",
+      " 14%|█▍        | 14/100 [00:04<00:28,  2.97it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:15, 568.87it/s]\n",
+      " 49%|████▉     | 49/100 [00:13<00:13,  3.67it/s]\n",
+      " 12%|█▏        | 1242/10000 [00:02<00:19, 451.42it/s]\n",
+      "  8%|▊         | 8/100 [00:02<00:29,  3.14it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:02<00:11, 710.14it/s]\n",
+      "100%|██████████| 100/100 [00:12<00:00,  7.71it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:03<00:13, 602.29it/s]\n",
+      " 80%|████████  | 80/100 [00:10<00:02,  7.33it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:04<00:20, 406.57it/s]\n",
+      "100%|██████████| 100/100 [00:21<00:00,  4.57it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:03<00:13, 594.83it/s]\n",
+      " 15%|█▌        | 15/100 [00:02<00:15,  5.58it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:03<00:13, 611.51it/s]\n",
+      " 46%|████▌     | 46/100 [00:07<00:08,  6.14it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:03<00:14, 549.49it/s]\n",
+      " 10%|█         | 10/100 [00:01<00:15,  5.66it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:03<00:13, 602.64it/s]\n",
+      "100%|██████████| 100/100 [00:13<00:00,  7.45it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:02<00:11, 684.95it/s]\n",
+      "100%|██████████| 100/100 [00:13<00:00,  7.48it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:02<00:12, 661.44it/s]\n",
+      "100%|██████████| 100/100 [00:14<00:00,  6.74it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:03<00:13, 590.15it/s]\n",
+      " 14%|█▍        | 14/100 [00:02<00:13,  6.15it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:02<00:11, 682.44it/s]\n",
+      " 51%|█████     | 51/100 [00:08<00:08,  5.97it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:02<00:12, 635.57it/s]\n",
+      " 11%|█         | 11/100 [00:02<00:16,  5.28it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:02<00:11, 695.68it/s]\n",
+      "100%|██████████| 100/100 [00:16<00:00,  6.07it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:02<00:12, 632.76it/s]\n",
+      "100%|██████████| 100/100 [00:17<00:00,  5.70it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:02<00:12, 655.98it/s]\n",
+      "100%|██████████| 100/100 [00:23<00:00,  4.35it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:04<00:18, 443.22it/s]\n",
+      " 14%|█▍        | 14/100 [00:03<00:22,  3.85it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:03<00:14, 574.25it/s]\n",
+      " 54%|█████▍    | 54/100 [00:13<00:11,  4.06it/s]\n",
+      " 18%|█▊        | 1845/10000 [00:03<00:16, 488.72it/s]\n",
+      "  9%|▉         | 9/100 [00:02<00:26,  3.44it/s]\n"
+     ]
+    }
+   ],
    "source": [
     "models = []\n",
     "for param in params:\n",
@@ -167,7 +266,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 35,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -186,7 +285,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 36,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -195,9 +294,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 37,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "For 0.25 quantile,\n",
+      "TD Concordance Index: 0.7758488928305234\n",
+      "Brier Score: 0.10949921706762251\n",
+      "ROC AUC  0.7840977530125496 \n",
+      "\n",
+      "For 0.5 quantile,\n",
+      "TD Concordance Index: 0.7084956219062457\n",
+      "Brier Score: 0.18207491533147557\n",
+      "ROC AUC  0.7298316693618319 \n",
+      "\n",
+      "For 0.75 quantile,\n",
+      "TD Concordance Index: 0.656044903399766\n",
+      "Brier Score: 0.22306554875644452\n",
+      "ROC AUC  0.7097073968619374 \n",
+      "\n"
+     ]
+    }
+   ],
    "source": [
     "cis = []\n",
     "brs = []\n",
@@ -221,6 +342,13 @@
     "    print(\"Brier Score:\", brs[0][horizon[0]])\n",
     "    print(\"ROC AUC \", roc_auc[horizon[0]][0], \"\\n\")"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {
@@ -239,7 +367,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.3"
+   "version": "3.8.12"
   }
  },
  "nbformat": 4,
diff --git a/examples/Phenotyping Survival Data.ipynb b/examples/Phenotyping Survival Data.ipynb
new file mode 100644
index 0000000..319c0eb
--- /dev/null
+++ b/examples/Phenotyping Survival Data.ipynb	
@@ -0,0 +1,152 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import sys\n",
+    "\n",
+    "sys.path.append('../')\n",
+    "from auton_survival import datasets\n",
+    "outcomes, features = datasets.load_support()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "horizons = [0.25, 0.5, 0.75]\n",
+    "times = np.quantile(outcomes.time[outcomes.event==1], horizons).tolist()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from auton_survival.phenotyping import IntersectionalPhenotyper"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/chiragn/Research/auton-survival/examples/../auton_survival/phenotyping.py:79: SettingWithCopyWarning: \n",
+      "A value is trying to be set on a copy of a slice from a DataFrame\n",
+      "\n",
+      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
+      "  features[num_var][features[num_var]>=var_max] = var_max\n",
+      "/Users/chiragn/Research/auton-survival/examples/../auton_survival/phenotyping.py:80: SettingWithCopyWarning: \n",
+      "A value is trying to be set on a copy of a slice from a DataFrame\n",
+      "\n",
+      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
+      "  features[num_var][features[num_var]<=var_min] = var_min\n"
+     ]
+    }
+   ],
+   "source": [
+    "phenotypes = IntersectionalPhenotyper(cat_vars=['race'], num_vars=['age']).fit_phenotype(features)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from auton_survival import reporting\n",
+    "\n",
+    "reporting.plot_kaplanmeier(outcomes, phenotypes)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from auton_survival.preprocessing import Preprocessor\n",
+    "\n",
+    "cat_feats = ['sex', 'dzgroup', 'dzclass', 'income', 'race', 'ca']\n",
+    "num_feats = ['age', 'num.co', 'meanbp', 'wblc', 'hrt', 'resp', \n",
+    "\t     'temp', 'pafi', 'alb', 'bili', 'crea', 'sod', 'ph', \n",
+    "             'glucose', 'bun', 'urine', 'adlp', 'adls']\n",
+    "\n",
+    "features_processed = Preprocessor().fit_transform(features, cat_feats=cat_feats, num_feats=num_feats)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Fitting the following Dimensionality Reduction Model:\n",
+      " PCA(n_components=8)\n",
+      "Fitting the following Clustering Model:\n",
+      " KMeans(n_clusters=3)\n"
+     ]
+    }
+   ],
+   "source": [
+    "from auton_survival.phenotyping import ClusteringPhenotyper\n",
+    "\n",
+    "phenotypes = ClusteringPhenotyper(n_clusters=3, dim_red_method='pca', n_components=8).fit_phenotype(features_processed)\n",
+    "\n",
+    "reporting.plot_kaplanmeier(outcomes, phenotypes.argmax(axis=1))"
+   ]
+  }
+ ],
+ "metadata": {
+  "interpreter": {
+   "hash": "c22fbbe4c37d04364aa4e785d7cd9729f94ca3cb878d5f955e35b0076c04a2d7"
+  },
+  "kernelspec": {
+   "display_name": "Python 3.8.12 ('base')",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.12"
+  },
+  "orig_nbformat": 4
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}