Merge pull request #41 from hammerlab/hammer_tf_backend

Add notebook that builds simple model running on TF backend

notebooks/TF exploration.ipynb

@@ -0,0 +1,185 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "os.environ['THEANO_FLAGS'] = \"'device=cpu'\"\n",
+    "os.environ['KERAS_BACKEND'] = \"tensorflow\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "%matplotlib inline\n",
+    "\n",
+    "import warnings\n",
+    "warnings.filterwarnings('ignore')\n",
+    "\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import tensorflow as tf\n",
+    "import seaborn as sns\n",
+    "\n",
+    "from mhcflurry.dataset import Dataset\n",
+    "from mhcflurry.peptide_encoding import indices_to_hotshot_encoding\n",
+    "from mhcflurry.regression_target import ic50_to_regression_target"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "file_to_explore=\"/root/.local/share/mhcflurry/2/class1_data/combined_human_class1_dataset.csv\"\n",
+    "dataset = Dataset.from_csv(\n",
+    "        filename=file_to_explore,\n",
+    "        sep=\",\",\n",
+    "        peptide_column_name=\"peptide\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "df = dataset.to_dataframe()\n",
+    "df.columns"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "df[df.species == 'human'].groupby('affinity').size().order().tail(10)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "df_reduced = df[df.allele.isin(['HLA-A0201', 'HLA-A2301', 'HLA-A2402', 'HLA-A1101'])][['allele','affinity']].reset_index(drop=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "sns.violinplot(x=df_reduced['allele'], y=np.log(df_reduced['affinity']))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "scaled_affinity = ic50_to_regression_target(df_reduced['affinity'])\n",
+    "sns.boxplot(x=df_reduced['allele'], y=scaled_affinity)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "df_reduced.groupby('allele').size()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "df_kmers = dataset.kmer_index_encoding()\n",
+    "training_hotshot = indices_to_hotshot_encoding(df_kmers[0])\n",
+    "training_labels = ic50_to_regression_target(df_kmers[1])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "from keras.models import Sequential\n",
+    "from keras.layers import Dense, Activation\n",
+    "\n",
+    "model = Sequential()\n",
+    "model.add(Dense(input_dim=189, output_dim=1))\n",
+    "model.add(Activation(\"sigmoid\"))\n",
+    "model.compile(loss=\"mse\", optimizer=\"rmsprop\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "model.fit(training_hotshot, training_labels, nb_epoch=5, batch_size=1)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 2",
+   "language": "python",
+   "name": "python2"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
+}