Skip to content

Instantly share code, notes, and snippets.

@jgc128

jgc128/test_keras.ipynb

Created October 26, 2015 02:49
Show Gist options
  • Save jgc128/0bad81b40787f15f6ced to your computer and use it in GitHub Desktop.
Save jgc128/0bad81b40787f15f6ced to your computer and use it in GitHub Desktop.
Download ZIP
Test Keras library
Display the source blob
Display the rendered blob
Raw
test_keras.ipynb
{
"cells": [
{
"cell_type": "code",
"execution_count": 79,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"import numpy as np\n",
"\n",
"from sklearn import datasets\n",
"from sklearn.cross_validation import train_test_split\n",
"\n",
"import theano.tensor as T\n",
"import theano\n",
"\n",
"from keras.models import Sequential\n",
"from keras.layers.core import Dense, Activation\n",
"from keras.optimizers import SGD\n",
"from keras.regularizers import l2, activity_l2"
]
},
{
"cell_type": "code",
"execution_count": 226,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"input_size = 4\n",
"hidden_size = 16\n",
"output_size = 3"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Create dataset"
]
},
{
"cell_type": "code",
"execution_count": 212,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# load iris dataset\n",
"iris = datasets.load_iris()\n",
"X = iris.data\n",
"Y = iris.target"
]
},
{
"cell_type": "code",
"execution_count": 227,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"# X1 = X[:,0:2]\n",
"X1 = X"
]
},
{
"cell_type": "code",
"execution_count": 214,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# create one-hot representation for Y\n",
"Y1 = np.zeros((len(Y), output_size), dtype=np.int)\n",
"Y1[np.arange(len(Y)), np.array(Y)] = 1"
]
},
{
"cell_type": "code",
"execution_count": 215,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# # create classifier for Y = 0\n",
"# Y1 = np.copy(Y)\n",
"# Y1[Y==0] = 1\n",
"# Y1[Y!=0] = 0"
]
},
{
"cell_type": "code",
"execution_count": 216,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([[1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1]])"
]
},
"execution_count": 216,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"Y1"
]
},
{
"cell_type": "code",
"execution_count": 217,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"X_train, X_test, Y_train, Y_test = train_test_split(X1, Y1, test_size=0.2, random_state=42)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Create Neural Network\n",
"==========="
]
},
{
"cell_type": "code",
"execution_count": 218,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"def class_activation_func(x):\n",
" return T.switch(T.lt(x, 0.5), 0.0, 1.0)"
]
},
{
"cell_type": "code",
"execution_count": 228,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"model = Sequential()\n",
"\n",
"# input\n",
"model.add(Dense(output_dim=hidden_size, input_dim=input_size, activation='sigmoid'))\n",
"\n",
"# hidden layers\n",
"model.add(Dense(output_dim=hidden_size, activation='sigmoid'))\n",
"\n",
"# output\n",
"model.add(Dense(output_dim=output_size, activation='sigmoid'))\n",
"# model.add(Activation(class_activation_func))"
]
},
{
"cell_type": "code",
"execution_count": 229,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"model.compile(class_mode='categorical', loss='categorical_crossentropy', optimizer='sgd')"
]
},
{
"cell_type": "code",
"execution_count": 230,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/10\n",
"120/120 [==============================] - 0s - loss: 1.1130 - acc: 0.3333\n",
"Epoch 2/10\n",
"120/120 [==============================] - 0s - loss: 1.1129 - acc: 0.3333\n",
"Epoch 3/10\n",
"120/120 [==============================] - 0s - loss: 1.1128 - acc: 0.3333\n",
"Epoch 4/10\n",
"120/120 [==============================] - 0s - loss: 1.1127 - acc: 0.3333\n",
"Epoch 5/10\n",
"120/120 [==============================] - 0s - loss: 1.1126 - acc: 0.3333\n",
"Epoch 6/10\n",
"120/120 [==============================] - 0s - loss: 1.1126 - acc: 0.3333\n",
"Epoch 7/10\n",
"120/120 [==============================] - 0s - loss: 1.1125 - acc: 0.3333\n",
"Epoch 8/10\n",
"120/120 [==============================] - 0s - loss: 1.1124 - acc: 0.3333\n",
"Epoch 9/10\n",
"120/120 [==============================] - 0s - loss: 1.1123 - acc: 0.3333\n",
"Epoch 10/10\n",
"120/120 [==============================] - 0s - loss: 1.1122 - acc: 0.3333\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7f9c12b27f28>"
]
},
"execution_count": 230,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model.fit(X_train, Y_train, nb_epoch=10, show_accuracy = True)\n",
"# model.fit(X, Y, nb_epoch=10, validation_split = 0.1, show_accuracy = True)"
]
},
{
"cell_type": "code",
"execution_count": 231,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"30/30 [==============================] - 0s\n"
]
},
{
"data": {
"text/plain": [
"array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
" 0, 0, 0, 0, 0, 0, 0])"
]
},
"execution_count": 231,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model.predict_classes(X_test)"
]
},
{
"cell_type": "code",
"execution_count": 232,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([[0, 1, 0],\n",
" [1, 0, 0],\n",
" [0, 0, 1],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [1, 0, 0],\n",
" [0, 1, 0],\n",
" [0, 0, 1],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 0, 1],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [1, 0, 0],\n",
" [0, 1, 0],\n",
" [0, 0, 1],\n",
" [0, 1, 0],\n",
" [0, 1, 0],\n",
" [0, 0, 1],\n",
" [1, 0, 0],\n",
" [0, 0, 1],\n",
" [1, 0, 0],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [0, 0, 1],\n",
" [1, 0, 0],\n",
" [1, 0, 0]])"
]
},
"execution_count": 232,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"Y_test"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
}
],
"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": "3.4.3"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment