StarMirror/data-science-ipython-notebooks
1
0
Fork 0
mirror of https://github.com/donnemartin/data-science-ipython-notebooks.git synced 2024-03-22 13:30:56 +08:00
Code Issues Projects Releases Wiki Activity
05dc9cffbb
data-science-ipython-notebooks/deep-learning/keras-tutorial/1.2 Introduction - Theano.ipynb

1019 lines
21 KiB
Python
Raw Normal View History

Add Keras tutorials resources #49 (#51)
2017-08-26 20:25:13 +08:00
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Credits: Forked from [deep-learning-keras-tensorflow](https://github.com/leriomaggio/deep-learning-keras-tensorflow) by Valerio Maggio"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"Theano \n",
"===\n",
"A language in a language"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Dealing with weights matrices and gradients can be tricky and sometimes not trivial.\n",
"Theano is a great framework for handling vectors, matrices and high dimensional tensor algebra. \n",
"Most of this tutorial will refer to Theano however TensorFlow is another great framework capable of providing an incredible abstraction for complex algebra.\n",
"More on TensorFlow in the next chapters."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [],
"source": [
"import theano\n",
"import theano.tensor as T"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"Symbolic variables\n",
"=========="
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Theano has it's own variables and functions, defined the following"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [],
"source": [
"x = T.scalar()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [],
"source": [
"x"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"Variables can be used in expressions"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "-"
}
},
"outputs": [],
"source": [
"y = 3*(x**2) + 1"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"y is an expression now "
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Result is symbolic as well"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "-"
}
},
"outputs": [
{
"data": {
"text/plain": [
"Shape.0"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"type(y)\n",
"y.shape"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"#####printing"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"As we are about to see, normal printing isn't the best when it comes to theano"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Elemwise{add,no_inplace}.0\n"
]
}
],
"source": [
"print(y)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"data": {
"text/plain": [
"'((TensorConstant{3} * (<TensorType(float32, scalar)> ** TensorConstant{2})) + TensorConstant{1})'"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"theano.pprint(y)"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Elemwise{add,no_inplace} [@A] '' \n",
" |Elemwise{mul,no_inplace} [@B] '' \n",
" | |TensorConstant{3} [@C]\n",
" | |Elemwise{pow,no_inplace} [@D] '' \n",
" | |<TensorType(float32, scalar)> [@E]\n",
" | |TensorConstant{2} [@F]\n",
" |TensorConstant{1} [@G]\n"
]
}
],
"source": [
"theano.printing.debugprint(y)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"Evaluating expressions\n",
"============\n",
"\n",
"Supply a `dict` mapping variables to values"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array(13.0, dtype=float32)"
]
},
"execution_count": 26,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"y.eval({x: 2})"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"Or compile a function"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"f = theano.function([x], y)"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"data": {
"text/plain": [
"array(13.0, dtype=float32)"
]
},
"execution_count": 28,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"f(2)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"Other tensor types\n",
"=========="
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {
"collapsed": true,
"slideshow": {
"slide_type": "-"
}
},
"outputs": [],
"source": [
"X = T.vector()\n",
"X = T.matrix()\n",
"X = T.tensor3()\n",
"X = T.tensor4()"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"Automatic differention\n",
"============\n",
"- Gradients are free!"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"x = T.scalar()\n",
"y = T.log(x)"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Elemwise{true_div}.0\n",
"0.5\n",
"Elemwise{mul,no_inplace}.0\n"
]
}
],
"source": [
"gradient = T.grad(y, x)\n",
"print gradient\n",
"print gradient.eval({x: 2})\n",
"print (2 * gradient)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"# Shared Variables\n",
"\n",
"- Symbolic + Storage"
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"import numpy as np\n",
"x = theano.shared(np.zeros((2, 3), dtype=theano.config.floatX))"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<CudaNdarrayType(float32, matrix)>"
]
},
"execution_count": 40,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"x"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"We can get and set the variable's value"
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "-"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(2, 3)\n",
"[[ 0. 0. 0.]\n",
" [ 0. 0. 0.]]\n"
]
}
],
"source": [
"values = x.get_value()\n",
"print(values.shape)\n",
"print(values)"
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"x.set_value(values)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"Shared variables can be used in expressions as well"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "-"
}
},
"outputs": [
{
"data": {
"text/plain": [
"Elemwise{pow,no_inplace}.0"
]
},
"execution_count": 43,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"(x + 2) ** 2"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Their value is used as input when evaluating"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([[ 4., 4., 4.],\n",
" [ 4., 4., 4.]], dtype=float32)"
]
},
"execution_count": 44,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"((x + 2) ** 2).eval()"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([[ 4., 4., 4.],\n",
" [ 4., 4., 4.]], dtype=float32)"
]
},
"execution_count": 45,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"theano.function([], (x + 2) ** 2)()"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"# Updates\n",
"\n",
"- Store results of function evalution\n",
"- `dict` mapping shared variables to new values"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "slide"
}
},
"outputs": [],
"source": [
"count = theano.shared(0)\n",
"new_count = count + 1\n",
"updates = {count: new_count}\n",
"\n",
"f = theano.function([], count, updates=updates)"
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"data": {
"text/plain": [
"array(0)"
]
},
"execution_count": 47,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"f()"
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"data": {
"text/plain": [
"array(1)"
]
},
"execution_count": 48,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"f()"
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {
"collapsed": false,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"data": {
"text/plain": [
"array(2)"
]
},
"execution_count": 49,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"f()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Warming up! Logistic Regression"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"%matplotlib inline"
]
},
{
"cell_type": "code",
"execution_count": 36,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Using Theano backend.\n"
]
}
],
"source": [
"import numpy as np\n",
"import pandas as pd\n",
"import theano\n",
"import theano.tensor as T\n",
"import matplotlib.pyplot as plt\n",
"from sklearn.preprocessing import StandardScaler\n",
"from sklearn.preprocessing import LabelEncoder \n",
"from keras.utils import np_utils"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"For this section we will use the Kaggle otto challenge.\n",
"If you want to follow, Get the data from Kaggle: https://www.kaggle.com/c/otto-group-product-classification-challenge/data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### About the data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The Otto Group is one of the world’s biggest e-commerce companies, A consistent analysis of the performance of products is crucial. However, due to diverse global infrastructure, many identical products get classified differently.\n",
"For this competition, we have provided a dataset with 93 features for more than 200,000 products. The objective is to build a predictive model which is able to distinguish between our main product categories. \n",
"Each row corresponds to a single product. There are a total of 93 numerical features, which represent counts of different events. All features have been obfuscated and will not be defined any further.\n",
"\n",
"https://www.kaggle.com/c/otto-group-product-classification-challenge/data"
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"def load_data(path, train=True):\n",
" \"\"\"Load data from a CSV File\n",
" \n",
" Parameters\n",
" ----------\n",
" path: str\n",
" The path to the CSV file\n",
" \n",
" train: bool (default True)\n",
" Decide whether or not data are *training data*.\n",
" If True, some random shuffling is applied.\n",
" \n",
" Return\n",
" ------\n",
" X: numpy.ndarray \n",
" The data as a multi dimensional array of floats\n",
" ids: numpy.ndarray\n",
" A vector of ids for each sample\n",
" \"\"\"\n",
" df = pd.read_csv(path)\n",
" X = df.values.copy()\n",
" if train:\n",
" np.random.shuffle(X) # https://youtu.be/uyUXoap67N8\n",
" X, labels = X[:, 1:-1].astype(np.float32), X[:, -1]\n",
" return X, labels\n",
" else:\n",
" X, ids = X[:, 1:].astype(np.float32), X[:, 0].astype(str)\n",
" return X, ids"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"def preprocess_data(X, scaler=None):\n",
" \"\"\"Preprocess input data by standardise features \n",
" by removing the mean and scaling to unit variance\"\"\"\n",
" if not scaler:\n",
" scaler = StandardScaler()\n",
" scaler.fit(X)\n",
" X = scaler.transform(X)\n",
" return X, scaler\n",
"\n",
"\n",
"def preprocess_labels(labels, encoder=None, categorical=True):\n",
" \"\"\"Encode labels with values among 0 and `n-classes-1`\"\"\"\n",
" if not encoder:\n",
" encoder = LabelEncoder()\n",
" encoder.fit(labels)\n",
" y = encoder.transform(labels).astype(np.int32)\n",
" if categorical:\n",
" y = np_utils.to_categorical(y)\n",
" return y, encoder"
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loading data...\n",
"[[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 3. 0. 0. 0. 3.\n",
" 2. 1. 0. 0. 0. 0. 0. 0. 0. 5. 3. 1. 1. 0.\n",
" 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 1. 0. 0.\n",
" 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
" 0. 0. 0. 0. 0. 0. 0. 3. 0. 0. 0. 0. 1. 1.\n",
" 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
" 0. 11. 1. 20. 0. 0. 0. 0. 0.]]\n",
"(9L, 'classes')\n",
"(93L, 'dims')\n"
]
}
],
"source": [
"print(\"Loading data...\")\n",
"X, labels = load_data('train.csv', train=True)\n",
"X, scaler = preprocess_data(X)\n",
"Y, encoder = preprocess_labels(labels)\n",
"\n",
"\n",
"X_test, ids = load_data('test.csv', train=False)\n",
"X_test, ids = X_test[:1000], ids[:1000]\n",
"\n",
"#Plotting the data\n",
"print(X_test[:1])\n",
"\n",
"X_test, _ = preprocess_data(X_test, scaler)\n",
"\n",
"nb_classes = Y.shape[1]\n",
"print(nb_classes, 'classes')\n",
"\n",
"dims = X.shape[1]\n",
"print(dims, 'dims')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now lets create and train a logistic regression model."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Hands On - Logistic Regression"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1\n",
"target values for Data:\n",
"[ 0. 0. 1. ..., 0. 0. 0.]\n",
"prediction on training set:\n",
"[0 0 0 ..., 0 0 0]\n"
]
}
],
"source": [
"#Based on example from DeepLearning.net\n",
"rng = np.random\n",
"N = 400\n",
"feats = 93\n",
"training_steps = 1\n",
"\n",
"# Declare Theano symbolic variables\n",
"x = T.matrix(\"x\")\n",
"y = T.vector(\"y\")\n",
"w = theano.shared(rng.randn(feats), name=\"w\")\n",
"b = theano.shared(0., name=\"b\")\n",
"\n",
"# Construct Theano expression graph\n",
"p_1 = 1 / (1 + T.exp(-T.dot(x, w) - b)) # Probability that target = 1\n",
"prediction = p_1 > 0.5 # The prediction thresholded\n",
"xent = -y * T.log(p_1) - (1-y) * T.log(1-p_1) # Cross-entropy loss function\n",
"cost = xent.mean() + 0.01 * (w ** 2).sum()# The cost to minimize\n",
"gw, gb = T.grad(cost, [w, b]) # Compute the gradient of the cost\n",
" # (we shall return to this in a\n",
" # following section of this tutorial)\n",
"\n",
"# Compile\n",
"train = theano.function(\n",
" inputs=[x,y],\n",
" outputs=[prediction, xent],\n",
" updates=((w, w - 0.1 * gw), (b, b - 0.1 * gb)),\n",
" allow_input_downcast=True)\n",
"predict = theano.function(inputs=[x], outputs=prediction, allow_input_downcast=True)\n",
"\n",
"#Transform for class1\n",
"y_class1 = []\n",
"for i in Y:\n",
" y_class1.append(i[0])\n",
"y_class1 = np.array(y_class1)\n",
"\n",
"# Train\n",
"for i in range(training_steps):\n",
" print('Epoch %s' % (i+1,))\n",
" pred, err = train(X, y_class1)\n",
"\n",
"print(\"target values for Data:\")\n",
"print(y_class1)\n",
"print(\"prediction on training set:\")\n",
"print(predict(X))"
]
},
{
"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
}
Reference in New Issue Copy Permalink