Chapter 10 – Introduction to Artificial Neural Networks with Keras
This notebook contains all the sample code and solutions to the exercises in chapter 10.
 Run in Google Colab
|
Setup
First, let’s import a few common modules, ensure MatplotLib plots figures inline and prepare a function to save the figures. We also check that Python 3.5 or later is installed (although Python 2.x may work, it is deprecated so we strongly recommend you use Python 3 instead), as well as Scikit-Learn ≥0.20 and TensorFlow ≥2.0.
# Python ≥3.5 is required
import sys
assert sys.version_info >= (3, 5)
# Scikit-Learn ≥0.20 is required
import sklearn
assert sklearn.__version__ >= "0.20"
try:
# %tensorflow_version only exists in Colab.
%tensorflow_version 2.x
except Exception:
pass
# TensorFlow ≥2.0 is required
import tensorflow as tf
assert tf.__version__ >= "2.0"
# Common imports
import numpy as np
import os
# to make this notebook's output stable across runs
np.random.seed(42)
# To plot pretty figures
%matplotlib inline
import matplotlib as mpl
import matplotlib.pyplot as plt
mpl.rc('axes', labelsize=14)
mpl.rc('xtick', labelsize=12)
mpl.rc('ytick', labelsize=12)
# Where to save the figures
PROJECT_ROOT_DIR = "."
CHAPTER_ID = "ann"
IMAGES_PATH = os.path.join(PROJECT_ROOT_DIR, "images", CHAPTER_ID)
os.makedirs(IMAGES_PATH, exist_ok=True)
def save_fig(fig_id, tight_layout=True, fig_extension="png", resolution=300):
path = os.path.join(IMAGES_PATH, fig_id + "." + fig_extension)
print("Saving figure", fig_id)
if tight_layout:
plt.tight_layout()
plt.savefig(path, format=fig_extension, dpi=resolution)
# Ignore useless warnings (see SciPy issue #5998)
import warnings
warnings.filterwarnings(action="ignore", message="^internal gelsd")
Perceptrons
Note: we set max_iter and tol explicitly to avoid warnings about the fact that their default value will change in future versions of Scikit-Learn.
import numpy as np
from sklearn.datasets import load_iris
from sklearn.linear_model import Perceptron
iris = load_iris()
X = iris.data[:, (2, 3)] # petal length, petal width
y = (iris.target == 0).astype(np.int)
per_clf = Perceptron(max_iter=1000, tol=1e-3, random_state=42)
per_clf.fit(X, y)
y_pred = per_clf.predict([[2, 0.5]])
y_pred
a = -per_clf.coef_[0][0] / per_clf.coef_[0][1]
b = -per_clf.intercept_ / per_clf.coef_[0][1]
axes = [0, 5, 0, 2]
x0, x1 = np.meshgrid(
np.linspace(axes[0], axes[1], 500).reshape(-1, 1),
np.linspace(axes[2], axes[3], 200).reshape(-1, 1),
)
X_new = np.c_[x0.ravel(), x1.ravel()]
y_predict = per_clf.predict(X_new)
zz = y_predict.reshape(x0.shape)
plt.figure(figsize=(10, 4))
plt.plot(X[y==0, 0], X[y==0, 1], "bs", label="Not Iris-Setosa")
plt.plot(X[y==1, 0], X[y==1, 1], "yo", label="Iris-Setosa")
plt.plot([axes[0], axes[1]], [a * axes[0] + b, a * axes[1] + b], "k-", linewidth=3)
from matplotlib.colors import ListedColormap
custom_cmap = ListedColormap(['#9898ff', '#fafab0'])
plt.contourf(x0, x1, zz, cmap=custom_cmap)
plt.xlabel("Petal length", fontsize=14)
plt.ylabel("Petal width", fontsize=14)
plt.legend(loc="lower right", fontsize=14)
plt.axis(axes)
save_fig("perceptron_iris_plot")
plt.show()
Activation functions
def sigmoid(z):
return 1 / (1 + np.exp(-z))
def relu(z):
return np.maximum(0, z)
def derivative(f, z, eps=0.000001):
return (f(z + eps) - f(z - eps))/(2 * eps)
z = np.linspace(-5, 5, 200)
plt.figure(figsize=(11,4))
plt.subplot(121)
plt.plot(z, np.sign(z), "r-", linewidth=1, label="Step")
plt.plot(z, sigmoid(z), "g--", linewidth=2, label="Sigmoid")
plt.plot(z, np.tanh(z), "b-", linewidth=2, label="Tanh")
plt.plot(z, relu(z), "m-.", linewidth=2, label="ReLU")
plt.grid(True)
plt.legend(loc="center right", fontsize=14)
plt.title("Activation functions", fontsize=14)
plt.axis([-5, 5, -1.2, 1.2])
plt.subplot(122)
plt.plot(z, derivative(np.sign, z), "r-", linewidth=1, label="Step")
plt.plot(0, 0, "ro", markersize=5)
plt.plot(0, 0, "rx", markersize=10)
plt.plot(z, derivative(sigmoid, z), "g--", linewidth=2, label="Sigmoid")
plt.plot(z, derivative(np.tanh, z), "b-", linewidth=2, label="Tanh")
plt.plot(z, derivative(relu, z), "m-.", linewidth=2, label="ReLU")
plt.grid(True)
#plt.legend(loc="center right", fontsize=14)
plt.title("Derivatives", fontsize=14)
plt.axis([-5, 5, -0.2, 1.2])
save_fig("activation_functions_plot")
plt.show()
def heaviside(z):
return (z >= 0).astype(z.dtype)
def mlp_xor(x1, x2, activation=heaviside):
return activation(-activation(x1 + x2 - 1.5) + activation(x1 + x2 - 0.5) - 0.5)
x1s = np.linspace(-0.2, 1.2, 100)
x2s = np.linspace(-0.2, 1.2, 100)
x1, x2 = np.meshgrid(x1s, x2s)
z1 = mlp_xor(x1, x2, activation=heaviside)
z2 = mlp_xor(x1, x2, activation=sigmoid)
plt.figure(figsize=(10,4))
plt.subplot(121)
plt.contourf(x1, x2, z1)
plt.plot([0, 1], [0, 1], "gs", markersize=20)
plt.plot([0, 1], [1, 0], "y^", markersize=20)
plt.title("Activation function: heaviside", fontsize=14)
plt.grid(True)
plt.subplot(122)
plt.contourf(x1, x2, z2)
plt.plot([0, 1], [0, 1], "gs", markersize=20)
plt.plot([0, 1], [1, 0], "y^", markersize=20)
plt.title("Activation function: sigmoid", fontsize=14)
plt.grid(True)
Building an Image Classifier
First let’s import TensorFlow and Keras.
import tensorflow as tf
from tensorflow import keras
tf.__version__
keras.__version__
Let’s start by loading the fashion MNIST dataset. Keras has a number of functions to load popular datasets in keras.datasets. The dataset is already split for you between a training set and a test set, but it can be useful to split the training set further to have a validation set:
fashion_mnist = keras.datasets.fashion_mnist
(X_train_full, y_train_full), (X_test, y_test) = fashion_mnist.load_data()
The training set contains 60,000 grayscale images, each 28x28 pixels:
X_train_full.shape
Each pixel intensity is represented as a byte (0 to 255):
X_train_full.dtype
Let’s split the full training set into a validation set and a (smaller) training set. We also scale the pixel intensities down to the 0-1 range and convert them to floats, by dividing by 255.
X_valid, X_train = X_train_full[:5000] / 255., X_train_full[5000:] / 255.
y_valid, y_train = y_train_full[:5000], y_train_full[5000:]
X_test = X_test / 255.
You can plot an image using Matplotlib’s imshow() function, with a 'binary'
color map:
plt.imshow(X_train[0], cmap="binary")
plt.axis('off')
plt.show()
The labels are the class IDs (represented as uint8), from 0 to 9:
y_train
Here are the corresponding class names:
class_names = ["T-shirt/top", "Trouser", "Pullover", "Dress", "Coat",
"Sandal", "Shirt", "Sneaker", "Bag", "Ankle boot"]
So the first image in the training set is a coat:
class_names[y_train[0]]
The validation set contains 5,000 images, and the test set contains 10,000 images:
X_valid.shape
X_test.shape
Let’s take a look at a sample of the images in the dataset:
n_rows = 4
n_cols = 10
plt.figure(figsize=(n_cols * 1.2, n_rows * 1.2))
for row in range(n_rows):
for col in range(n_cols):
index = n_cols * row + col
plt.subplot(n_rows, n_cols, index + 1)
plt.imshow(X_train[index], cmap="binary", interpolation="nearest")
plt.axis('off')
plt.title(class_names[y_train[index]], fontsize=12)
plt.subplots_adjust(wspace=0.2, hspace=0.5)
save_fig('fashion_mnist_plot', tight_layout=False)
plt.show()
model = keras.models.Sequential()
model.add(keras.layers.Flatten(input_shape=[28, 28]))
model.add(keras.layers.Dense(300, activation="relu"))
model.add(keras.layers.Dense(100, activation="relu"))
model.add(keras.layers.Dense(10, activation="softmax"))
keras.backend.clear_session()
np.random.seed(42)
tf.random.set_seed(42)
model = keras.models.Sequential([
keras.layers.Flatten(input_shape=[28, 28]),
keras.layers.Dense(300, activation="relu"),
keras.layers.Dense(100, activation="relu"),
keras.layers.Dense(10, activation="softmax")
])
model.layers
model.summary()
keras.utils.plot_model(model, "my_mnist_model.png", show_shapes=True)
hidden1 = model.layers[1]
hidden1.name
model.get_layer(hidden1.name) is hidden1
weights, biases = hidden1.get_weights()
weights
weights.shape
biases
biases.shape
model.compile(loss="sparse_categorical_crossentropy",
optimizer="sgd",
metrics=["accuracy"])
This is equivalent to:
model.compile(loss=keras.losses.sparse_categorical_crossentropy,
optimizer=keras.optimizers.SGD(),
metrics=[keras.metrics.sparse_categorical_accuracy])
history = model.fit(X_train, y_train, epochs=30,
validation_data=(X_valid, y_valid))
history.params
print(history.epoch)
history.history.keys()
import pandas as pd
pd.DataFrame(history.history).plot(figsize=(8, 5))
plt.grid(True)
plt.gca().set_ylim(0, 1)
save_fig("keras_learning_curves_plot")
plt.show()
model.evaluate(X_test, y_test)
X_new = X_test[:3]
y_proba = model.predict(X_new)
y_proba.round(2)
y_pred = model.predict_classes(X_new)
y_pred
np.array(class_names)[y_pred]
y_new = y_test[:3]
y_new
plt.figure(figsize=(7.2, 2.4))
for index, image in enumerate(X_new):
plt.subplot(1, 3, index + 1)
plt.imshow(image, cmap="binary", interpolation="nearest")
plt.axis('off')
plt.title(class_names[y_test[index]], fontsize=12)
plt.subplots_adjust(wspace=0.2, hspace=0.5)
save_fig('fashion_mnist_images_plot', tight_layout=False)
plt.show()
Regression MLP
Let’s load, split and scale the California housing dataset (the original one, not the modified one as in chapter 2):
from sklearn.datasets import fetch_california_housing
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
housing = fetch_california_housing()
X_train_full, X_test, y_train_full, y_test = train_test_split(housing.data, housing.target, random_state=42)
X_train, X_valid, y_train, y_valid = train_test_split(X_train_full, y_train_full, random_state=42)
scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_valid = scaler.transform(X_valid)
X_test = scaler.transform(X_test)
np.random.seed(42)
tf.random.set_seed(42)
model = keras.models.Sequential([
keras.layers.Dense(30, activation="relu", input_shape=X_train.shape[1:]),
keras.layers.Dense(1)
])
model.compile(loss="mean_squared_error", optimizer=keras.optimizers.SGD(lr=1e-3))
history = model.fit(X_train, y_train, epochs=20, validation_data=(X_valid, y_valid))
mse_test = model.evaluate(X_test, y_test)
X_new = X_test[:3]
y_pred = model.predict(X_new)
plt.plot(pd.DataFrame(history.history))
plt.grid(True)
plt.gca().set_ylim(0, 1)
plt.show()
y_pred
Functional API
Not all neural network models are simply sequential. Some may have complex topologies. Some may have multiple inputs and/or multiple outputs. For example, a Wide & Deep neural network (see paper) connects all or part of the inputs directly to the output layer.
np.random.seed(42)
tf.random.set_seed(42)
input_ = keras.layers.Input(shape=X_train.shape[1:])
hidden1 = keras.layers.Dense(30, activation="relu")(input_)
hidden2 = keras.layers.Dense(30, activation="relu")(hidden1)
concat = keras.layers.concatenate([input_, hidden2])
output = keras.layers.Dense(1)(concat)
model = keras.models.Model(inputs=[input_], outputs=[output])
model.summary()
model.compile(loss="mean_squared_error", optimizer=keras.optimizers.SGD(lr=1e-3))
history = model.fit(X_train, y_train, epochs=20,
validation_data=(X_valid, y_valid))
mse_test = model.evaluate(X_test, y_test)
y_pred = model.predict(X_new)
What if you want to send different subsets of input features through the wide or deep paths? We will send 5 features (features 0 to 4), and 6 through the deep path (features 2 to 7). Note that 3 features will go through both (features 2, 3 and 4).
np.random.seed(42)
tf.random.set_seed(42)
input_A = keras.layers.Input(shape=[5], name="wide_input")
input_B = keras.layers.Input(shape=[6], name="deep_input")
hidden1 = keras.layers.Dense(30, activation="relu")(input_B)
hidden2 = keras.layers.Dense(30, activation="relu")(hidden1)
concat = keras.layers.concatenate([input_A, hidden2])
output = keras.layers.Dense(1, name="output")(concat)
model = keras.models.Model(inputs=[input_A, input_B], outputs=[output])
model.compile(loss="mse", optimizer=keras.optimizers.SGD(lr=1e-3))
X_train_A, X_train_B = X_train[:, :5], X_train[:, 2:]
X_valid_A, X_valid_B = X_valid[:, :5], X_valid[:, 2:]
X_test_A, X_test_B = X_test[:, :5], X_test[:, 2:]
X_new_A, X_new_B = X_test_A[:3], X_test_B[:3]
history = model.fit((X_train_A, X_train_B), y_train, epochs=20,
validation_data=((X_valid_A, X_valid_B), y_valid))
mse_test = model.evaluate((X_test_A, X_test_B), y_test)
y_pred = model.predict((X_new_A, X_new_B))
Adding an auxiliary output for regularization:
np.random.seed(42)
tf.random.set_seed(42)
input_A = keras.layers.Input(shape=[5], name="wide_input")
input_B = keras.layers.Input(shape=[6], name="deep_input")
hidden1 = keras.layers.Dense(30, activation="relu")(input_B)
hidden2 = keras.layers.Dense(30, activation="relu")(hidden1)
concat = keras.layers.concatenate([input_A, hidden2])
output = keras.layers.Dense(1, name="main_output")(concat)
aux_output = keras.layers.Dense(1, name="aux_output")(hidden2)
model = keras.models.Model(inputs=[input_A, input_B],
outputs=[output, aux_output])
model.compile(loss=["mse", "mse"], loss_weights=[0.9, 0.1], optimizer=keras.optimizers.SGD(lr=1e-3))
history = model.fit([X_train_A, X_train_B], [y_train, y_train], epochs=20,
validation_data=([X_valid_A, X_valid_B], [y_valid, y_valid]))
total_loss, main_loss, aux_loss = model.evaluate(
[X_test_A, X_test_B], [y_test, y_test])
y_pred_main, y_pred_aux = model.predict([X_new_A, X_new_B])
The subclassing API
class WideAndDeepModel(keras.models.Model):
def __init__(self, units=30, activation="relu", **kwargs):
super().__init__(**kwargs)
self.hidden1 = keras.layers.Dense(units, activation=activation)
self.hidden2 = keras.layers.Dense(units, activation=activation)
self.main_output = keras.layers.Dense(1)
self.aux_output = keras.layers.Dense(1)
def call(self, inputs):
input_A, input_B = inputs
hidden1 = self.hidden1(input_B)
hidden2 = self.hidden2(hidden1)
concat = keras.layers.concatenate([input_A, hidden2])
main_output = self.main_output(concat)
aux_output = self.aux_output(hidden2)
return main_output, aux_output
model = WideAndDeepModel(30, activation="relu")
model.compile(loss="mse", loss_weights=[0.9, 0.1], optimizer=keras.optimizers.SGD(lr=1e-3))
history = model.fit((X_train_A, X_train_B), (y_train, y_train), epochs=10,
validation_data=((X_valid_A, X_valid_B), (y_valid, y_valid)))
total_loss, main_loss, aux_loss = model.evaluate((X_test_A, X_test_B), (y_test, y_test))
y_pred_main, y_pred_aux = model.predict((X_new_A, X_new_B))
model = WideAndDeepModel(30, activation="relu")
Saving and Restoring
np.random.seed(42)
tf.random.set_seed(42)
model = keras.models.Sequential([
keras.layers.Dense(30, activation="relu", input_shape=[8]),
keras.layers.Dense(30, activation="relu"),
keras.layers.Dense(1)
])
model.compile(loss="mse", optimizer=keras.optimizers.SGD(lr=1e-3))
history = model.fit(X_train, y_train, epochs=10, validation_data=(X_valid, y_valid))
mse_test = model.evaluate(X_test, y_test)
model.save("my_keras_model.h5")
model = keras.models.load_model("my_keras_model.h5")
model.predict(X_new)
model.save_weights("my_keras_weights.ckpt")
model.load_weights("my_keras_weights.ckpt")
Using Callbacks during Training
keras.backend.clear_session()
np.random.seed(42)
tf.random.set_seed(42)
model = keras.models.Sequential([
keras.layers.Dense(30, activation="relu", input_shape=[8]),
keras.layers.Dense(30, activation="relu"),
keras.layers.Dense(1)
])
model.compile(loss="mse", optimizer=keras.optimizers.SGD(lr=1e-3))
checkpoint_cb = keras.callbacks.ModelCheckpoint("my_keras_model.h5", save_best_only=True)
history = model.fit(X_train, y_train, epochs=10,
validation_data=(X_valid, y_valid),
callbacks=[checkpoint_cb])
model = keras.models.load_model("my_keras_model.h5") # rollback to best model
mse_test = model.evaluate(X_test, y_test)
model.compile(loss="mse", optimizer=keras.optimizers.SGD(lr=1e-3))
early_stopping_cb = keras.callbacks.EarlyStopping(patience=10,
restore_best_weights=True)
history = model.fit(X_train, y_train, epochs=100,
validation_data=(X_valid, y_valid),
callbacks=[checkpoint_cb, early_stopping_cb])
mse_test = model.evaluate(X_test, y_test)
class PrintValTrainRatioCallback(keras.callbacks.Callback):
def on_epoch_end(self, epoch, logs):
print("\nval/train: {:.2f}".format(logs["val_loss"] / logs["loss"]))
val_train_ratio_cb = PrintValTrainRatioCallback()
history = model.fit(X_train, y_train, epochs=1,
validation_data=(X_valid, y_valid),
callbacks=[val_train_ratio_cb])
TensorBoard
root_logdir = os.path.join(os.curdir, "my_logs")
def get_run_logdir():
import time
run_id = time.strftime("run_%Y_%m_%d-%H_%M_%S")
return os.path.join(root_logdir, run_id)
run_logdir = get_run_logdir()
run_logdir
keras.backend.clear_session()
np.random.seed(42)
tf.random.set_seed(42)
model = keras.models.Sequential([
keras.layers.Dense(30, activation="relu", input_shape=[8]),
keras.layers.Dense(30, activation="relu"),
keras.layers.Dense(1)
])
model.compile(loss="mse", optimizer=keras.optimizers.SGD(lr=1e-3))
tensorboard_cb = keras.callbacks.TensorBoard(run_logdir)
history = model.fit(X_train, y_train, epochs=30,
validation_data=(X_valid, y_valid),
callbacks=[checkpoint_cb, tensorboard_cb])
To start the TensorBoard server, one option is to open a terminal, if needed activate the virtualenv where you installed TensorBoard, go to this notebook’s directory, then type:
$ tensorboard --logdir=./my_logs --port=6006
You can then open your web browser to localhost:6006 and use TensorBoard. Once you are done, press Ctrl-C in the terminal window, this will shutdown the TensorBoard server.
Alternatively, you can load TensorBoard’s Jupyter extension and run it like this:
%load_ext tensorboard
%tensorboard --logdir=./my_logs --port=6006
run_logdir2 = get_run_logdir()
run_logdir2
keras.backend.clear_session()
np.random.seed(42)
tf.random.set_seed(42)
model = keras.models.Sequential([
keras.layers.Dense(30, activation="relu", input_shape=[8]),
keras.layers.Dense(30, activation="relu"),
keras.layers.Dense(1)
])
model.compile(loss="mse", optimizer=keras.optimizers.SGD(lr=0.05))
tensorboard_cb = keras.callbacks.TensorBoard(run_logdir2)
history = model.fit(X_train, y_train, epochs=30,
validation_data=(X_valid, y_valid),
callbacks=[checkpoint_cb, tensorboard_cb])
Notice how TensorBoard now sees two runs, and you can compare the learning curves.
Check out the other available logging options:
help(keras.callbacks.TensorBoard.__init__)
Hyperparameter Tuning
keras.backend.clear_session()
np.random.seed(42)
tf.random.set_seed(42)
def build_model(n_hidden=1, n_neurons=30, learning_rate=3e-3, input_shape=[8]):
model = keras.models.Sequential()
model.add(keras.layers.InputLayer(input_shape=input_shape))
for layer in range(n_hidden):
model.add(keras.layers.Dense(n_neurons, activation="relu"))
model.add(keras.layers.Dense(1))
optimizer = keras.optimizers.SGD(lr=learning_rate)
model.compile(loss="mse", optimizer=optimizer)
return model
keras_reg = keras.wrappers.scikit_learn.KerasRegressor(build_model)
keras_reg.fit(X_train, y_train, epochs=100,
validation_data=(X_valid, y_valid),
callbacks=[keras.callbacks.EarlyStopping(patience=10)])
mse_test = keras_reg.score(X_test, y_test)
y_pred = keras_reg.predict(X_new)
np.random.seed(42)
tf.random.set_seed(42)
from scipy.stats import reciprocal
from sklearn.model_selection import RandomizedSearchCV
param_distribs = {
"n_hidden": [0, 1, 2, 3],
"n_neurons": np.arange(1, 100),
"learning_rate": reciprocal(3e-4, 3e-2),
}
rnd_search_cv = RandomizedSearchCV(keras_reg, param_distribs, n_iter=10, cv=3, verbose=2)
rnd_search_cv.fit(X_train, y_train, epochs=100,
validation_data=(X_valid, y_valid),
callbacks=[keras.callbacks.EarlyStopping(patience=10)])
rnd_search_cv.best_params_
rnd_search_cv.best_score_
rnd_search_cv.best_estimator_
rnd_search_cv.score(X_test, y_test)
model = rnd_search_cv.best_estimator_.model
model
model.evaluate(X_test, y_test)
Exercise solutions
1. to 9.
See appendix A.
10.
TODO