Learn KERAS with Real Code Examples

import numpy as np
from tensorflow import keras
from tensorflow.keras import layers

# Sample data
x_train = np.array([1, 2, 3, 4], dtype=float)
y_train = np.array([2, 4, 6, 8], dtype=float)

# Define model
model = keras.Sequential([layers.Dense(units=1, input_shape=[1])])
model.compile(optimizer='sgd', loss='mean_squared_error')

# Train the model
model.fit(x_train, y_train, epochs=500)

# Predict
y_pred = model.predict([10.0])
print("Prediction for 10:", y_pred)

A minimal Keras example for linear regression using a single dense layer.

import numpy as np
from tensorflow import keras
from tensorflow.keras import layers

# Sample data
x_train = np.random.rand(100,3)
y_train = np.random.randint(0,2,100)

# Define model
model = keras.Sequential([
    layers.Dense(8, activation='relu', input_shape=[3]),
    layers.Dense(1, activation='sigmoid')
])

model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
model.fit(x_train, y_train, epochs=50)

Basic feedforward neural network for binary classification.

from tensorflow import keras
from tensorflow.keras import layers

(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()
x_train, x_test = x_train/255.0, x_test/255.0

model = keras.Sequential([
    layers.Flatten(input_shape=(28,28)),
    layers.Dense(128, activation='relu'),
    layers.Dense(10, activation='softmax')
])

model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
model.fit(x_train, y_train, epochs=5)
model.evaluate(x_test, y_test)

Train a simple fully-connected network on MNIST digits.

from tensorflow import keras
from tensorflow.keras import layers

model = keras.Sequential([
    layers.Conv2D(32, (3,3), activation='relu', input_shape=(28,28,1)),
    layers.MaxPooling2D((2,2)),
    layers.Flatten(),
    layers.Dense(64, activation='relu'),
    layers.Dense(10, activation='softmax')
])

model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])

A simple CNN for image classification.

import numpy as np
from tensorflow import keras
from tensorflow.keras import layers

x_train = np.random.rand(100,10,1)
y_train = np.random.rand(100,1)

model = keras.Sequential([
    layers.LSTM(50, input_shape=(10,1)),
    layers.Dense(1)
])

model.compile(optimizer='adam', loss='mse')
model.fit(x_train, y_train, epochs=20)

A simple LSTM network for sequence prediction.

from tensorflow import keras
from tensorflow.keras import layers

input_dim = 20
model = keras.Sequential([
    layers.Dense(10, activation='relu', input_shape=(input_dim,)),
    layers.Dense(5, activation='relu'),
    layers.Dense(10, activation='relu'),
    layers.Dense(input_dim, activation='sigmoid')
])
model.compile(optimizer='adam', loss='mse')

A simple autoencoder for dimensionality reduction.

import numpy as np
from tensorflow import keras
from tensorflow.keras import layers

x_train = np.random.rand(100,3)
y_train = np.dot(x_train, [1.5,-2.0,1.0])+0.5

model = keras.Sequential([layers.Dense(1, input_shape=[3])])
model.compile(optimizer='sgd', loss='mse')
model.fit(x_train, y_train, epochs=100)

Linear regression with multiple features.

from tensorflow import keras
base_model = keras.applications.MobileNetV2(input_shape=(128,128,3), include_top=False, weights='imagenet')
base_model.trainable = False
model = keras.Sequential([
    base_model,
    layers.GlobalAveragePooling2D(),
    layers.Dense(10, activation='softmax')
])
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])

Use a pre-trained MobileNetV2 for image classification.

from tensorflow import keras

class PrintLossCallback(keras.callbacks.Callback):
    def on_epoch_end(self, epoch, logs=None):
        print(f"Epoch {epoch+1}: loss = {logs['loss']}")

model = keras.Sequential([keras.layers.Dense(1, input_shape=[1])])
model.compile(optimizer='sgd', loss='mse')
model.fit([1,2,3],[2,4,6], epochs=5, callbacks=[PrintLossCallback()])

Custom callback to print loss at each epoch.

from tensorflow import keras
from tensorflow.keras import layers

# Generator
generator = keras.Sequential([
    layers.Dense(128, activation='relu', input_shape=(100,)),
    layers.Dense(784, activation='sigmoid')
])

# Discriminator
discriminator = keras.Sequential([
    layers.Dense(128, activation='relu', input_shape=(784,)),
    layers.Dense(1, activation='sigmoid')
])
discriminator.compile(optimizer='adam', loss='binary_crossentropy')

Minimal GAN architecture structure in Keras.