how to design neural network architecture using keras sequential code example
Example: keras ann code
import numpy as np
from keras.utils import to_categorical
from keras import models
from keras import layers
from keras.datasets import imdb
(training_data, training_targets), (testing_data, testing_targets) = imdb.load_data(num_words=10000)
data = np.concatenate((training_data, testing_data), axis=0)
targets = np.concatenate((training_targets, testing_targets), axis=0)
def vectorize(sequences, dimension = 10000):
results = np.zeros((len(sequences), dimension))
for i, sequence in enumerate(sequences):
results[i, sequence] = 1
return results
data = vectorize(data)
targets = np.array(targets).astype("float32")
test_x = data[:10000]
test_y = targets[:10000]
train_x = data[10000:]
train_y = targets[10000:]
model = models.Sequential()
model.add(layers.Dense(50, activation = "relu", input_shape=(10000, )))
model.add(layers.Dropout(0.3, noise_shape=None, seed=None))
model.add(layers.Dense(50, activation = "relu"))
model.add(layers.Dropout(0.2, noise_shape=None, seed=None))
model.add(layers.Dense(50, activation = "relu"))
model.add(layers.Dense(1, activation = "sigmoid"))
model.summary()
model.compile(
optimizer = "adam",
loss = "binary_crossentropy",
metrics = ["accuracy"]
)
results = model.fit(
train_x, train_y,
epochs= 2,
batch_size = 500,
validation_data = (test_x, test_y)
)
print("Test-Accuracy:", np.mean(results.history["val_acc"]))