stratify=labels,
random_state=42)
# # Account for skew in the labeled data
# from sklearn.utils import class_weight
# class_weights = class_weight.compute_sample_weight('balanced', np.unique(train_y), train_y)
# # class_totals = labels.sum(axis=0)
# # class_weights = class_totals.max() / class_totals
# # class_weights = dict(enumerate(class_weights))
# print("class_weight: ", class_weights)
# class_weight_dict = dict(enumerate(np.unique(y_train), class_weights))
# print("class_weight: ", class_weight_dict)
# Initialize the model
print("[INFO]: Compiling model....")
model = LeNet.build(width=28, height=28, depth=1, classes=2)
model.compile(loss="binary_crossentropy",
optimizer="adam",
metrics=["accuracy"])
# Train the network
print("[INFO]: Training....")
H = model.fit(
train_x,
train_y,
validation_data=(test_x, test_y),
# class_weight=class_weight,
batch_size=64,
epochs=15,
verbose=1)
from utilities.nn.cnn import LeNet from keras.utils import plot_model # Initialize LeNet and then write the network architecture visualization grpah to disk model = LeNet.build(28, 28, 1, 10) plot_model(model, to_file="output/lenet_architecture.png", show_shapes=True)