space-codes / asar.core

This is our graduation project for arabic manuscript analysis and recognition
0 stars 0 forks source link

Encoding the dataset labels #18

Closed aboelkassem closed 2 years ago

aboelkassem commented 2 years ago
aboelkassem commented 2 years ago

Added through 1b9e4965f62e00332a2387724c4698add3b99559

aboelkassem commented 2 years ago

or you can use


# Converting the class vector in integers to binary class matrix
y_train = to_categorical(y_train)
y_test = to_categorical(y_test)
print(y_train.shape, y_test.shape)
for i in zip(y_train[0], arabic_characters):
    print(i)
aboelkassem commented 2 years ago

Endoded using PHOC label