9951 explained code solutions for 126 technologies


python-kerasHow do I set the batch size when using Python and Keras?


The batch size is an important hyperparameter that controls the number of samples propagated through the network before the weights are updated.

In Python and Keras, the batch size can be set using the batch_size argument of the fit() method.

For example, if we want to set the batch size to 32, we can do so as follows:

model.fit(x_train, y_train, batch_size=32)

The batch_size argument can also be passed to the fit_generator() method if using a generator to supply data.

model.fit_generator(generator, batch_size=32)

The batch size can also be set when creating the model. For example, if we are using a Sequential model, we can pass the batch_input_shape argument to specify the batch size.

model = Sequential(batch_input_shape=(32, x_train.shape[1], x_train.shape[2]))

Helpful links

Edit this code on GitHub