The generator should return the same kind of data as accepted by
predict_on_batch().
predict_generator(object, generator, steps, max_q_size = 10, workers = 1,
pickle_safe = FALSE, verbose = 0)Keras model object
Generator yielding batches of input samples.
Total number of steps (batches of samples) to yield from
generator before stopping.
Maximum size for the generator queue.
Maximum number of processes to spin up when using process based threading
If TRUE, use process based threading. Note that because
this implementation relies on multiprocessing, you should not pass non
picklable arguments to the generator as they can't be passed easily to
children processes.
verbosity mode, 0 or 1.
Numpy array(s) of predictions.
ValueError: In case the generator yields data in an invalid format.
Other model functions: compile,
evaluate_generator, evaluate,
fit_generator, fit,
get_config, get_layer,
keras_model_sequential,
keras_model, pop_layer,
predict.keras.engine.training.Model,
predict_on_batch,
predict_proba,
summary.keras.engine.training.Model,
train_on_batch