Parameters not supported

The following parameters currently cant be assigned for both TensorFlow and PyTorch in the training plan:

• steps_per_epoch
• initial_epoch
• validation_data
• validation_steps
• validation_batch_size
• validation_freq
• max_queue_size
• workers
• use_multiprocessing
• metrics
• loss_weights
• weighted_metrics
• run_eagerly
• steps_per_execution
• jit_compile
• zca_epsilon
• zca_whitening
• data_format
• interpolation_order
• dtype
• featurewise_center
• featurewise_std_normalization

Additionally, these parameters are not supported for PyTorch specifically:

• Adaptive and Custom learning rate
• Custom loss function
• horizontal_flip
• samplewise_center
• samplewise_std_normalization
• vertical_flip