[features] Add pipeline

[okbalefthanded]

• evaluation pipeline

[okbalefthanded]

• Cross validation strategies available:
• Kfold (only for SingleSubject evaluation)
• Shufflesplit (for both types of evaluation: Single/Cross subject)

[okbalefthanded]

Long running evaluations may face interruptions, a checkpoint mechanism is required to resume operations from interruption point

• Evaluation checkpoint and resume

[okbalefthanded]

Add Cross dataset evaluation for domain adaptation

• CrossSet

[okbalefthanded]

• Hyperparameter tuning with Keras-tuner

[okbalefthanded]

• Experimental:
• Add preprocessing layers before any model
• Break pipeline : delete preprocessing methods and leave it to the model

[okbalefthanded]

• Set image date format in the evaluation for model's training/inference consistency on GPU/TPU/CPU.

[okbalefthanded]

• Add Debug mode to evaluation:
  • Keras Tensorboard callback.
  • TF Debugging V2