Google Net

[ArturAzarskyy]

Try the GoogleNet architecture:

• inceptionV3 training basic : https://github.com/JordiCorbilla/ocular-disease-intelligent-recognition-deep-learning/blob/master/odir_inception_v3_training_basic.py
• inceptionV3 training : https://github.com/JordiCorbilla/ocular-disease-intelligent-recognition-deep-learning/blob/master/odir_inception_v3_training.py
• ResNetV2 training basic : https://github.com/JordiCorbilla/ocular-disease-intelligent-recognition-deep-learning/blob/master/odir_inception_ResNetV2_training_basic.py
• InceptionV1 model: https://github.com/JordiCorbilla/ocular-disease-intelligent-recognition-deep-learning/blob/f67ed307c3963e8e06c676f708e448b6663b604c/odir_model_inception_v1.py#L22

This could be a good idea to implement in general later on, will probably need to have a separate issue later on

• testing interface https://github.com/JordiCorbilla/ocular-disease-intelligent-recognition-deep-learning/blob/master/odir_inception_v3_testing_inference.py