DeepProfiler with TFv1 and TFv2 training strategies Latest

This is a stable version of DeepProfiler with features:

• Label smoothing
• Online labelling training strategy
• TFv2 basic training (separate module)

And some other improvements and bug fixes.

Last TensorFlow v1 stable version

This version of DeepProfiler is the last one based on TensorFlow 1.15.

• Improved training speed and stability.
• Added EfficientNet plugins.
• Support for full image analysis.
• Sampling single cells in separate files for training.
• Fixed minor bugs.

Version 0.1.0

This is the first beta release for DeepProfiler. The following features have been added:

• Plugins system for models, generators, and metrics
• 6 new models
• Crop generators for autoencoders, mixup, and repeated channels
• Improved set of commands
• Hyperparameter optimization