Lightning Support, Better visualizations

Glad to announce the release 0.3.0 of PyTorch CNN Trainer.

Features

• PyTorch Lightning trainer is now avaliable. It supports all torchvision models as of now. (#26)
• [BC BREAKING] Dictionary object with metrics is returned from fit instead of True #23
• Better plotting utility to visualize results quickly (#25)
• Utility to turn off NVIDIA Profilers. This might results in faster training (#21)

Examples

• New examples for lightning based trainer (#26)

Tests and CI

• Tests are added for lightning (#26)
• Dropped Black code formatting.

Contributors

This release was possible due to contributions from @premalrupnur, @oke-aditya . Glad to have more people here.

0.3.0rc2

New beta release

0.3.0rc1

Bumpy version for next release

CNN Training and Fine-Tuning with PyTorch 1.6

PyTorch CNN Trainer

A complete CNN trainer package that can train all CNNs in torchvision and timm models.
It can fine-tune as well as train from scratch.

Features

• Support PyTorch image models (timm) training and transfer learning.
• Quantization Aware training example.
• Early stopping with patience.
• Support torchvision models trainging and transfer learning.
• Support torchvision quantized models transfer learning.
• Support for Mixed Precision Training.
• L2 Norm Gradient Penalty.
• SWA Stochastic weighted Averaging support for training.
• Add Keras Like fit method.
• Sanity Check method.

0.2.0rc1

Tries for 0.2.0release

v0.1.0rc4

Update settings.ini

v0.1.0rc3: Merge pull request #17 from oke-aditya/add_swa

Updates readme, manfiest, Tries adding SWA

v0.1.0rc2: Merge pull request #17 from oke-aditya/add_swa

Updates readme, manfiest, Tries adding SWA

v0.1.0-rc1: Merge pull request #17 from oke-aditya/add_swa

Updates readme, manfiest, Tries adding SWA