Segmentation Models - v0.1.2

Fixes

• Fix pytorch-efficientnet package version in requirements.txt to strict 0.6.3 (#260)

Segmentation Models - v0.1.1

Updates

• New decoders DeepLabV3, DeepLabV3+, PAN
• New backbones (encoders) timm-efficientnet*
• New pretrained weights (ssl, wsl) for resnets
• New pretrained weights (advprop) for efficientnets

And some small fixes.

Thanks @IlyaDobrynin @gavrin-s @lizmisha @suitre77 @thisisiron @phamquiluan and all other contributers!

Segmentation Models - v0.1.0

Updates

1. New backbones (mobilenet, efficientnet, inception)
2. depth and in_channels options for all models
3. Auxiliary classification output

Note!

Model architectures have been changed, use previous versions for weights compatibility!

Segmentation Models - v0.0.3

Updates

• Conv2D Initialization
  • kaiming_normal -> kaiming_uniform;
  • fan_out -> fan_in;
  • bias -> 0
• package dependencies

Segmentation Models - v0.0.2

Updates

• New backbones:
  • resnext50_32x4d
  • resnext101_32x8d
  • resnext101_32x16d
  • resnext101_32x32d
  • resnext101_32x48d
• Unet scSE attention block (optional)
• torchvision version update
• get_preprocessing_params function

Thanks @laol777 for contribution!