Releases | Open Source Science

- Python
Published by jgrss over 2 years ago

- Python
Published by jgrss almost 3 years ago

- Python
Published by jgrss almost 3 years ago

Improved and fixed issues in the ResUNet 3+ Psi architecture, which was introduced in v1.6.5
More flexible user arguments. The user can now specify:
- the model architecture
- convolution blocks
- dilations
- attention weights
Improvements in the train optimizer stability
Deep supervision
- Cultionet uses the UNet 3+ style of deep supervision along three decoders
  - These are optional during training
Improved training efficiency using PyTorch’s parallel data loader
Improved inference efficiency using PyTorch’s batch loader

- Python
Published by jgrss almost 3 years ago

We upgraded geowombat to v2.1.4 following a bug fix.

- Python
Published by jgrss almost 3 years ago

- Python
Published by jgrss about 3 years ago

- Python
Published by jgrss about 3 years ago

setuptools was bumped from >=59.5.0 to >=65.5.1

- Python
Published by jgrss about 3 years ago

- Python
Published by jgrss about 3 years ago

New architecture design based on UNet 3+ and residual convolutions
- The new design is a multi-head connection of the UNet 3+ architecture
- Added optional crop-type model for finer crop learning
Modified total loss quantification with deep supervision of crop type in RNN layer
- The tanimoto loss is used on all layers
Added num_workers option in DataLoader for faster train/predict
Added .pt data compression by changing torch.save|load to joblib.dump|load