1. Network Type
This tab define the deep learning network we would like to use.
Network Selections
Options | Description |
|---|---|
RCAN | For denoising and super-resolution. This is also the model used by our Nature Methods papaer. https://www.nature.com/articles/s41592-021-01155-x |
UNet | For virtual staining and segmentation |
Network Shape
Options: 2D or 3D
Description: By default, you should choose 2D model for 2D image data and 3D model for 3D data. But you can train a 2D model using 3D data. It will process the image slice-by-slice.
RCAN Specific Parameters
Number of Filters
Default: 32
Description: Number of features (i.e. number of output channels of each convolution layer). Increase the number for model complexity, reduce for speed up the training.
Number of Residual Blocks
Default: 3
Description: Number of residual blocks in each residual group
Number of Residual Groups
Default: 3
Description: Number of residual groups
Channel Reduction Factor
Default: 8
Description: Channel reduction factor for the squeeze-and-excitation module.
UNet Specific Parameters
Depth
Number of Initial Filter
Filter Growth Factor
Normalization Type
Channel Reduction Factor
Use Attention Gate
Activation Type at the Last Layer
2. Training Parameters
This tab define how are we going to update the model weights during training
3. Apply Parameters
This tab define how are we going to update the model weights during training
