segmentation_models.pytorch

segmentation_models.pytorch copied to clipboard

Docs are sometimes a unclear

3

https://smp.readthedocs.io/en/latest/losses.html#segmentation_models_pytorch.losses.constants.MULTILABEL_MODE It is not clear in the description of multilabel mode what C=1..N means and the sentence is not correct English, which is confusing when reading it.

kaspar-gestalt

Add Mask2Former to SMP

4

Mask2Former model was introduced in the paper Masked-attention [Mask Transformer for Universal Image Segmentation](https://arxiv.org/abs/2112.01527) and first released in [this repository](https://github.com/facebookresearch/Mask2Former/). Mask2Former addresses instance, semantic and panoptic segmentation with the same...

caxel-ap

How can I perform data training on a custom binary segmentation dataset?

3

I tried to build my own dataset for training by following the code example at https://github.com/qubvel-org/segmentation_models.pytorch/blob/main/examples/cars%20segmentation%20(camvid).ipynb, but failed. How can I replace my own binary segmentation dataset in the code?

MrLangZe

How to add a mamba backbone as an encoder for training

3

ljb-1

Ensure ONNX support

1

I would like to request the addition of a feature to support exporting models to the ONNX (Open Neural Network Exchange) format within the library. As ONNX is becoming a...

johnnv1

I have worked with FPN in the past. The same code now runs without error, but the training F-score/IoU becomes zero. I've tried switching FPN with U-Net/LinkNet in the same...

arcskyrider

Pytorch Lightning Checkpoint saving and Earlystopping

4

Recently I have been trying to train an SMP model using pytorch lightning Checkpoint saving and Earlystopping features in my training loop to save progress. it uses a torch.save() method...

omarequalmars

Add example of amp training & qunatization-aware training

5

nieyan

TypeError: __init__() got an unexpected keyword argument 'output_stride'

1

``` smp.DeepLabV3Plus(encoder_name='tu-maxvit_base_tf_224', encoder_depth=5, encoder_weights='imagenet', encoder_output_stride=16, decoder_channels=256, decoder_atrous_rates=(12, 24, 36), in_channels=3, classes=1, activation=None, upsampling=4, aux_params=None) ``` when I run the above code,I get error, TypeError: __init__() got an unexpected keyword argument...

LuShuaie

Error when trying to parse the encoder output to the decoder

1

Hello, I am trying to remove the segmentation head and add some more layers before adding segmentation head to my model. But whenever I pass my encoder output to the...

mvp-aadesh