Hello! Thank you for making such a great segmentation toolkit. I was interested in using the pre-trained weights of the organ segmentation pipeline for a transfer learning problem; however, for this I would need the model architecture, not just the weights. The pre-trained weights are in a .MODEL file, which seems to contain a state_dict with the weights, as well as some other information related to model training, but not the architecture itself. Is the model architecture somewhere in this repository?
The model is normal nnunet model. You can check the nnunet code + documentation (https://github.com/MIC-DKFZ/nnUNet) for some more information. I would have to search for the model architecture myself.
Hello! Thank you for making such a great segmentation toolkit. I was interested in using the pre-trained weights of the organ segmentation pipeline for a transfer learning problem; however, for this I would need the model architecture, not just the weights. The pre-trained weights are in a .MODEL file, which seems to contain a state_dict with the weights, as well as some other information related to model training, but not the architecture itself. Is the model architecture somewhere in this repository?