xmindflow / deformablelka Goto Github PK

Hello author, I noticed a strange phenomenon while reproducing the results of DAEFormer on the ISIC2018 dataset. After starting training on the preprocessed dataset, I noticed a gradient explosion in the loss value section. May I ask if this is caused by some issues with the code of the processed dataset

Hello, your work is excellent, and I'm eager to delve deeper into learning. However, I encountered an issue while running the training code, which indicates that the pre-training weight file is missing.

As shown below:
FileNotFoundError: [Errno 2] No such file or directory: './pretrained_pth/maxvit/maxvit_rmlp_small_rw_224_sw-6ef0ae4f.pth'

I couldn't find this particular pre-training weight file on the code homepage. Would you mind providing it? If possible, here's my email: [email protected].

Hello, while I was training the model, the system reminded me that it couldn't find the file “maxvit_rmlp_small_rw_224_sw-6ef0ae4f.pth”. How can I solve it? thanks ( :

I find the code finally output 13 classes' results, but in inference_synpase.py, you wanna 8 classes like other papers. i wonder is the final result save in "summary.json"? and when i reproduce the code, i find inference_synpase.py didnt work , as well as write in the text. Then my final result is different from your paper, the result is in the image

Hello! Thank you for resolving my issue earlier. I now realize that I ran it first but couldn't find the file. Where does it exist? Thank you very much!

Dear author：
Due to the poor performance of my computer, it is difficult to finish running the train_MaxViT_deform_LKA.py file.
When I run the test.py file it shows that I don't have best_model.pth and best_model.pth.txt
Do I need to run train_MaxViT_deform_LKA.py completely to obtain these two files? Or can these two files be downloaded directly?

what is the role of 2D/deformable_LKA/deformable_2dconvolution.py and 2D/deformable_LKA/deformable_3dconvolution.py ?
Are they python implementation of dcn? If that's the case, why you need to use D3D?

Futhermore, as far as I konwn, D3D process THW data, they don't deform at T. I can't see the any modifications to D3D's dcn library.

So how do you deform at three spatial dimension? I mean, D, H, W. Thank you very much.

I followed all the steps you have including downloading the same Python and PyTorch versions for running the pancreas code.

The train_pancreas.py ran well and the weights were saved as intended. However, I am getting an error when running the test_pancreas.py and I cannot seem to fix it.

Hello author, while reproducing ISIC2018 and ISIC2018, I found a significant difference between the final results and those in your paper. May I ask if you have done any processing on the dataset, or could you upload the dataset you have processed.

Dear author, may I ask if you can tell us how to obtain such visual results after obtaining the predicted results? Is it through any software?



Thank you very much! Thank you for your generous sharing!

note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for D3D
ERROR: Could not build wheels for D3D, which is required to install pyproject.toml-based projects

When training synapse data in 2D, processing 12% and then error：

Traceback (most recent call last):
File "train_MaxViT_deform_LKA.py", line 134, in
trainer[dataset_name](args, net, args.output_dir)
File "/home/sjh/mydisk/project/deformableLKA/2D/trainer_MaxViT_deform_LKA.py", line 168, in trainer_synapse
image = image_batch[1, 0:1, :, :]
IndexError: index 1 is out of bounds for dimension 0 with size 1

Who can handle this problem？

because i dont have an identification of the from identity provider 'live.com' does not exist in tenant 'Mohamed Bin Zayed University of Artificial Intelligence' and cannot access the application '00000003-0000-0ff1-ce00-000000000000'(Office 365 SharePoint Online) in that tenant. The account needs to be added as an external user in the tenant first. Sign out and sign in again with a different Azure Active Directory user account. can you change the source? thank u!

How to solve this? Or can I download D3D in other ways?

Hello, I would like to know where the complete code flow of the Deformable LKA module in Figure 2 in your paper is, but I didn't find it

Dear author， When I want to use synapse code to train kipa22 dataset, I meet this problem. But I don't know what's wrong with this. I do not know whether there is a problem with the size format of my data, or the parameters in the code need to be modified? Thanks for your help, may be this is a easy problem, but I don't know how to dealt with it.
I processed the date just use the nnUnet's code, then I copy the processed data to this project's folder.

Traceback (most recent call last):
File "d_lka_former/run/run_training.py", line 211, in
main()
File "d_lka_former/run/run_training.py", line 195, in main
trainer.run_training()
File "/home/javier/Desktop/deformableLKA-main/3D/d_lka_former/training/network_training/d_lka_former_trainer_synapse.py", line 490, in run_training
ret = super().run_training()
File "/home/javier/Desktop/deformableLKA-main/3D/d_lka_former/training/network_training/Trainer_synapse.py", line 321, in run_training
super(Trainer_synapse, self).run_training()
File "/home/javier/Desktop/deformableLKA-main/3D/d_lka_former/training/network_training/network_trainer_synapse.py", line 485, in run_training
l = self.run_iteration(self.tr_gen, True)
File "/home/javier/Desktop/deformableLKA-main/3D/d_lka_former/training/network_training/d_lka_former_trainer_synapse.py", line 295, in run_iteration
output = self.network(data)
File "/home/javier/conda3/envs/d_lka_net_3d/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "/home/javier/Desktop/deformableLKA-main/3D/d_lka_former/network_architecture/synapse/d_lka_former_synapse.py", line 153, in forward
x_output, hidden_states = self.d_lka_former_encoder(x_in)
File "/home/javier/conda3/envs/d_lka_net_3d/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "/home/javier/Desktop/deformableLKA-main/3D/d_lka_former/network_architecture/synapse/model_components.py", line 69, in forward
x, hidden_states = self.forward_features(x)
File "/home/javier/Desktop/deformableLKA-main/3D/d_lka_former/network_architecture/synapse/model_components.py", line 56, in forward_features
x = self.stages0
File "/home/javier/conda3/envs/d_lka_net_3d/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "/home/javier/conda3/envs/d_lka_net_3d/lib/python3.8/site-packages/torch/nn/modules/container.py", line 139, in forward
input = module(input)
File "/home/javier/conda3/envs/d_lka_net_3d/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "/home/javier/Desktop/deformableLKA-main/3D/d_lka_former/network_architecture/synapse/transformerblock.py", line 60, in forward
x = x + self.pos_embed
RuntimeError: The size of tensor a (71680) must match the size of tensor b (32768) at non-singleton dimension 1

what is the differences among DeformConv, DeformConvPack, DeformConvPack_experimental,
DeformConv_d, and DeformConvPack_d?

Dear author：
When I run train_MaxViT_deform_LKA.py
the result：FileNotFoundError: [Errno 2] No such file or directory: './pretrained_pth/maxvit/maxvit_rmlp_small_rw_224_sw-6ef0ae4f.pth'

Dear author：When I run the train_MaxViT_deform_LKA.py file, it shows OSError: [Errno 22] Invalid argument: '.\model_out\MaxViT_improved_deform_LKA_two_layer\MaxViT_improved_deform_LKA_2024-03-16 12:43:30.810350dice.png'. How can I solve it?

Dear author, while I was training the model designed by myself, "mean_dice" and "mean_hd95" will arise zero values. How can I solve it? Thanks.