Closed spa-yhson closed 1 month ago
Thanks a lot for your interest in our work!
Did you run the commands in the "Install dependencies using pip" section of the readme file and were there any errors?
I have resolved the previous issue. Now I have a question regarding evaluation process.
I have used python run.py --env local evaluate --config_path ./heal_swin/run_configs/segmentation/evaluate_all_config.py
command.
Below is the error I received:
File "/home/Documents/HEAL-SWIN/heal_swin/utils/utils.py", line 136, in get_artifacts_path
raise RuntimeError(f"The specified run_id {run_id} wasn't found")
RuntimeError: The specified run_id wasn't found
How to resolve this?
Great to read that you could run the training. I’ve also pushed a fix with an updated flake8 version in setup.py
.
The run id refers to the id in MLFlow under which the run was logged. It is read from the environment variable RUN_ID
. So, if you run
export RUN_ID=<<MLFLow run id>>; python run.py --env local evaluate --config_path ./heal_swin/run_configs/segmentation/evaluate_all_config.py
it should work.
Thank you so much for your assistance. I have one final question regarding the sharing of model weights. Would it be possible to obtain the saved weights referenced in the paper?
At the moment, we don’t plan to release the weights, sorry. But we provide the exact training configurations which produced the published results, see the documentation here: https://github.com/JanEGerken/HEAL-SWIN/tree/master/heal_swin#training-config-files . If you have trouble reproducing our results, please let me know!
Dear author,
Hello, I am writing to express my gratitude for your excellent research showcased in your paper. Along with my appreciation, I would like to ask about the error encountered when run your code.
I used
python run.py --env local train
command to train your model, but I encountered following error:Could you please tell me how to resolve this?
Thank you so much for your time.