junyahayashida / mpm Goto Github PK

ValueError: Using config_path to specify the config name is not supported, specify the config name via config_name.
See https://hydra.cc/docs/next/upgrades/0.11_to_1.0/config_path_changes

Hi~ and first of all thank you for sharing your great work on github!
I could apply the MPM approach for my own research project and so far it is working really well.

To compare my results to your results from the MPM paper I would like to use the same metric definition.
However from the paper I am not 100% sure, how you calculated the metrics. Is it possible to share your code/ more detailed definition of cell detection performance, association accuracy and target effectiveness?

F.e. for cell detection performance: how did you compare the predicted vs ground truth likelihood maps to get recall, precision and f1 score?

Thank you already~

Hi, great work! Thanks for sharing the code~
I am a little bit confused about the dataset structure of the training set.
As for the training command is 'python3 mpm_train.py 100 40 sample/train', how can I generate the train directory and what is the exact structure of the train directory?
It will be very helpful for you to introduce these details~ Thanks for your time.

Thank you for you amazing work! Will the code for the cell tracking part after MPM be available?

Hi, great work! I wonder if the dataset is available for us.