Closed virlyananda closed 1 year ago
Hmm, not sure. Can you paste the whole error here with the command you used to run the prediction step and the predict json file please? Thanks
Hello, I was just trying the package with sample dataset, I didn't change much of anything except the path file. This is the issue I got.
Can you change False to "False" in the overwrite row in the json? If that is the issue sorry!
Thank you. It works.
Also, do you know if the program works for different type of bin (binning) size?
Should work, yes
Hello, Do you know if there's an alternative to align, split and generate even/odd tomograms more efficiently? I notice I could no longer use the cryoCARE_T2T align & split due to different IMOD version issue. As recommended from the cryocare_pip, I tried using alignframes -debug 10000 but it only gave me the frames that have been split but not the .rec tomograms on both ends (which is what we need for the training inputs). Looking forward to hear your thoughts! Thanks!
I don't sorry, I have a personalised way of dealing with generating even/odd tomos which isn't publicly available yet and won't be for a while. Can you not just install the correct version of IMOD?
Ok. Thank you!
I guess this is solved.
Hello, I tried running the latest version of cryocare (for CUDA 11) and was able to implement the training part (here I was able to obtain .tar.gz file). However, I encountered an issue with the prediction step where it mentions json.decoder.JSONDecodeError. I installed all the necessary packages from cryocare_11 environment, so I'm not sure where the problem is. Please let me know if anyone encountered similar issue or if there's anyway I could fix this. Thank you!