I followed the instructions under the "training" section of the readme, but executing the train.py command results in an error which reads 'Access is denied.' This is the command I am referring to:
python bpartis/train.py --data-dir=<path> --device=cuda --epochs=300 --save-dir=bpartis/saved_models/
I have tried running cmd as an administrator but this does not solve the issue. Is there something I am doing incorrectly?
Another question I had was whether it was possible to disregard the EMPS dataset and to feed my own images into the program for BPartIS to better recognize the particles I am working with specifically. I have tried to analyze my personal microscopy images using imagedataextractor, but it does not seem to pick up any of the particles I am interested in; sometimes even picking up the absence of a particle as a supposed particle while disregarding what actually interests me. Is it possible to personalize this program by feeding it custom particle images or would this require overhauling the entirety of the code?
Thank you.
I followed the instructions under the "training" section of the readme, but executing the train.py command results in an error which reads 'Access is denied.' This is the command I am referring to:
python bpartis/train.py --data-dir=<path> --device=cuda --epochs=300 --save-dir=bpartis/saved_models/I have tried running cmd as an administrator but this does not solve the issue. Is there something I am doing incorrectly?
Another question I had was whether it was possible to disregard the EMPS dataset and to feed my own images into the program for BPartIS to better recognize the particles I am working with specifically. I have tried to analyze my personal microscopy images using imagedataextractor, but it does not seem to pick up any of the particles I am interested in; sometimes even picking up the absence of a particle as a supposed particle while disregarding what actually interests me. Is it possible to personalize this program by feeding it custom particle images or would this require overhauling the entirety of the code?
Thank you.