Pat Gunn
Pat Gunn
One good way to do this is to reduce the number of processes; most of the notebooks have a call to setup_cluster that takes a n_processes flag, and a nearby...
Unfortunately the error log above is from the jupyter/spawning level rather than anything caiman itself emits; you can try increasing the loglevel to the INFO priority; this will produce way...
Is there anything more in that cell than that (below "Using Keras")?
It's near the end of the day; I'm working through the call tree to figure out where it might be crashing and how to investigate further, but it may take...
Right after that "Using Keras" message, it's going to try to load the model file and it will throw an exception if that fails, printing something if it succeeds. In...
Oh, that's surprising and interesting. I'll have to see if I can replicate that; can you send me the output of `conda list` for your environment? It's okay to pastebin...
Oh, also that chunk of code is small enough that you could try running it in a script outside of jupyter (but still in the right environment), which may give...
@jessegmeyerlab Can you break up your test cell above into the part that loads the json and the part that loads the hdf5 so we can see if it fails...
The latest version of Caiman replaces Tensorflow with Pytorch; might be worth seeing if it still comes up.