"the same problem" is I posted above but I was using jupyter lab, I simplified hoping the problem was with jupyter

the models tab has nothing

AgitatedDove14 other things are logging, I'm using task.connect to log hyperparameters logger to log the loss

it's as simple as yours,-- class myCNN(nn.Module):

torch.save(model.state_dict(), " http://test_cnn.pt ")

from clearml import OutputModel, Task
task = Task.init(project_name='ClearML test', task_name='test 1')

version 1.7.0-232 I think

also attempted: model = task.connect(model)

from clearml import Task
task = Task.init(project_name='ClearML test', task_name='test 10')

AgitatedDove14 quick question, do I need to alter the name of the dict file I save with torch.save for each run? Or is clearml able to version it out?

hmmm, yes that worked

this is in the code but still no logged model

thanks for your help!

now that I know the logging is working, it must be something dumb I've overlooked in my code