Reputation
Badges 1
14 × Eureka!from clearml import OutputModel, Task
task = Task.init(project_name='ClearML test', task_name='test 1')
also attempted: model = task.connect(model)
torch.save(model.state_dict(), " http://test_cnn.pt ")
"the same problem" is I posted above but I was using jupyter lab, I simplified hoping the problem was with jupyter
this is in the code but still no logged model
AgitatedDove14 other things are logging, I'm using task.connect to log hyperparameters logger to log the loss
from clearml import Task
task = Task.init(project_name='ClearML test', task_name='test 10')
now that I know the logging is working, it must be something dumb I've overlooked in my code
it's as simple as yours,-- class myCNN(nn.Module):
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?