Hello MotionlessCoral18 ,
Can you please add a log with the failure?
As my clearml server is run using docker, i have no idea where http://clearml.10.68.0.250.nip.io/projects/300ec77013504f51a7f295226c3f7e40/experiments/5418cf58b64f425a9a17fbd4af6cfee8/output/log is exactly stored.
Hello MotionlessCoral18 . I have a few questions that might help us find out why you experience this problem:
Is there any chance you are running the program in offline mode? Is there any other message being logged that might help? The error messages might include Action failed
, Failed sending
, Retrying, previous request failed
, contains illegal schema
Are you able to connect to the backend at all from the program you are trying to get the dataset?
Thank you!
ClearML results page:
http://clearml.10.68.0.250.nip.io/projects/300ec77013504f51a7f295226c3f7e40/experiments/5418cf58b64f425a9a17fbd4af6cfee8/output/logTraceback (most recent call last):
File "/app/.clearml/venvs-builds/3.8/code/__init__.py", line 287, in <module>
[train_data, test_data, train_loader, test_loader, nb_class] = import_data(root_database, train_path, test_path,
File "/app/.clearml/venvs-builds/3.8/code/__init__.py", line 153, in import_data
root = Dataset.get(
File "/app/.clearml/venvs-builds/3.8/lib/python3.8/site-packages/clearml/datasets/dataset.py", line 1153, in get
tasks = Task.get_tasks(
File "/app/.clearml/venvs-builds/3.8/lib/python3.8/site-packages/clearml/task.py", line 967, in get_tasks
return cls.__get_tasks(task_ids=task_ids, project_name=project_name, tags=tags,
File "/app/.clearml/venvs-builds/3.8/lib/python3.8/site-packages/clearml/task.py", line 3749, in __get_tasks
queried_tasks = cls._query_tasks(
File "/app/.clearml/venvs-builds/3.8/lib/python3.8/site-packages/clearml/task.py", line 3817, in _query_tasks
ret_tasks.extend(res.response.tasks)
TypeError: 'NoneType' object is not iterable
But i don't want to create a new dataset, my dataset exists and has already been downloaded by a previous task.
what versions do you have for the clearml packages ?
Can you try to add the flagauto_create=True
when you call Dataset.get ?