search

ArrichM / dynaflow

Apache License 2.0
4 stars 2 forks source link

Issues with MLFlow 2.0 #2

Open gvelimir opened 2 years ago

gvelimir commented 2 years ago

The latest version of MLFlow 2.0 does not function with Dynaflow. Since MLFlow removed the list_experiments method, the tables from the backend store could not be fetched. Here is the stack trace:

2022/11/17 07:42:02 ERROR mlflow.server: Exception on /ajax-api/2.0/mlflow/experiments/search [GET] Traceback (most recent call last): File "/home/velimir/anaconda3/envs/intertec-mlops/lib/python3.10/site-packages/flask/app.py", line 2525, in wsgi_app response = self.full_dispatch_request() File "/home/velimir/anaconda3/envs/intertec-mlops/lib/python3.10/site-packages/flask/app.py", line 1822, in full_dispatch_request rv = self.handle_user_exception(e) File "/home/velimir/anaconda3/envs/intertec-mlops/lib/python3.10/site-packages/flask/app.py", line 1820, in full_dispatch_request rv = self.dispatch_request() File "/home/velimir/anaconda3/envs/intertec-mlops/lib/python3.10/site-packages/flask/app.py", line 1796, in dispatch_request return self.ensure_sync(self.view_functions[rule.endpoint])(view_args) File "/home/velimir/anaconda3/envs/intertec-mlops/lib/python3.10/site-packages/mlflow/server/handlers.py", line 452, in wrapper return func(*args, *kwargs) File "/home/velimir/anaconda3/envs/intertec-mlops/lib/python3.10/site-packages/mlflow/server/handlers.py", line 522, in wrapper return func(args, kwargs) File "/home/velimir/anaconda3/envs/intertec-mlops/lib/python3.10/site-packages/mlflow/server/handlers.py", line 1017, in _search_experiments response_message.experiments.extend([e.to_proto() for e in experiment_entities]) TypeError: 'NoneType' object is not iterable

BenWilson2 commented 1 year ago

Hi @ArrichM great package, by the way! When in the process of updating for MLflow2.0, note that we added a top-level attribute to Run wherein we include run_name (as well as preserve the tag-based tags.mlflow.runName convention for backwards compatibility). Feel free to reach out to the team on the MLflow slack with any other questions you might have around breaking changes in 2.0 :)