Logging #11
Logging #11
Conversation
umarteauowkin
requested review from
BorisMuzellec and
tanguy-marchand
as code owners
There was a problem hiding this comment.
Real change : description of the test paths json.
There was a problem hiding this comment.
Real change: add docformatter
There was a problem hiding this comment.
adding this file to allow devs to fully install everything (testing datasets included)
There was a problem hiding this comment.
Adding this file in the doc to describe how logging is handled
There was a problem hiding this comment.
Big change: the logging is activated only if the logging path points to a file that is not empty.
There was a problem hiding this comment.
- utils to log iterations and organisation methods.
There was a problem hiding this comment.
Readme on logging
There was a problem hiding this comment.
Adding 2 files to describe logging and install for devs including dataset download and logging config
There was a problem hiding this comment.
Adding docformatter
There was a problem hiding this comment.
Adding logging paths and simplifying processed_data -> either tmp if nothing is set in the paths.json or not if set (so that locally a user can avoid rebuilding the datasets all the time).
There was a problem hiding this comment.
New file to test logging with the workflow file.
What does your PR implement? Be specific.
This PR cleans the logging utilities of fedpydeseq2.
Logging is now configured using a configuration file, and is not performed by default. It is only performed in subprocess mode.
Moreover, we add the possibility to create a workflow file in subprocess mode (configured by the configuration file)
This file essentially logs the input and outputs of all remote and remote_data methods (it actually logs the keys of the dictionaries and their types and shapes), as well as when a for loop is used, and when we structure the calling to these remote/remote_data methods using organisation methods.
This PR also adds docformatter, hence the enormous number of changed files. Sorry about this.
Finally: I also modified the readmes and the docs to reflect these new features, even though they are honestly meant for developpers. Note that these features are tested.
I also simplified the test logic of processed data, to avoid, locally, to have to reprocess the data all the time.