ramp-workflow
ramp-workflow copied to clipboard
paris-saclay-cds
Toolkit for building predictive workflows on top of pydata (pandas, scikit-learn, pytorch, keras, etc.).
Sometimes, the bagging CV scores can be slow and/or lead to a MemoryError. This of course depends on the dataset, the problem and the machine I use. This PR adds...
When RAMP is used for developing models for a problem, we may want to tag certain versions of a submission, and even `problem.py`, together with the scores. One idea is...
While developing a new classifier class which has `fit`, `predict `and `predict_proba ` methods, it worked fine locally on jupyter but unfortunately it did not work with ramp tests. You...
Failing on Travis for Python 3.6 at the test kit 'digits': https://travis-ci.org/paris-saclay-cds/ramp-workflow/jobs/637847038?utm_medium=notification&utm_source=github_status Likely due to dependency version as the latest PR (#199) was green when merged. cc @kegl @agramfort
`ramp-test` should have an optional `--verbose` that is passed to `workflow.train_submission` and `workflow.test_submission` which can pass it to fit and predict.
Regarding #236 but also more generally: 1. What's the policy regarding backward compatibility with the [ramp-kits](https://github.com/ramp-kits)? Any change should be compatible with the kits in ramp-kits (or any change made...
`ramp-test` will expect to find the file `problem.py` and will fail when it is run and it will not fail gracefully in other case. It could be nicer to have...
I think that navigating and debugging the code could be made easier if there were different filenames for the workflow and the submission (for instance `classifier.py` is both a workflow...
This works with the master branch of ramp-board (only ramp-engine and ramp-utils is needed). The usage is like: ``` ramp-test --submission starting_kit_local2 --worker conda_config.yml # or ramp-test --submission starting_kit_local2 --worker...
Reverting some changes from my previous PR. We should use the built-in functionalities from pandas in this case.
Metadata
Owner
Metadata
Toolkit for building predictive workflows on top of pydata (pandas, scikit-learn, pytorch, keras, etc.).