Benjamin Bossan
Benjamin Bossan
Currently, I have only added a failing test and show a way to circumvent the error. I'm not sure what the proper way to fix this should be. BTW, using...
Loading features now matches the shapes of the layers using python's difflib.
As mentioned [here](https://github.com/dnouri/nolearn/pull/69), there are situations where the user wants more frequent feedback from the net than just after each epoch. Especially so with the arrival of RNNs, which are...
As discussed, the "Tests" and "Doc Tests" tasks are redundant. Since CI is already quite slow as is, we really don't want to run all the tests twice. Therefore, the...
Solves #111 [This](https://huggingface.co/skops-ci/hf_hub_example-c57afd8c-0f9f-4dcc-9fd0-29325a03cf83/tree/main) is the outcome of using `plot_hf_hub.py` with one added line: ```python hub_utils.add_files([__file__], dst=local_repo) ``` As can be seen, the script itself is now included in the upload.
Try 4 parallel, remove redundant tests
See if setting max parallelism affects the timeout issue.
Let's discuss the different options and their pros and cons. ## Problem sklearn models are typically persisted as pickle files. This can be done directly through `pickle.dump` or indirectly through...
This is an RFC for a suggested solution for the issues discussed on #72 ## Description The proposed model card implementation would allow to dynamically add sections or overwrite them....
Certain scipy functions cannot be loaded correctly. Here is a test to reproduce: ```python from scipy import stats def test_scipy_stats(tmp_path): estimator = FunctionTransformer(func=stats.zipf) save_load_round(estimator, tmp_path / "file.skops") ``` `zipf` is...