Olivier Grisel
Olivier Grisel
The new Firefox "Inspect Element" dark modal background is very neat and would probably fit the create edition pattern very well (along with the out of area toolbar). Make sure...
When the number of features is much bigger than the number of samples I get a segmentation fault. The following script can reproduce the problem: import numpy as np from...
At the moment we do not have systematic support for very large sparse matrices in our Cython code. That would be useful when the data is passed as a sparse...
Basically combining `TimeSeriesSplit` with the `Group` awareness of other CV strategies such as `GroupKFold`. I think it's a good first issue for first time contributors that are already familiar with...
Observed several times: - ` linux_py310_distributed` in #1272: https://dev.azure.com/joblib/joblib/_build/results?buildId=1849&view=logs&j=e366bf14-feff-5258-9913-c0ec433aa72d&t=db9d9cf0-46c0-52ca-1b5f-c25993207165&l=1312 - `linux_py37_distributed` in #1264 - `linux_py310_distributed` in https://dev.azure.com/joblib/joblib/_build/results?buildId=1874&view=logs&j=e366bf14-feff-5258-9913-c0ec433aa72d ``` 2022-02-23T13:12:02.7548426Z =================================== FAILURES =================================== 2022-02-23T13:12:02.7549232Z _________ test_nested_loop_error_in_grandchild_resource_tracker_silent _________ 2022-02-23T13:12:02.7549590Z 2022-02-23T13:12:02.7549891Z @with_numpy 2022-02-23T13:12:02.7550280Z...
It looks like a real deadlock. Observed in: - macos_py37_no_numpy #1254 https://dev.azure.com/joblib/joblib/_build/results?buildId=1868&view=logs&j=e4a78a51-caef-5527-76d7-fb6a1ee31c5a&t=bdaca537-7df6-5ee5-a116-96a6b8e69e1a ``` 2022-02-25T14:59:04.4467470Z =================================== FAILURES =================================== 2022-02-25T14:59:04.4469870Z _ test_parallel_with_unpicklable_functions_in_args[kwargs-def gen_square():\n def square(x):\n return x ** 2\n return square\nsquare =...
Observed after the merge of loky 3.1.0. - `windows_py310` on #1274 https://dev.azure.com/joblib/joblib/_build/results?buildId=1852&view=logs&j=a35b4612-962d-5bf3-96de-7e8bc5750ea4&t=0182bc6c-d367-55f8-ab6b-9f319778ff9d and: - `windows_py310` on #1272 https://dev.azure.com/joblib/joblib/_build/results?buildId=1850&view=logs&j=8008bb8c-6b4f-56d5-66f4-b1d037aba1bf&t=7af8808d-3494-56e0-c1dc-6fbf5bb2bc40 ``` 2022-02-23T13:35:43.2669420Z ================================== FAILURES =================================== 2022-02-23T13:35:43.2670528Z _______ test_multithreaded_parallel_termination_resource_tracker_silent _______ 2022-02-23T13:35:43.2671117Z 2022-02-23T13:35:43.2671822Z @with_numpy...
Trying to reproduce the `test_nested_exception_dispatch[loky]` failure observed in #1055 as it seems unrelated to dask.
The development version of scikit-learn now uses a unified loss module for several estimators including Logistic Regression: https://github.com/scikit-learn/scikit-learn/pull/21808 This causes the following problem when using the latest version of scikit-learn-intelex...
As implemented in the Cython version in scikit-learn: https://github.com/scikit-learn/scikit-learn/pull/18242