dask
dask copied to clipboard
dask
Parallel computing with task scheduling
In #9483 @jrbourbeau added `median` and `median_approximate` to dask.dataframe. It would also be nice to have `DataFrame.groupby(...).median()`. This is challenging to do with an ACA-based approach for the same reason...
Looks like Scipy modify what is accepted as `popmean` in `scipy.stats.ttest_1samp` [`scipy=1.9.1`](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.ttest_1samp.html) >popmean:float or array_like > Expected value in null hypothesis. If array_like, then it must have the same shape...
**What happened**: If there's no statistics in parquet file I get unclear error message when trying to apply filter `TypeError: '
We are having an import problem in `compatibility.py` more specifically in this line https://github.com/dask/dask/blob/6c151021c313e5d29d8ecca37d98456e7a3e36c2/dask/compatibility.py#L21 ```python AttributeError: module 'importlib' has no attribute 'metadata' ``` https://github.com/dask/dask/actions/runs/3055182028/jobs/4927954949#step:5:66970 I'm not able to reproduce it...
This PR serves as a reference implementation for (what is expected to be) "Dask Design Proposal 1": https://github.com/dask/design-docs/pull/1 Required cudf branch for Dask-DataFrame dispatching to cudf: https://github.com/rjzamora/cudf/tree/backend-class
**What happened**: When using `npartitions="auto"` in `DataFrame.set_index()` on a local distributed cluster, a "Could not deserialize task" error occurs (see code and output below). This happens only when: 1. Instantiating...
We have a number of places where we issue user warnings during graph construction time that a user may not be doing what they want to be doing (e.g., [here](https://github.com/dask/dask/blob/785c8456ea2f62e41b0110b07e883eea42f969d2/dask/array/reshape.py#L261-L272))....
NumPy's NaN-skipping aggregation functions issue warnings when they encounter all NaN slices, e.g., ``` In [2]: np.nanmean([np.nan]) /Users/shoyer/miniconda3/envs/xarray-py36/bin/ipython:1: RuntimeWarning: Mean of empty slice #!/Users/shoyer/miniconda3/envs/xarray-py36/bin/python Out[2]: nan ``` As far as...
here is the bug: svd_compressed() fails for complex input #7639 and I have figured out where are the bugs and explaned how to fix this bug (maybe not the best...
