Mike McCarty

[BUG] 'normalize' in LinearRegression deprecated in scikit-learn 1.0

1

**Describe the bug** PR scikit-learn/scikit-learn#17743 released in sklearn v1.0 makes an API change that is not reflected in `cuml`. See [release notes](https://scikit-learn.org/stable/whats_new/v1.0.html#id15) for detail. **Steps/Code to reproduce bug** Follow this...

**What happened**: After following the [authentication instructions](https://cloudprovider.dask.org/en/latest/azure.html#overview) using the Azure CLI, I'm unable to create a cluster on Azure. **Minimal Complete Verifiable Example**: ```python cluster = AzureVMCluster( location=, resource_group=, vnet=,...

**What happened**: Compute calls with `schedule="single-threaded"` while using a distributed client may lead to unexpected results. This needs to be fixed in a few places. So far: - `dask_ml.datasets.make_classification` -...

azureml-sdk downgrades pyarrow to 3.0.0 which breaks cudf

2

### Steps to reproduce 1. Create a fresh RAPIDS conda environment `conda create -n rapids-22.06 -c rapidsai -c nvidia -c conda-forge rapids=22.06 python=3.8 cudatoolkit=11.5` 2. `conda activate rapids-22.06` 3. `conda...

Following up from #122

Versioned Caching

14

Intake caching was added to avoid the need to re-download remote data each time a catalog is loaded. Another use case for caching could be to provide versioning datasets for...