-
-
Currently, RAPIDS deploys a forked version of this repository with additions that support a few things:
1. New Random Forests (RF) interface in XGBoost
2. Ingest of `dask-cudf` objects
3. Ingest …
-
http://scikit-learn.org/stable/developers/contributing.html#rolling-your-own-estimator
It'd be nice if these passed.
-
**Minimal Complete Verifiable Example**:
```python
import dask.dataframe as dd
df = dd.demo.make_timeseries()
# Works with divisions
df.x.rolling("50D").mean().compute()
# timestamp
# 200…
-
Graphs like this are not currently scheduled well:
```
. . . . . . . . . . . . . . . .
|\|\|\|\|/|/|/| |\|\|\|\|/|/|/|
| | | | a | | | | | | | b | | |
* * * * * * * * * * * * * * * *
```…
-
In https://github.com/dask/dask/pull/9175 we added automatic retries when using remote filesystems with `dd.read_parquet` / `dd.to_parquet`. It'd be nice for other I/O functions (e.g. `dd.read_csv` / …
-
Currently if `Ensemble.from_dask_dataframe` loads a dataframe with an index with the same label as the column mapper, the following warning is produced:
`dask/dataframe/core.py:5251: UserWarning: N…
-
As a follow-up to https://github.com/coiled/dask-mongo/pull/2, we should update `dask-mongo.read_mongo` to use `HighLevelGraph` layers (specifically `DataFrameIOLayer`). This should look very similar …
-
Newly cloned, and running pytest results in
`16 failed, 979 passed, 496 skipped, 23 xfailed, 112 xpassed, 115 warnings, 102 error`
This is with pytest 3.10 and pytest-tornado installed. Full vers…
-
I'm interesting in writing distributed code that runs using Dask, but that checkpoints at certain points, so that it can resume from these points in case of a failure. This is distinct from the [oppor…