Skip to content

Commit

Permalink
Cross-link cudf.pandas profiler documentation. (#17668)
Browse files Browse the repository at this point in the history
Adds a cross-link to the cudf.pandas profiler docs. This cross-linking would have helped answer a user question about how to profile.

Authors:
  - Bradley Dice (https://github.com/bdice)
  - GALI PREM SAGAR (https://github.com/galipremsagar)

Approvers:
  - GALI PREM SAGAR (https://github.com/galipremsagar)

URL: #17668
  • Loading branch information
bdice authored Jan 5, 2025
1 parent 07ee82b commit 955b1f4
Show file tree
Hide file tree
Showing 2 changed files with 6 additions and 5 deletions.
10 changes: 5 additions & 5 deletions docs/cudf/source/cudf_pandas/faq.md
Original file line number Diff line number Diff line change
Expand Up @@ -63,11 +63,11 @@ keyword arguments, cuDF is not able to provide GPU acceleration and
`cudf.pandas` will fall back to the CPU.

The most accurate way to assess which functions run on the GPU is to try
running the code while using the `cudf.pandas` profiling features. The
profiler will indicate which functions ran on GPU / CPU. To improve
performance, try to use only functionality that can run entirely on GPU.
This helps reduce the number of memory transfers needed to fallback to
CPU.
running the code while using the `cudf.pandas` [profiling
features](cudf-pandas-profiling). The profiler will indicate which functions
ran on GPU / CPU. To improve performance, try to use only functionality that
can run entirely on GPU. This helps reduce the number of memory transfers
needed to fallback to CPU.

## How can I improve performance of my workflow with `cudf.pandas`?

Expand Down
1 change: 1 addition & 0 deletions docs/cudf/source/cudf_pandas/usage.md
Original file line number Diff line number Diff line change
Expand Up @@ -75,6 +75,7 @@ with Pool(4) as pool:
...
```

(cudf-pandas-profiling)=
## Profiling `cudf.pandas`

`cudf.pandas` will attempt to use the GPU whenever possible and fall
Expand Down

0 comments on commit 955b1f4

Please sign in to comment.