[FEAT] Add DataFrame.to_torch_map_dataset and .to_torch_iter_dataset.
#1086
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Codecov Report
Additional details and impacted files@@ Coverage Diff @@
## main #1086 +/- ##
==========================================
- Coverage 88.87% 88.33% -0.55%
==========================================
Files 53 54 +1
Lines 5439 5477 +38
==========================================
+ Hits 4834 4838 +4
- Misses 605 639 +34
|
clarkzinzow
reviewed
Jun 23, 2023
There was a problem hiding this comment.
@xcharleslin Meta comment: we might want to subclass the new canonical Torch data abstractions, IterDataPipe and MapDataPipe.
github-actions
bot
added
the
documentation
|
@clarkzinzow Hahaha, was just gonna ask for your input on this one :) I'll do that, thanks! Do the API naming and docs look good to you as well? |
xcharleslin
changed the title
DataFrame.to_torch_map_dataset and .to_torch_iter_dataset.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adds two new top-level APIs to DataFrame: to_torch_map_dataset and to_torch_iter_dataset, that returns respective PyTorch datasets (TODO: make these DataPipes).
to_torch_map_datasetwill execute the whole dataframe before returning, since it needs random access.to_torch_iter_datasetwill return immediately; results are returned via streaming execution.Both are only meant for use in a single-node setting, with Ray Datasets being the recommended data loading abstraction for distributed training.