Support Parquet files in ShardedDataSource #764
Merged
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
percyliang
reviewed
Oct 14, 2024
tests/test_sharded_dataset.py
Outdated
| table = pa.Table.from_pydict(data) | ||
| pq.write_table(table, f.name) | ||
|
|
||
| try: |
There was a problem hiding this comment.
Why not put this within the context manager and set delete to True?
percyliang
approved these changes
Oct 14, 2024
dlwh
reviewed
Oct 14, 2024
| @@ -238,6 +243,11 @@ def open_shard_at_row(self, shard_name: str, row: int) -> Iterator[str]: | |||
| data = json.load(f) | |||
| for doc in data[row:]: | |||
| yield doc[self.text_key] | |||
| case ".parquet": | |||
| table = pq.read_table(f) | |||
There was a problem hiding this comment.
read_table is expensive in the general case and it would be better to look at the metadata to figure out which row group to start on and then use read_row_group I think
There was a problem hiding this comment.
Thanks for suggesting this, incorporated this in #766
| @@ -417,6 +427,24 @@ def open_shard_at_row(self, shard_name: str, row: int) -> Iterator[dict]: | |||
| return iter(data[row:]) | |||
|
|
|||
|
|
|||
| class ParquetDataSource(ShardedDataSource[dict]): | |||
There was a problem hiding this comment.
ideally the TextUrlDataSource would also work with parquet files. That is what we use for training configs typically
There was a problem hiding this comment.
Makes sense, I added a new test for this in #766
This was referenced Oct 15, 2024
dlwh
pushed a commit
that referenced
this pull request
Oct 15, 2024
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Outdated suggestions cannot be applied.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.
This PR creates a
ParquetDataSourceclass to support loading.parquetfiles.Closes #763