meta-pytorch / meta-pytorch/data

[DataPipe] Ensure all DataPipes Meet Testing Requirements

Open
#106 5 comments 2 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
1.3k
Forks
179
Avg merge
6d 1h
Merged PRs (30d)
2

Description

🚀 Feature

We have many tests for existing DataPipes (both in PyTorch Core and TorchData). However, over time, they have become less organized. Moreover, as the testing requirements expand, older DataPipes may not have tests to cover the newly added requirements.

This issue aims to track the status of tests for all DataPipes.

Motivation

We want to ensure test coverage for all DataPipe is complete to reduce bugs and unexpected behavior.

Alternative

We also should create some testing templates for IterDataPipe and MapDataPipe that can be widely applied.

IterDataPipe Tracker

X - Done
NA - Not Applicable
Blank - Not Done/Unclear

Test definitions:
Functional - unit test to ensure that the DataPipe works properly with various input arguments
Reset - DataPipe can be reset/restart after being read
__len__ - the __len__ method is implemented whenever possible (or explicitly not implemented)
Serializable - DataPipe is serializable
Graph (future) - can be traversed as part of a DataPipe graph
Snapshot (future) - can be saved/loaded as a checkpoint/snapshot

Name Module Functional Test Reset __len__ Serializable (Pickable) Graph Snapshot
Batcher Core X X X X
Collator Core X X X X
Concater Core X X X X
Demultiplexer Core X X X X
FileLister Core X X X X
FileOpener Core X X X X
Filter Core X X X X
Forker Core X X X X
Grouper Core X X X
IterableWrapper Core X X X X
Mapper Core X X X X
Multiplexer Core X X X X
RoutedDecoder Core X X X X
Sampler Core X X X X
Shuffler Core X X X X
StreamReader Core X X X X
UnBatcher Core X X X
Zipper Core X X X X
BucketBatcher Data X X X X
CSVDictParser Data X X X X
CSVParser Data X X X X
Cycler Data X X X X
DataFrameMaker Data X X X X
Decompressor Data X X X X
Enumerator Data X X X X
FlatMapper Data X X X X
FSSpecFileLister Data X X X X
FSSpecFileOpener Data X X X X
FSSpecSaver Data X X X X
GDriveReader Data X X X X
HashChecker Data X X X X
Header Data X X X X
HttpReader Data X X X X
InMemoryCacheHolder Data X X X X
IndexAdder Data X X X X
IoPathFileLister Data X X X X
IoPathFileOpener Data X X X X
IoPathSaver Data X X X X
IterKeyZipper Data X X X X
JsonParser Data X X X X
LineReader Data X X X X
MapKeyZipper Data X X X X
OnDiskCacheHolder Data X X X X
OnlineReader Data X X X X
ParagraphAggregator Data X X X X
ParquetDataFrameLoader Data X X X X
RarArchiveLoader Data X X X X
Rows2Columnar Data X X X X
SampleMultiplexer Data X X X X
Saver Data X X X X
TarArchiveLoader Data X X X X
UnZipper Data X X X X
XzFileLoader Data X X X X
ZipArchiveLoader Data X X X X

MapDataPipe Tracker

X - Done
NA - Not Applicable
Blank - Not Done/Unclear

Name Module Functional Test __len__ Serializable (Pickable) Graph Snapshot
Batcher Core X X
Concater Core X X
Mapper Core X X X
SequenceWrapper Core X X X
Shuffler Core X X
Zipper Core X X

cc: @ejguan @VitalyFedyunin @NivekT

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading the linked DataPipes testing requirements and comparing the IterDataPipe and MapDataPipe trackers in this issue. Identify an individual blank or unclear requirement, inspect its existing tests in PyTorch Core or TorchData, and consider the work done when the selected DataPipe meets the applicable functional, reset, length, or serialization requirement.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
data-engineering, testing-qa
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.