NVIDIA / DALI

A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.
https://docs.nvidia.com/deeplearning/dali/user-guide/docs/index.html
Apache License 2.0
4.98k stars 609 forks source link

Use "device" and "preserve" built-in arguments in OpGraph2. #5516

Closed mzient closed 3 weeks ago

mzient commented 3 weeks ago

Category:

New feature (non-breaking change which adds functionality)

Description:

New OpNode has fields op_type and keep corresponding to OpSpec's "device" and "preserve" arguments. These are now parsed when adding a node to the graph.

Additional information:

Affected modules and functionalities:

New OpGraph and tests.

Key points relevant for the review:

Tests:

op_graph2_test.cc updated.

Checklist

Documentation

DALI team only

Requirements

REQ IDs: N/A

JIRA TASK: N/A

dali-automaton commented 3 weeks ago

CI MESSAGE: [15734268]: BUILD STARTED

dali-automaton commented 3 weeks ago

CI MESSAGE: [15734268]: BUILD PASSED