About obtaining ET before execution
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 196
- Forks
- 84
- PR merge metrics
- No merged PRs in 30d
Description
hello,I wonder if I can get an ET or some other information that can be used to generate chakra before the actual execution of the program, as I saw “We extended PyTorch to support trace collection with the Execution Graph Observer [1], which does not require intrusive modifications to ML models. Simply enabling the Execution Graph Observer before running a model yields traces that closely reflect real-world execution.” when I read the paper ，
but in the example,
the code use ExecutionTraceObserver instead of Execution Graph Observer，I think I only got the profile information after actually running the code here, whether to abandon the use of the Execution Graph Observer
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by comparing the paper's Execution Graph Observer with the example's PyTorch ExecutionTraceObserver, focusing on when each produces execution information. Trace how the resulting ET is used to generate Chakra data. Done means the repository documents whether ET can be obtained before execution and, if not, clearly identifies the supported workflow or required design work.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100