PennyLaneAI / PennyLaneAI/catalyst
Improve error handling / add type promotion to AutoGraph loops
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- Python
- Stars
- 234
- Forks
- 84
- Avg merge
- 2d 15h
- Merged PRs (30d)
- 66
Description
Two issues are uncovered in the referenced discussion:
- Which errors are raised during a fallback to Python? Sometimes it is clear the user intended to perform an autograph conversion, but the fallback to Python can obscure the error messaging.
- Can we support type promotion for autograph captured control flow, on pre-initialized variables?
Is type promotion supported here?
I suppose the following code should pass:
def f(pred: bool):
x = 1
while pred:
x = 3.2
return x
Originally posted by @maliasadi in https://github.com/PennyLaneAI/catalyst/pull/352#discussion_r1384077617
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
Read the referenced discussion in PR #352 first, then reproduce the Python loop example and inspect the AutoGraph fallback behavior it describes. Done means the intended conversion produces clear errors and the captured control flow supports the proposed type promotion for pre-initialized variables.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- compilers
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100