Lucas Kabela

[DRAFT] Initial LLM collector

1

## Description Add LLM Collector ## Motivation and Context #2872 ## Types of changes What types of changes does your code introduce? Remove all that do not apply: - [...

[GraphPartition] cache get_free_symbol_uses (#166338)

3

Graph partition relies on `get_free_symbol_uses()` to collect symbol inputs. https://github.com/pytorch/pytorch/blob/ee7434be822cf6e75b4566d8159f550ee233d8ae/torch/_inductor/scheduler.py#L4869-L4885 I empirically observed that `get_free_symbol_uses()` becomes slower for larger graphs. Specifically, I tried to aten fallback for torchtitan which results...

[dynamo] fix keyerror in resume_execution, fix store attr

1

Fixes #166176 See also stack on https://github.com/pytorch/pytorch/pull/166036 cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang @amjames

[BE][Typing][Dynamo] Type torch/_dynamo/variables/dicts.py

2

Provides type coverage to torch/_dynamo/variables/dicts.py Coverage report: `mypy torch/_dynamo/variables/dicts.py --linecount-report /tmp/coverage_log` Compare before to after - we go from 0 lines and 0 funcs covered to XYZ lines and ABC...

[BE][Typing][Dynamo] Type torch/_dynamo/variables/user_defined.py

20

Provides type coverage to torch/_dynamo/variables/user_defined.py Coverage report: mypy torch/_dynamo/variables/user_defined.py --linecount-report /tmp/coverage_log Compare before to after - we go from 0 lines and 0 funcs covered to 2278 lines and 106...