Add GPU telemetry with fallback and test coverage - #3
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This pull request introduces a new, decoupled GPU and wall-clock telemetry profiler for Patchwork, along with comprehensive tests and a manual verification script. The profiler is designed to be robust in environments without a GPU (such as CI), ensuring no crashes and clear error reporting. The main features include a context manager for VRAM profiling, a decorator for easy integration, and a Pydantic model for telemetry results.
Telemetry Profiler Implementation:
patchwork/telemetry/profiler.pyimplementing aGPUProfilercontext manager that tracks peak VRAM usage and duration, gracefully handling the absence of an NVIDIA GPU or driver, and providing aTelemetryResultPydantic model for structured results. Includes a decorator (profile_vram) and a utility (profile_call) for easy integration.Testing and Verification:
tests/test_profiler.pyto verify all fallback and error-handling paths, ensuring correct behavior when no GPU is present and validating JSON serialization of telemetry results.scripts/manual_profiler_test.pyfor manual, on-GPU verification of the profiler in real inference scenarios, allowing developers to confirm correct telemetry when a GPU is available.