- The Matrix
- https://marksaroufim.substack.com/
- @marksaroufim
Block or Report
Block or report msaroufim
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abusePinned
-
pytorch/pytorch Public
Tensors and Dynamic neural networks in Python with strong GPU acceleration
-
-
pytorch/torchx Public
TorchX is a universal job launcher for PyTorch applications. TorchX is designed to have fast iteration time for training/research and support for E2E production ML pipelines when you're ready.
-
pytorch/data Public
A PyTorch repo for data loading and utilities to be shared by the PyTorch domain libraries.
-
pytorch/examples Public
A set of examples around pytorch in Vision, Text, Reinforcement Learning, etc.
-
pytorch/torchdynamo Public
A Python-level JIT compiler designed to make unmodified PyTorch programs faster.
1,005 contributions in the last year
Activity overview
Contribution activity
April 2023
Created 12 commits in 2 repositories
Created 1 repository
- msaroufim/xla C++
Created a pull request in pytorch/serve that received 22 comments
Refactor PT2 code changes
Description Note: the link check lint keeps timing out so i might ignore it for now Now that PT 2.0 was official released there's a few updates we …
Opened 5 other pull requests in 1 repository
Reviewed 35 pull requests in 6 repositories
pytorch/serve
24 pull requests
-
Enable
torch.compilesupport for torch 2.0.0 pre release - open github issue on regression failure
- Refactoring dependencies for installing TorchServe
- FAQ on slow model initialization
- Revert "[easy] lint fix broken windows conda link"
- Updates to TorchServe page on pytorch.org
- Large model inference
- move misplaced config-file arg from torchserve to torch-model-archiver
- http stream response via http 1.1 chunked encoding
- Refactor PT2 code changes
- fix core pinning enabling in cpu nightly benchmark
- fixes error thrown while loading multiple models
- Automate benchmark validation
- Enable worker core pinning in CPU nightly benchmark
- Remove generation of graph plots from benchmark runs
- deepspeed base handler and example
- feat: add PyTorch/XLA support
- Refactor sanity tests
- Make sanity test output useable again
- Added test cases for SageMaker MME requirements
- TorchServe decompress tgz format model
- Refactor Dockerfile: Fix Python version; Add CI workflow to test Docker container
- Fix model-archiver to fail if extra files are missing
- Enable regression tests to run daily






