Mark Saroufim
Mark Saroufim
Tried to run the test locally and discovered that `AOTAutograd` has been changed to `AotAutograd` which broke 7 of the dynamo cuda graphs tests I guess this definitively answers the...
**Is your feature request related to a problem? Please describe.** Right now the way to run scalene is to assume some input `.py` script to do `scale my_prog.py` and get...
## Description Adds an alert on failed Github scheduled Github Action. This is the exact same trick used in [pytorch/examples ](https://github.com/pytorch/examples) Code borrowed from: https://github.com/pytorch/examples/blob/main/.github/workflows/main.yml#L34 And a test showing it...
Fixes https://github.com/pytorch/serve/issues/1776 Test: https://colab.research.google.com/drive/1eKMiheF50YYGtPmi0WqVkeRqLk8Gbwg5?usp=sharing (would probably be better to start writing unit tests for the python backend when we have time) The meaningful code is at the very end of...
## Description This PR supports a manually triggerable Github Action that will make an official release. The way this works is assuming a code freeze, we will run a bunch...
After looking into #1744 I noticed we don't actually use our `docs/sphinx/requirements.txt` in CI, this is not great because if there's issues with upstream dependencies like `markdown` it means we...
### 🐛 Describe the bug Our latest doc build is crashing https://github.com/pytorch/serve/runs/7361642275?check_suite_focus=true with this error `ModuleNotFoundError: No module named 'ts_scripts'` @svekars does this look familiar? ### Error logs ``` Run...
## Modularize `base_handler.py` into `handler.utils` This is a follow-up note on #1440 where we initially noted a few problems that were making it difficult to add code to TorchServe Initially,...
### 🚀 The feature We're close to fully automated releases with our new nightly build setup by leveraging nightly builds. The remaining gaps are * [x] #1620 * [ ]...
### 🚀 The feature Wrote this after an offline discussion with @lxning We've recently standardized on using `benchmarks/benchmark-ab.py` as **the** preferred way to benchmark torchserve models and used it to...