Guang Yang

Fail fast when dynamo attempts to add unspecialized int/float as additional graph inputs

11

Summary: Verified the changes to catch unspecialized int/floats being added as additional graph in D44037548 prior to RP(https://github.com/pytorch/pytorch/pull/95621). However, with #95621 the issue to be solved originally is no longer...

[v0.2.0] Release Tracker

91

v0.2.0 release branch was cut. The branch name is "[release/0.2](https://github.com/pytorch/executorch/tree/release/0.2)", with [tag "v0.2.0-rc4"](https://github.com/pytorch/executorch/tree/v0.2.0-rc4) atm. Our release plan from this point is roughly: * **Phase 1 (until 4/5/24):** -- work on...

This PR is a prototype to showcase the minimal changes required to lower Gemma-2b to ExecuTorch w/ static kv cache and run it directly in [llama runner](https://github.com/pytorch/executorch/tree/main/examples/models/llama2) w/o single line...

Support devices and delegates parameters in android-perf workflow

4

### Summary: The workflow can support parameters `models`, `devices`, `delegates`. The workflow is able to translate `devices` to `device-pool-arn` and pass to the `mobile_job` for scheduling The model is packed...

Clean up llava deps and consolidate all HF deps

3

`python3 -m examples.portable.scripts.export --model_name="llava"` works fine on my mac M1. No strict mode is needed. And can directly working on the most recent transformers lib. It seems like the old...

The apk downloaded from S3 can not be unzipped

1

### 🐛 Describe the bug `Android/test-llama-app (mobile-job)` fails due to corrupted apk downloaded from S3. See logs: ``` Exception: Couldn't upload ci.apk: {'upload': {'arn': 'arn:aws:devicefarm:us-west-2:308535385114:upload:02a2cf0f-6d9b-45ee-ba1a-a086587469e6/9efb530a-d0ab-4776-96b8-b3ef3cd25848', 'name': 'mobile-job-android-9954110435-1-2024-07-16-ZOfaUHWN_ci.apk', 'created': datetime.datetime(2024, 7,...

[Android] Set mobile-job workflow as a periodic job

1

### 🚀 The feature, motivation and pitch benchmark job to run ET models on device farm ### Alternatives _No response_ ### Additional context _No response_ ### RFC (Optional) _No response_

[Not To Land][HF][Gemma] Lower and run HF Gemma2b in ExecuTorch

3

This PR is a prototype to showcase the minimal changes required to lower Gemma-2b to ExecuTorch w/ static kv cache and run it directly in [llama runner](https://github.com/pytorch/executorch/tree/main/examples/models/llama2) w/o single line...

### 🐛 Describe the bug Follow up guide here: https://pytorch.org/executorch/main/build-run-qualcomm-ai-engine-direct-backend.html#setting-up-your-developer-environment Then run `./backends/qualcomm/scripts/build.sh --release`, will hit the error: ``` ~/executorch/examples/qualcomm/llama2/qaihub_runner/runner.cpp:21:10: fatal error: 'executorch/examples/models/llama2/runner/util.h' file not found 21 | #include |...