huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 20:34:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

Trying a few things.

Browse Source
This commit is contained in:
Nicolas Patry 2024-10-02 15:49:45 +02:00
parent f9abe4cd40
commit 2c12e937ce
No known key found for this signature in database
GPG Key ID: E939E8CC91A1C674
2 changed files with 6 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
.github/workflows/build.yaml vendored
View File

@ -193,5 +193,7 @@ jobs:
export DOCKER_DEVICES=${{ needs.build-and-push.outputs.docker_devices }} export DOCKER_DEVICES=${{ needs.build-and-push.outputs.docker_devices }}
export EXTRA_PYTEST="${{ needs.build-and-push.outputs.extra_pytest }}" export EXTRA_PYTEST="${{ needs.build-and-push.outputs.extra_pytest }}"
export HF_TOKEN=${{ secrets.HF_TOKEN }} export HF_TOKEN=${{ secrets.HF_TOKEN }}
export ROCR_VISIBLE_DEVICES=0,1
echo $DOCKER_IMAGE echo $DOCKER_IMAGE
rocm-smi
pytest -s -vv integration-tests ${PYTEST_FLAGS} ${EXTRA_PYTEST} pytest -s -vv integration-tests ${PYTEST_FLAGS} ${EXTRA_PYTEST}

5
integration-tests/conftest.py
View File

@ -534,6 +534,10 @@ def launcher(event_loop):
yield ContainerLauncherHandle(client, container.name, port) yield ContainerLauncherHandle(client, container.name, port)
import time
time.sleep(600)
if not use_flash_attention: if not use_flash_attention:
del env["USE_FLASH_ATTENTION"] del env["USE_FLASH_ATTENTION"]
@ -589,7 +593,6 @@ def generate_multi():
max_new_tokens: int, max_new_tokens: int,
seed: Optional[int] = None, seed: Optional[int] = None,
) -> List[Response]: ) -> List[Response]:
import numpy as np import numpy as np
arange = np.arange(len(prompts)) arange = np.arange(len(prompts))