You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@beam.apache.org by da...@apache.org on 2024/11/19 15:52:59 UTC
(beam) branch revert-33122-users/damccorm/vllmColab created (now 7f5a575d11e)
This is an automated email from the ASF dual-hosted git repository.
damccorm pushed a change to branch revert-33122-users/damccorm/vllmColab
in repository https://gitbox.apache.org/repos/asf/beam.git
at 7f5a575d11e Revert "Update vllm colab to work with asyncio requirements (#33122)"
This branch includes the following new commits:
new 7f5a575d11e Revert "Update vllm colab to work with asyncio requirements (#33122)"
The 1 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
(beam) 01/01: Revert "Update vllm colab to work with asyncio requirements (#33122)"
Posted by da...@apache.org.
This is an automated email from the ASF dual-hosted git repository.
damccorm pushed a commit to branch revert-33122-users/damccorm/vllmColab
in repository https://gitbox.apache.org/repos/asf/beam.git
commit 7f5a575d11e9c7c419f88a77273c4aebddde14c4
Author: Danny McCormick <da...@google.com>
AuthorDate: Tue Nov 19 10:52:52 2024 -0500
Revert "Update vllm colab to work with asyncio requirements (#33122)"
This reverts commit ca92726144f5cd789c4f0e0ad1989c09ddedf36e.
---
.../notebooks/beam-ml/run_inference_vllm.ipynb | 27 +---------------------
1 file changed, 1 insertion(+), 26 deletions(-)
diff --git a/examples/notebooks/beam-ml/run_inference_vllm.ipynb b/examples/notebooks/beam-ml/run_inference_vllm.ipynb
index 5608bb4cb0b..13b4a915c08 100644
--- a/examples/notebooks/beam-ml/run_inference_vllm.ipynb
+++ b/examples/notebooks/beam-ml/run_inference_vllm.ipynb
@@ -117,35 +117,10 @@
"source": [
"!pip install openai>=1.52.2\n",
"!pip install vllm>=0.6.3\n",
- "!pip install apache-beam[gcp]==2.61.0\n",
- "!pip install nest_asyncio # only needed in colab\n",
+ "!pip install apache-beam[gcp]==2.60.0\n",
"!pip check"
]
},
- {
- "cell_type": "markdown",
- "source": [
- "## Colab only: allow nested asyncio\n",
- "\n",
- "The vLLM model handler logic below uses asyncio to feed vLLM records. This only works if we are not already in an asyncio event loop. Most of the time, this is fine, but colab already operates in an event loop. To work around this, we can use nest_asyncio to make things work smoothly in colab. Do not include this step outside of colab."
- ],
- "metadata": {
- "id": "3xz8zuA7vcS3"
- }
- },
- {
- "cell_type": "code",
- "source": [
- "# This should not be necessary outside of colab.\n",
- "import nest_asyncio\n",
- "nest_asyncio.apply()\n"
- ],
- "metadata": {
- "id": "sUqjOzw3wpI3"
- },
- "execution_count": null,
- "outputs": []
- },
{
"cell_type": "markdown",
"source": [