search

cohere-ai / cohere-aws

MIT License
58 stars 9 forks source link

Update Deploy rerank multilingual v3.0 model.ipynb #153

Closed pradeep-prabhakaran closed 2 months ago

pradeep-prabhakaran commented 2 months ago
  1. Updated the endpoint name from 'cohere-rerank-multilingual-v3.0. to 'cohere-rerank-multilingual-v3-0
  2. Updated code with supported instance type ml.g5.2xlarge
pradeep-prabhakaran commented 2 months ago

Yes. I was going to do it tonight. Thanks

On Wed, Aug 7, 2024, 6:34 PM Leila @.***> wrote:

@.**** approved this pull request.

In notebooks/sagemaker/rerank_v3_notebooks/Deploy rerank multilingual v3.0 model.ipynb https://github.com/cohere-ai/cohere-aws/pull/153#discussion_r1708102387:

@@ -138,7 +138,7 @@ "outputs": [], "source": [ "co = Client(region_name=region)\n",

  • "co.create_endpoint(arn=model_package_arn, endpoint_name=\"cohere-rerank-multilingual-v3.0\", instance_type=\"ml.g4dn.xlarge\", n_instances=1)\n",
  • "co.create_endpoint(arn=model_package_arn, endpoint_name=\"cohere-rerank-multilingual-v3-0\", instance_type=\"ml.g5.2xlarge\", n_instances=1)\n",

can you update the english notebook to match, please? and let's use ml.g5.xlarge since that's 'vendor recommended' https://aws.amazon.com/marketplace/pp/prodview-ydysc72qticsw

— Reply to this email directly, view it on GitHub https://github.com/cohere-ai/cohere-aws/pull/153#pullrequestreview-2226354112, or unsubscribe https://github.com/notifications/unsubscribe-auth/BC2DLN7SW64QGXZXFWKIUCDZQKVJRAVCNFSM6AAAAABMFKGAWCVHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZDEMRWGM2TIMJRGI . You are receiving this because you authored the thread.Message ID: @.***>