-
Notifications
You must be signed in to change notification settings - Fork 1.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Evaluating distill models #3446
Conversation
@microsoft-github-policy-service agree company="Microsoft" |
sdk/python/distillation/benchmarking/pipelines/conversation/hellaswag.yaml
Outdated
Show resolved
Hide resolved
sdk/python/distillation/benchmarking/evaluating_distill_models.ipynb
Outdated
Show resolved
Hide resolved
sdk/python/distillation/benchmarking/evaluating_distill_models.ipynb
Outdated
Show resolved
Hide resolved
sdk/python/distillation/benchmarking/evaluating_distill_models.ipynb
Outdated
Show resolved
Hide resolved
sdk/python/distillation/benchmarking/pipelines/conversation/hellaswag.yaml
Outdated
Show resolved
Hide resolved
I believe it would be great for customers if we could include the code to deploy the distilled model, create the connection, and then submit the evaluation job. The idea is to simplify the customer journey, otherwise they will have to manually deploy the model and manually create connection, which introduces chances of error while doing these operations. All of this can be part of the distillation notebook, post getting the distilled model. What do you think? |
...on/foundation-models/system/distillation/conversation/distillation_conversational_task.ipynb
Outdated
Show resolved
Hide resolved
...on/foundation-models/system/distillation/conversation/distillation_conversational_task.ipynb
Outdated
Show resolved
Hide resolved
sdk/python/foundation-models/system/distillation/conversation/pipelines/hellaswag.yaml
Outdated
Show resolved
Hide resolved
Benchmark validation runs
|
Description
Checklist