-
Notifications
You must be signed in to change notification settings - Fork 402
Use remote models for test #324
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
Conversation
switch variant { | ||
case .largev3: | ||
modelPath = try largev3ModelPath() | ||
modelName = "large-v3" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
should we consider updating model here to v20240930 variant and compressed to download less?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This code path isnt actively tested in the unit tests, and has been superseded but the regression test suite, so probably just a todo to remove it in the future
default: | ||
modelPath = try tinyModelPath() | ||
modelName = "tiny" | ||
} | ||
let computeOptions = ModelComputeOptions( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
why cpu only? should this be only if target is simulator?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Pretty old commit, may have been intended to keep the unit test output consistent across chips.
This will allow the unit tests to be run from the example WhisperAX app without needing to run
make setup
or download any models manually, which is useful for testing environments that use physical devices and require a host app.