-
Notifications
You must be signed in to change notification settings - Fork 599
Implement unary_ufunc functions using elementwise_util #9386
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/9386
Note: Links to docs will display an error until the docs builds have been completed. ⏳ 6 Pending, 1 Unrelated FailureAs of commit d14a503 with merge base 2dedc9e ( BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
ScalarType compute_type = in.scalar_type() == ScalarType::Double | ||
? ScalarType::Double | ||
: ScalarType::Float; | ||
ET_SWITCH_FLOAT_TYPES(compute_type, ctx, op_name, CTYPE_COMPUTE, [&] { |
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 is the only ET_SWITCH change remaining -- we are switching over the possible compute types, not the input/output types.
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
tests looking good; few stragglers but it's only llava and then one each of qnn/arm which shouldn't be affected |
This reverts commit dffbd42.
One less set of independent implementations to worry about going forward (e.g., we don't have to vectorize these separately from elementwise_util and they get all benefits of elementwise_util).