Ncnn load net refactor #1023
Merged
Ncnn load net refactor #1023
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
As with the ONNX sessions, we were loading the NCNN model weights from cache to the Net object every time the upscale node was run. This refactor just reuses the ONNX pattern to cache the Net the first time it is loaded, and use that afterwards. I've been getting iterator run times 60-75% as long as pre-fix for smaller images. Also removes the CPU spikes from loading weights to the Net, though does not change high CPU usage for very fast iterator processing with small image/small model combinations.
Also fixed a bug with ncnn_auto_split_process not passing the input and output names when recursing.