Debug/Logging #41
Open
Debug/Logging #41
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
* Peppa Pig Teacher Face Landmark Detector. (Thanks to PINTO0309) * 2d Fan 4 Face Landmark Detector. (Thanks to PINTO0309) * Yunet N 640x640 Face Detector. (Thanks to ShiqiYu) * 3d68 Face Landmark Detector. (Thanks to InsightFace) * 106 Face Landmark Detector. (Thanks to InsightFace) * Google FaceMesh V2. (Thanks to Google and PINTO0309)
* Peppa Pig Teacher Face Landmark Detector. (Thanks to PINTO0309) * 2d Fan 4 Face Landmark Detector. (Thanks to PINTO0309) * Yunet N 640x640 Face Detector. (Thanks to ShiqiYu) * 3d68 Face Landmark Detector. (Thanks to InsightFace) * 106 Face Landmark Detector. (Thanks to InsightFace) * Google FaceMesh V2. (Thanks to Google and PINTO0309)
…ctor keypoints, the image will be rotated before processing by Landmarks Detector. * Add support for GPEN 1024 restorer. * Fix issue regarding settings not automatically applied when changed. * Fix typo-error. * Improve Landmarks Detectors code. * General bug fixing.
…rks when the image processed by the 'Landmark' is smaller than the required size and from points setting was not enabled. * Add support for 5-points Landmark detector. * Add Show Landmarks setting.
…t keypoints value exceed the image size.
through the CPU. This significantly increases DFL model inference performance. Fix issues for histogram_matching_DFL_Orig and histogram_matching_DFL_test functions.
Add trt_max_workspace_size property to trt_ep_options variable.
to work with TensorRT/TensorRT-Engine providers, even for those with morph_value:0 as an input parameter.
Add two functions in order to use kornia.geometry.transform.warpaffine.
histogram_matching_DFL_test functions. Now they work without permute operation.
Optimize the apply_face_parser function for performance.
into the proper files.
* Lay the groundwork for supporting new providers. * Optimize CLIPSwitch inference to work entirely on CUDA tensors or CPU tensors.
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.
Because my GPU is ancient EVGA Geforce GTX 1080ti, I often run into issues where my display or my computer shuts down. Sometimes I am able to use ctrl-shift-windows-[b] to reset my graphics but not all the time.
I am implemented a debug mode that allows for monitoring of VRAM, virtual memory, GPU Utilization and GPU temps. With this info, I can better determine what is causing the failure. Thank you for your consideration.