[Filter] Allow dynamic framerate changes #4078
Open
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
…on found for /usr/local/lib/libtensorflow.so.2 '
…on found for /usr/local/lib/libtensorflow.so.2 '
** (process:615338): CRITICAL **: 07:37:24.833: Exception while loading the model: Legacy model format is not supported on mobile.
… == true. This fixes: "RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!"
Merge upstream changes
Allow tensor_filter to be configured with smaller number of output tensors than PyTorch returns natively. Signed-off-by: Torsten Schulz <torsten@eyepop.ai>
* Non-maintainer upload. * Allow PyTorch filter to be configured with less output tensors than PyTorch natively returns
One might want to dynamically change the framerate at any time while the pipeline is running. This is prevented by code in `tensor_converter` and `tensor_filter` that, upon reconfigure events, requre the new caps to be 'compatible' with the orignal caps on the sink pads. Requiring the caps to have matching framerates seems unnecessary and too restrictive. This patch removes that restriction from both `tensor_converter` and `tensor_filter` elements. A new unit test for this case has been added. Signed-off-by: Torsten Schulz <torsten@eyepop.ai>
[Filter] Allow dynamic framerate changes
|
📝 TAOS-CI Version: 1.5.20200925. Thank you for submitting PR #4078. Please a submit 1commit/1PR (one commit per one PR) policy to get comments quickly from reviewers. Your PR must pass all verificiation processes of cibot before starting a review process from reviewers. If you are new member to join this project, please read manuals in documentation folder and wiki page. In order to monitor a progress status of your PR in more detail, visit http://ci.nnstreamer.ai/. |
|
|
cibot: 
tschulz
changed the title
myungjoo
changed the title
|
myungjoo
reviewed
Mar 28, 2024
| @@ -3,19 +3,20 @@ Section: libs | |||
| Priority: optional | |||
| Maintainer: MyungJoo Ham <myungjoo.ham@samsung.com> | |||
| Build-Depends: gcc-9 | gcc-8 | gcc-7 | gcc-6 | gcc-5 (>=5.4), | |||
| <<<<<<< HEAD | |||
| ninja-build, meson (>=0.50), debhelper (>=9), | |||
| ======= | |||
myungjoo
reviewed
Mar 28, 2024
| @@ -1,14 +1,82 @@ | |||
| nnstreamer (2.4.1.0) unstable bionic focal jammy sid; urgency=medium | |||
| nnstreamer (2.3.0.0+ep1) UNRELEASED; urgency=medium | |||
There was a problem hiding this comment.
Not merged properly. Overwritten by older version descriptions.
myungjoo
reviewed
Mar 28, 2024
There was a problem hiding this comment.
Please remove this empty file.
myungjoo
reviewed
Mar 28, 2024
|
|
||
| third_party/package_tflite/ | ||
| third_party/build_tflite/ | ||
| third_party/embedded_root/ |
myungjoo
reviewed
Mar 28, 2024
There was a problem hiding this comment.
Please don't remove these files. We have flatbuf/protobuf users.
myungjoo
reviewed
Mar 28, 2024
| @@ -681,6 +681,7 @@ gst_join_request_new_pad (GstElement * element, GstPadTemplate * templ, | |||
| return sinkpad; | |||
| } | |||
|
|
|||
|
|
|||
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.
One might want to dynamically change the framerate at any time while the pipeline is running. This is prevented by code in
tensor_converterandtensor_filterthat, upon reconfigure events, require the new caps to be 'compatible' with the original caps on the sink pads. Requiring the caps to have matching framerates seems unnecessary and too restrictive.This patch removes that restriction from both
tensor_converterandtensor_filterelements.A new unit test for this case has been added.