Feature/sg 747 support predict video full pipeline master #829
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
…dd_full_pipeline_with_preprocessing
Louis-Dupont
commented
Apr 10, 2023
shaydeci
suggested changes
Apr 10, 2023
src/super_gradients/examples/predict/detection_predict_image_folder.py
Outdated
Show resolved
Hide resolved
src/super_gradients/training/models/detection_models/customizable_detector.py
Outdated
Show resolved
Hide resolved
src/super_gradients/training/models/detection_models/customizable_detector.py
Outdated
Show resolved
Hide resolved
|
Regarding processing video/image folder, I wonder whether we can extract common logic via generators: So processing logic for a collection of images is in one place, regardless what if it is an image folder or video. So now even if we want to process and save entire LoTR movie we can do this without storing all frames in memory: |
Louis-Dupont
commented
Apr 11, 2023
BloodAxe
reviewed
Apr 11, 2023
shaydeci
suggested changes
Apr 16, 2023
Louis-Dupont
deleted the
feature/SG-747-support_predict_video_full_pipeline_master
branch
Louis-Dupont
added a commit
that referenced
this pull request
Apr 17, 2023
* wip * move to imageprocessors * wip * add back changes * making it work fully for yolox and almost for ppyoloe * minor change * working for det * cleaning * clean * undo * replace empty with none * add _get_shift_params * minor doc change * cleaning wip * working for multiple images * add ppyoloe * replace pydantic with dataclasses and fix typing * add docstrings * doc improvment and use get_shift_params in transforms * add tests * improve comment * rename * wip * add option to keep ratio in rescale * make functions private * remove DetectionPaddedRescale * fix doc * big commit with wrong things * try undo bad change * doc * minor doc * add a lot of doc * fix comment * minor change * first draft of load_video * adding save_video, some parts are still to be checked * wip * add __init__.py to pipelines * replace size with shape * wip * cleaning * wip * fix rgb to bgr and remove check * almost working, missing batch * proposal of predict_video * wip working on dete * add yolox * add flag to visualize * update * add streaming * improve streaming code * docstring update * fix stream example * rename Results * cleaning * rename stream to predict_webcam * doc fixes * improve docstring and homogenize some names * rename _images_prediction_lst * improve doc * add doc * minore change * fix image * fix ci * fix merge * reverse channel properly
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.
Adding support for image folder and video as part of the base
predictfunction.Also adding support for
predict_webcam.This also introduced some refactoring on the Pipeline to support lazy batch processing.