* initial checkins for mllama image process
* fix some tests
* some fixings
* add more image
* More test assertions
* parity test passed
* code clean up
* code refinement
* Remove OpenCV dependency from C_API model
* fix build on Windows
* switch ci build flag
* try to fix the macOS build issue
* more fixing
* fix the macOS build issue
* list jpeg source
* verified on MacOS
* update the pp_api too
* avoid the codecs library conflicts
* Add the unit tests
* move the codec test
* add the missing dl lib for extensions test
* refine the code
* a smaller fixing for Windows Python
* optimize the tokenizer for efficiency
* fix the unit test failures.
* fix the api test case failures
* removed the unused code.
* More test cases fixings
* One more fixing
* fix macOS build issues
* refine the test
* add more diagnosis info.
* fix unit test in CI Linux
* fix the pp_api test failure
* initial api for tokenizer
* More fixings and test data refinement
* add a simple wrapper for pre-processing APIs
* fix the test issues
* test if the tokenizer is spm based
* fix the failed test cases
* json pointer does not work
* reimplement resize cpu kernel for image processing
* accuracy fixing and code refinement
* fix the build issues
* fix Linux build issue
* more fixings
* Fix the pipeline issue
* fix the ci script
* try to fix CUDA machine pool
* Feature extraction C API for whipser model
* Update the docs
* Update the docs2
* refine the code
* fix some issues
* fix the Linux build
* fix more data consistency issue
* More code refinements
This commit updates `HFTokenizerConverter` to handle cases where the `hf_tokenizer` object might not have a `vocab_file` attribute.
Changes:
* Uses `getattr` to retrieve the `vocab_file` attribute for flexibility
* Stores the retrieved value in a separate variable `vocab_file` for clarity
* Checks if `vocab_file` is `None` before checking its existence
This ensures the converter works correctly even with tokenizers that don't define a `vocab_file` attribute.
* Unify the spm/bpe tokenizers
* fix the build error
* fix the decoding issue
* add model name in exported onnx
* fixing the unit tests
* revert the unneccesary file format changes
* Add support for YOLO v8 Pose post-processing.
The output has additional values in the 'mask' data for the keypoints.
- Update the post processing steps to support extracting and scaling the keypoints.
- Simplify the existing step to split out the boxes and scores by using a basic Split operator if there is no confidence score for a bounding box to apply to the class scores.
- Confidence score for a bounding box is YOLO versions prior to 8.
- Update existing tests
TODO: Add unit tests for new Steps. They have been manually validated with the real model for now.
* Changes to support pre-decoded input.
Needs cleanup.
* Support an overall max number of detections as well as per-class detections.
* Expand Identity to support multiple inputs
Fix issue with incorrect score being selected by NMS (was max and not score for selected clas)
Fix TopK usage so result ordering is consistent when it is not used
Add unit tests.
* Update docs and some cleanups
* Use Union
* Fix CodeGenTokenizer issues and the related code refactoring.
* refactor the trie-tree
* temp check-ins
* code complete
* correctness fixing
* Update _hf_cvt.py
* more test cases fixing
* more refinement
* linux crash fixing
* Update test_autotokenizer.py
* Fix GPT2 and Falcon tokenizer cvt for AutoTokenizer imp
* fix fast tokenizer issue
* small fix
* use slow tokenizer in test script
---------
Co-authored-by: Sayan Shaw <sayanshaw@microsoft.com>
* Add TrieTokenizer for RWKV-like LLM models
* add more tests
* fix the windows build
* downloading file instead of check in the vocab file
* a small bug fixing
* Add Falcon-7b and Falcon-40b tokenizer support
* fix alignment and add tokenizer file in test/data to speed up compute
---------
Co-authored-by: Sayan Shaw <sayanshaw@microsoft.com>
* initial checkins
* test pass
* basic impl
* first unit test pass
* merge error
* refine a little bit
* add more unit test
* fix unit test
* Fix the unit test.
* add one more whisper audiodecoder test case
* update the docs
* More updates
* Calculate and specify ir_version so we use the oldest possible for maximum compatibility
* Don't use `ignore_unknown` in call to `find_min_ir_version_for` as it's only supported in the most recent ONNX release.
* initial draft
* second
* third
* polishing
* fix the M_PI name in LINUX platform
* fix bessel function issue
* add a unit test case
* fix the unit test name
* object detection
* Unit test
add e2e fastestdet model test
---------
Co-authored-by: Changming Sun <chasun@microsoft.com>
Co-authored-by: Scott McKay <skottmckay@gmail.com>
* built-in bounding box op
* update boundary check
* assert policy
* more boundary test and check
* XYXY--> X horizon
---------
Co-authored-by: Scott McKay <skottmckay@gmail.com>
* evaluate the audio decoder library
* MP3 Decoder
* rename it to test_audio_codec
* add the audio decoder to whisper model
* whisper end-to-end draft
* fix the mp3 decoder
* Running with ONNX models
* Add more audio format supports
* refine the end-to-end script
* Update operators/audio/audio_decoder.hpp
Co-authored-by: Edward Chen <18449977+edgchen1@users.noreply.github.com>
* Update operators/audio/audio_decoder.hpp
Co-authored-by: Edward Chen <18449977+edgchen1@users.noreply.github.com>
* Update operators/audio/audio_decoder.hpp
Co-authored-by: Edward Chen <18449977+edgchen1@users.noreply.github.com>
* some fixings of comments and more test cases.
* changes for review comments.
* Update audio_decoder.hpp
* Update audio_decoder.hpp
* code refinement
* Update operators/audio/audio_decoder.hpp
Co-authored-by: Edward Chen <18449977+edgchen1@users.noreply.github.com>
---------
Co-authored-by: Sayan Shaw <52221015+sayanshaw24@users.noreply.github.com>
Co-authored-by: Edward Chen <18449977+edgchen1@users.noreply.github.com>
* add a stft-norm custom op for log-mel spectrum.
* undo the debug change
* Support ONNX standard STFT op signature.
* Add a unit test onnx STFT compatible mode.
* add whisper pre-/post- processing example
* Update dlib.cmake
* undo test code changes
* Update setup.cfg
* update the end2end example with STFT op
Wenbing Li
cao lei
Kaz Nishimura
Scott McKay
Rachel Guo
Danylo Ulianych
Sayan Shaw
Sam Webster
JiCheng
Vishal Jain