feat: add new benchmark Stable Diffusion

[anhappdev]

Relevant files can be downloaded from here:
https://drive.google.com/drive/folders/10zCF7_ctUIM7xVPyhw5P60utnvWVUcSy?usp=sharing

[github-actions]

MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅

[anhappdev]

@RSMNYS You should be able to integrate the Stable Diffusion pipeline for mobile_back_tflite based on this branch.
I will implement the download of multiple models in a separated PR. But for now, you should be able to just use a local folder path instead of a file path, (more here) e.g.:

model_path: "local:///path/to/folder"

then the model_path will be a folder path in:

mlperf_backend_ptr_t mlperf_backend_create(const char* model_path, mlperf_backend_configuration_t* configs, const char* native_lib_path)

[freedomtan]

my old resize() with TFLite resize op (https://github.com/tensorflow/tensorflow/blob/2ca3e974e350664c0e18b2ff5a52867490a15222/tensorflow/lite/examples/label_image/bitmap_helpers_impl.h#L30)

[freedomtan]

@freedomtan to check this implementation

[RSMNYS]

@RSMNYS You should be able to integrate the Stable Diffusion pipeline for mobile_back_tflite based on this branch. I will implement the download of multiple models in a separated PR. But for now, you should be able to just use a local folder path instead of a file path, (more here) e.g.:

model_path: "local:///path/to/folder"

then the model_path will be a folder path in:

mlperf_backend_ptr_t mlperf_backend_create(const char* model_path, mlperf_backend_configuration_t* configs, const char* native_lib_path)

Hi guys! I've pushed the code partially here:

https://github.com/mlcommons/mobile_app_open/tree/features/pipeline_backend

The idea was to create the pipeline (stable_diffusion_backend.cc). And initiate it from the dart-run-benchmark.cc, by checking the dataset type for example. Then in issueQuery we can process the stable diffusion steps.

@anhappdev @freedomtan

[anhappdev]

@RSMNYS I refactored the TFLite backend to support multiple pipelines.
You can add your implementation to mobile_back_tflite/cpp/backend_tflite/stable_diffusion_pipeline.cc.
Please remember to update init_pipeline() in tflite_c.cc.

[sonarqubecloud]

Quality Gate passed

Issues
54 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarCloud

[anhappdev]

We can merge this PR into master now. I disabled the tests for stable_diffusion task, so we have it passed the CI when merging into master and will enable it again in the submission branch.

[freedomtan]

LGTM. @mohitmundhragithub if you are OK with this. Let's merge it.