How do I run inferences with the models in this repo?

[ianhua2022]

i.e. is there any instruction to use the mobile benchmark?

[freedomtan]

it depends types of models:

• for .tflite: you can start from https://www.tensorflow.org/lite/performance/measurement
• for .dlc: please check Qualcomm's https://developer.qualcomm.com/software/qualcomm-neural-processing-sdk
• for .nnc: please Samsung's https://developer.samsung.com/neural/overview.html

[ianhua2022]

@freedomtan Thanks for the pointers but I mean if there are any instructions to run the MLPerf Mobile benchmark rather than model's frameworks?

[anhappdev]

@ianhua2022 Please take a look at this repo https://github.com/mlcommons/mobile_app_open/

[anhappdev]

If you still have related questions, please reopen this issue.