quic / ai-hub-models

The Qualcomm® AI Hub Models are a collection of state-of-the-art machine learning models optimized for performance (latency, memory etc.) and ready to deploy on Qualcomm® devices.
https://aihub.qualcomm.com
BSD 3-Clause "New" or "Revised" License
438 stars 60 forks source link

[BUG] Improvements for apps/Android/ImageClassification/build_apk.py #38

Closed fangxingSR closed 4 months ago

fangxingSR commented 4 months ago

Describe the bug There is an issue with build_apk.py: when only the qnnsdk and model_path are input, the model selection menu should not be printed using printmenu(), because the model will invariably be assigned the value of 'classification.tflite'.

I have modified the decision rule: the model selection menu will only be printed if neither model_name nor model_path has been provided. It is suggested that line 45 be changed to: if not args.model_name and not args.model_path:

mestrona-3 commented 4 months ago

Hi @fangxingSR, thanks for reporting this issue. We'll look into this and provide an update here.

fangxingSR commented 4 months ago

I created and submitted a branch to https://github.com/quic/ai-hub-models/pull/41 . This submission aims to optimize the logic of the build_apk.py script and update the README.md file. I hope this contribution will bring positive benefits to the community and look forward to it being helpful to everyone.