Cedric Chee

The steps to download libcaffe2.a and other library files are in the [README](https://github.com/cedrickchee/pytorch-android#documentation). Scroll down a bit and you'll see: > _If you prefer not to build PyTorch from master,...

> would it work seamlessly after replacing the _init_net.pb and _predict_net.pb or do I need to make other modifications? It should. MobileNet v2 (or SqueezeNet) input (data)/output (prediction) tensor dimension...

Unless I understand it incorrectly, isn't MobileNet/ResNet18 a classification model?

My tutorials used image classification models. It _should_ not require any modifications to the code if you put aside and do not count my previous reply where the MobileNet v2...

The steps to download libcaffe2.a and other library files are in the [README](https://github.com/cedrickchee/pytorch-android#documentation). > If you prefer not to build PyTorch from master, I have committed the files for the...

What is the Android NDK version? I have not tested with CMake 3.10.2. I'm using NDK r18 and CMake 3.6. I'm not too sure what caused this error.

I have tried running the quantized MobileNet v2 from your link. According to the docs there, I have ensure my Caffe2 has QNNPACK based Int8 operators before running inference on...

What is the training speed looks like for you? I mean, how many minutes per epoch and what GPU are you using? I will try to optimize the code soon....

Sorry for the terribly late response. If I understand your intention correctly, you are trying to do inference using the model. Unfortunately, the code has no *proper* support for inference...

Hi @arshiyaaggarwal Thank you for reporting this. My plate is quite full at the moment. I will try my best to look into this tomorrow. I don't fully understand the...