QAT for LSTM

[schmiph2]

Hi everyone

I saw in your documentation that you support post training quantization of LSTMs by replacing the PyTorch modules by your own implementation. Do you also support quantization aware training of LSTMs with this technique? Do you already have a tutorial for this?

Thanks a lot for your help!