search

ModelTC / MQBench

Model Quantization Benchmark
Apache License 2.0
769 stars 140 forks source link

About the quantization of recurrent neural network! #241

Closed larenzhang closed 1 year ago

larenzhang commented 1 year ago

It is a nice work to benchmark quantization methods for various CNN architectures. Recurrent neural network is another mainstrain architecture, which is widly used as a times series model in edge-device. Quantization can also be used to recurrent neural network, such as LSTM, GRU and etc. I am curious that is there any plan to benchmark the quantization of recurrent neural network.

Look forward to your response. Best wishes!

wimh966 commented 1 year ago

Hi, we have not planned to support RNNs. But we plan to support transformer models, which are also popular in NLP field.

github-actions[bot] commented 1 year ago

This issue has not received any updates in 120 days. Please reply to this issue if this still unresolved!