HIGHLIGHTS
- who: Yuejiao Wang and collaborators from the Xi`an Microelectronics Technology Institute, Xi`an, China have published the research work: Sequential Characteristics Based Operators Disassembly Quantization Method for LSTM Layers, in the Journal: (JOURNAL)
- what: The authors design an operators disassembly quantization process for the LSTM layer, so that the neural_network accelerator can support the LSTM layer on the embedded platform. Aiming at this bottleneck, the goal of the proposed quantization algorithm is to determine the quantization parameters of neural_network models with the LSTM layer and deploy them on the NPU. Most importantly, the authors . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.