fattyyyyyyyyyy

Results 2 issues of fattyyyyyyyyyy

输入的维度是(1, 398, 120),会报batch_sizes!=1 is not supported for lstm in this version的错误。 ![image](https://user-images.githubusercontent.com/73877725/153349893-da9688b1-a80d-4235-aa00-516885ca7f32.png) 如果将输入的维度改成(1, 398, 120, 1),增加一个reshape层,则会报下面的错误。 ![image](https://user-images.githubusercontent.com/73877725/153350302-a7554953-d401-46b5-a0e0-e3d5ce9989ef.png) 也尝试了先转成onnx模型,再转rknn,会报Calc node LSTM : LSTM_29 output shape fail ![image](https://user-images.githubusercontent.com/73877725/153350493-d88d661e-aaa9-4bfc-9eb7-276e2a85cdb3.png) 想问一下是哪里出了问题,如果有需要,pt和onnx模型我放在了自己的仓库里

I tried to train the model, but the average SDR is only about 6. Here are my specific results: