Shen Haoyang comments

Results 5 comments of


                                            Shen Haoyang

ValueError: Cannot take the length of shape with unknown rank.

> I got an error message like this, it seems to dynamic input caused. Have you solved this problem?

ValueError: Cannot take the length of shape with unknown rank. during evaluation

我也遇到了相同的问题，请问您解决了吗

Error converting ONNX to TF file: Cannot take the length of shape with unknown rank.

> **Describe the bug** > > Hi, I want to convert an onnx model to Tensorflow, but when I try to convert it, using either the CLI or by script,...

ValueError: Cannot take the length of shape with unknown rank.

Thanks! ________________________________ 发件人: magic_shuang ***@***.***> 发送时间: 2023年8月28日 15:09 收件人: onnx/onnx-tensorflow ***@***.***> 抄送: Shen Haoyang ***@***.***>; Comment ***@***.***> 主题: Re: [onnx/onnx-tensorflow] ValueError: Cannot take the length of shape with unknown rank....

怎么进行多卡训练

> > 请问怎么进行多卡训练 > > 你好，论文中的实验都是单卡训练和测试的。至于多卡训练，可以参考PyTorch教程进行修改。根据我们之前的实验，目前代码中的超参数在多卡训练的情况下可能需要调整，否则精度会偏低一些。您好，我将代码实现了多卡运行（4gpus）总的batchsize为12，请问学习率需要进行相关修改吗，btw我在自己修改超参后与您的结果还是有0.5%的差距，请问您能否提供一些超参的修改建议呢？谢谢