Henry

您好，想请问一下时间层面比同等的resnet模型，额外增加了多少时间呢？

我用MJ数据集也没有训起来，报错：`assert len(datasets) > 0, 'datasets should not be an empty iterable' `，这是因为什么呢

关于量化训练和推理问题

1

苏老师，向您请教一个问题。 因为我这边用模型是在芯片上面，所以所用到的模型都要进行量化操作，并且不是后量化，是先把模型中的算子修改成我们自己基于pytorch or tensorflow所开发的训练插件中的同等算子，训8bit，4bit这样子。然后我看您代码里面有重写Embedding等等层，如果我把这些层替换成keras原始的层，会不会对代码运行或者推理效果有影响呢？

### Describe the bug Traceback (most recent call last): File "TTS/bin/compute_attention_masks.py", line 14, in from TTS.tts.utils.text.characters import make_symbols, phonemes, symbols ImportError: cannot import name 'make_symbols' from 'TTS.tts.utils.text.characters' ### To Reproduce...

### Describe the bug I've checked the history issues about speedyspeech, can not find the same question. In the past days, i have trained the speedyspeech model on ljspeech with...

DDP training question

2

Hi, I'm using the tutorial [https://github.com/pytorch/tutorials/blob/master/intermediate_source/ddp_tutorial.rst](url) for DDP train,using 4 gpus in myself code, reference Basic Use Case. But when I finished the modification, it was stuck during run the...

Code: lenet.fit(x_train, y_train, batch_size=64, epochs=50, validation_data=[x_test, y_test]) there is not list,so you should change to tuple as: lenet.fit(x_train, y_train, batch_size=64, epochs=50, validation_data=(x_test, y_test))