XiaoSX comments

Results 6 comments of


                                            XiaoSX

运行时出现在from text import keras_detect as detect中，tensorflow.python.framework.errors_impl.InvalidArgumentError: You must feed a value for placeholder tensor 'Placeholder_395' with dtype float and shape [2]，是不是版本的问题

tf.concat also works with tensorflow 1.14

EfficientNet backbone

@SahadevPoudel Hi, I want to try the EfficientNet backbone, where could I find your code?

Only noise be generated

@devinroth hi, can i have some advise from you, i do the 99999 steps , but could not hear the clear sound.

Only noise be generated

@devinroth ok, here is the result file. [generated2_wav.zip](https://github.com/ibab/tensorflow-wavenet/files/1090402/generated2_wav.zip)

Which layers of the model are finetuned?

I am inspecting the backend.h5 file, found 249 variables of yolo inside the h5 file. only 10 variables are missed, they are ['conv_81', 'input_3', 'input_2', 'conv_93', 'input_4', 'conv_105', 'input_5', 'yolo_layer_1',...