search

pannous / tensorflow-speech-recognition

🎙Speech recognition using the tensorflow deep learning framework, sequence-to-sequence neural networks
Other
2.17k stars 638 forks source link

What should dtype of placeholder y_ in training be? #4

Closed codegeniur closed 8 years ago

codegeniur commented 8 years ago

From speech_encoder.py, batch_xs, batch_ys = speech.train.next_batch(100) batch_xs=[flatten(matrix) for matrix in batch_xs] feed = {x: batch_xs, y_: batch_ys}

The above has the following error:

ValueError: invalid literal for float(): 2 14 68 6 32 14 73 6 47 14 73 3

What should placeholder of y_ be?

Thanks!

pannous commented 8 years ago

bug fixed