search

EN10 / Speech-to-Text-WaveNet

Speech to Text
5 stars 1 forks source link

train.py error #3

Closed atanumandal0491 closed 5 years ago

atanumandal0491 commented 5 years ago

I was having a error "ValueError: Shape must be rank 1 but is rank 0 for 'CTCLoss' (op: 'CTCLoss') with input shapes: [?,16,28], [?,2], [?], []."

I just chaged "seq_len = tf.not_equal(x.sg_sum(dims=2), 0.).sg_int().sg_sum(dims=1)" to "seq_len = tf.not_equal(x.sg_sum(axis=2), 0.).sg_int().sg_sum(axis=1)"

It worked