bucket iterator with ctc error

[zhangjiulong]

I want to test flexible length input with warpctc
because the rnn is an example of bucket iterator
So I modified the rnn/lstm.py as follows
I modified def lstm_unroll funciton

#sm = mx.sym.SoftmaxOutput(data=pred, label=label, name='softmax')
sm = mx.sym.WarpCTC(data=pred, label=label, label_length = num_label, input_length = seq_len)

But got the following error:

bucket of len  46 : 34 samples
bucket of len  74 : 27 samples
[17:32:46] src/operator/./reshape-inl.h:249: Using target_shape will be deprecated.
Traceback (most recent call last):
  File "lstm_bucketing.py", line 76, in <module>
    batch_end_callback=mx.callback.Speedometer(batch_size, 50),)
  File "../../python/mxnet/model.py", line 746, in fit
    self._init_params(dict(data.provide_data+data.provide_label))
  File "../../python/mxnet/model.py", line 486, in _init_params
    arg_shapes, _, aux_shapes = self.symbol.infer_shape(**input_shapes)
  File "../../python/mxnet/symbol.py", line 453, in infer_shape
    return self._infer_shape_impl(False, *args, **kwargs)
  File "../../python/mxnet/symbol.py", line 513, in _infer_shape_impl
    ctypes.byref(complete)))
  File "../../python/mxnet/base.py", line 77, in check_call
    raise MXNetError(py_str(_LIB.MXGetLastError()))
mxnet.base.MXNetError: InferShape Error in warpctc0's label argument

Please help me to locate the error.

[firestonelib]

Hi, @zhangjiulong I've also met this error, can you tell me how to solve that?

[szha]

This issue is closed due to lack of activity in the last 90 days. Feel free to ping me to reopen if this is still an active issue. Thanks!
Also, do please check out our forum (and Chinese version) for general "how-to" questions.