Regression in nightly: TypeError: squeeze() missing 1 required positional argument: 'axis'

[leezu]

Description

squeeze() stopped working recently.

http://ci.mxnet.io/blue/rest/organizations/jenkins/pipelines/GluonNLP-py3-master-gpu-integration/branches/PR-842/runs/13/nodes/82/steps/126/log/?start=0

  File "./scripts/word_embeddings/train_sg_cbow.py", line 324, in <module>
    train(args_)
  File "./scripts/word_embeddings/train_sg_cbow.py", line 235, in train
    loss = embedding(*batch)
  File "/var/lib/jenkins/workspace/gluon-nlp-gpu-py3-master/conda/gpu/py3-master/lib/python3.6/site-packages/mxnet/gluon/block.py", line 548, in __call__
    out = self.forward(*args)
  File "/var/lib/jenkins/workspace/gluon-nlp-gpu-py3-master/conda/gpu/py3-master/lib/python3.6/site-packages/mxnet/gluon/block.py", line 916, in forward
    return self._call_cached_op(x, *args)
  File "/var/lib/jenkins/workspace/gluon-nlp-gpu-py3-master/conda/gpu/py3-master/lib/python3.6/site-packages/mxnet/gluon/block.py", line 805, in _call_cached_op
    self._build_cache(*args)
  File "/var/lib/jenkins/workspace/gluon-nlp-gpu-py3-master/conda/gpu/py3-master/lib/python3.6/site-packages/mxnet/gluon/block.py", line 757, in _build_cache
    data, out = self._get_graph(*args)
  File "/var/lib/jenkins/workspace/gluon-nlp-gpu-py3-master/conda/gpu/py3-master/lib/python3.6/site-packages/mxnet/gluon/block.py", line 749, in _get_graph
    out = self.hybrid_forward(symbol, *grouped_inputs, **params)  # pylint: disable=no-value-for-parameter
  File "/var/lib/jenkins/workspace/gluon-nlp-gpu-py3-master/scripts/word_embeddings/model.py", line 135, in hybrid_forward
    pred_pos = F.batch_dot(emb_center, emb_context).squeeze()
TypeError: squeeze() missing 1 required positional argument: 'axis'

Will add more info shortly

[mxnet-label-bot]

Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so that the appropriate MXNet community members can help resolve it.

[leezu]

@eric-haibin-lin pointed out #14053 is likely at fault
CC: @szha

[szha]

Yes, it was the cause as in symbol the default value is missing for the axis.

[szha]

https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/symbol/symbol.py#L2542

[vrakesh]

@mxnet-label-bot add [Bug]