Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

'NoneType' object has no attribute 'get_shape' #98

Open
rishabh-shah opened this issue Jun 16, 2018 · 0 comments
Open

'NoneType' object has no attribute 'get_shape' #98

rishabh-shah opened this issue Jun 16, 2018 · 0 comments

Comments

@rishabh-shah
Copy link

rishabh-shah commented Jun 16, 2018

Hi,

I have been really struggling to get sentence completion work with encoder and decoder logic. After getting introduced to Seq2seq and recurrentshop, I could not imagine it being any easier. I have tried SimpleSeq2Seq() and Seq2Seq() to create a model but not able to get past the following issues,

After installing Seq2seq I used the Seq2seq API as follows,

Seq2Seq(
            input_dim=embedded_features_len,
            input_length=input_seq_length,
            hidden_dim=n_units,
            output_dim=vocab_size,
            output_length=output_seq_length,
            depth=1,
            batch_size = 64,
            dropout=0.2
)

At first I got

'_OptionalInputPlaceHolder' object has no attribute 'inbound_nodes'

error. This is a known issue which was fixed in #91 After I changed "inbound_nodes" to "_inbound_nodes" I get the following error,

'NoneType' object has no attribute 'get_shape'

After some debugging this is what I found,

image

Since the inputs list has 'None' elements, when we call line #487 (engine.py)

_collect_input_shape(inputs)

because we iterate through all elements of inputs, I now see this error and I am not able to get past this. Any help would be really appreciated.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant