Omnia Zayed

icon indicating a website link https://www.linkedin.com/in/omnia-zayed-296b3726/

icon company Insight Centre for Data Analytics - National University of Ireland Galway icon location Galway, Ireland icon location PhD Student @Insight_Centre @nuigalway. Passionate about NLP, Biomedical Engineering, Human Cognition and Psychology.

Results 5 comments of Omnia Zayed

Minor issues - but nothing's broken

Hi @thushv89, I didnt really get why did you initialize a new `Input Layer` for the `encoder_inf_states` [here](https://github.com/thushv89/attention_keras/blob/322a16ee147122026b63305aaa5e899d9e5de883/src/examples/nmt/model.py#L56). why didn't you do the attention between the `encoder_inf_out` and the `decoder_inf_out`...

Minor issues - but nothing's broken

@thushv89 It will be great if you can clarify to me what is the role of the `K.rnn` in computing the energy output and the context vector [here ](https://github.com/thushv89/attention_keras/blob/322a16ee147122026b63305aaa5e899d9e5de883/src/layers/attention.py#L101)and [here](https://github.com/thushv89/attention_keras/blob/322a16ee147122026b63305aaa5e899d9e5de883/src/layers/attention.py#L106)....

while training got Error

I face the same error with theano (version 1.0.4) while running different code. @aimaaonline suggestion didn't work. I was able to fix the problem when I passed a list to...

Questions on statistics

> Why are the CleanEval results different for the Java & Python implementations if it's the same algorithm? Yes, as Ivan said, but not only due to the different library...

Semi-deterministic output even though randon_state is set

Hi, I get the same warning. Any ideas on resolving it?