GitHub - sotosd/Bi-dir_RNN_for_Slot_Filling_in_SLU: Bidirectional Recurrent Neural Networks for Slot Filling in Spoken Language Understanding

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commits
AtisData		AtisData
src		src
paper_report.pdf		paper_report.pdf
readMe		readMe
results.txt		results.txt

Repository files navigation

In order to run (train and validate) the model you should run on the following commands on terminal
>Model built using Basic RNN cell
$python pain.py BasicRNN 

>Model built using Basic LSTM cell
$python pain.py LSTM 

After you run one of the previous commands,the model starts training for 40 epochs. The ATIS fold which is currently used is the fold0.

File results.txt, contains the results we obtained during the experiments. 
For each ATIS fold, we compared the 3 models. (These two that we iplemented, plus the Elman Forward model that was given to us)