Skip to content

Latest commit

 

History

History
31 lines (21 loc) · 1 KB

File metadata and controls

31 lines (21 loc) · 1 KB

Text-Translation-using-RNN-and-GRU-

Problem : Text to text translation Approach : Used attention mechanism with teacher forcing for more better results.

To reproduce results:

  1. Download the datasets for language translation from this site.
  2. Set hyperparameters according to your need and run main.py as python main.py

Results:

Following is the results for French to english translation

je ne suis pas si occupe d habitude . = i m usually not this busy .

vous n y etes pas en securite . = you re not safe here .

je suis content de mon travail . = i m satisfied with my work .

tu me fais peur . = you re scaring me .

References:

Sequence to Sequence Learning with Neural Networks
Neural Machine Translation by Jointly Learning to Align and Translate
A Neural Conversational Model
RNN and LSTM in pytorch