https://github.com/vvkmnn/rnndl
⚜️ A recurrent neural network that translates English and French.
https://github.com/vvkmnn/rnndl
deep-learning recurrent-neural-networks tensorflow translation
Last synced: about 2 months ago
JSON representation
⚜️ A recurrent neural network that translates English and French.
- Host: GitHub
- URL: https://github.com/vvkmnn/rnndl
- Owner: Vvkmnn
- Created: 2017-09-09T17:41:11.000Z (almost 9 years ago)
- Default Branch: master
- Last Pushed: 2017-09-11T01:03:15.000Z (almost 9 years ago)
- Last Synced: 2025-10-10T14:21:18.284Z (9 months ago)
- Topics: deep-learning, recurrent-neural-networks, tensorflow, translation
- Language: Jupyter Notebook
- Homepage:
- Size: 4.84 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
## babelfishRNN
The fourth project in the [Deep Learning Foundations Nanodegree program](https://www.udacity.com/course/deep-learning-nanodegree-foundation--nd101), further refining the nuances of Recurrent Neural Networks, but this time, focused on **translating between two languages**.
Here, we **translate English to French** using **Word Embedding, Encoding, Decoding**, and **Interference**. The resulting network achieved 0.0177 loss, but on a relatively small dataset 227 English words. The details of the methodology are in `babelfishRNN`**,
### Setup
This project requires Python 3 (Probably as distributed by [Anaconda](https://www.continuum.io/downloads)) and [Tensorflow](https://www.tensorflow.org/):
```python
appnope==0.1.0
autopep8==1.3.2
backports.weakref==1.0rc1
bleach==1.5.0
decorator==4.1.2
html5lib==0.9999999
ipykernel==4.6.1
ipython==6.1.0
ipython-genutils==0.2.0
jedi==0.10.2
jupyter-client==5.1.0
jupyter-core==4.3.0
Markdown==2.6.8
numpy==1.13.1
pexpect==4.2.1
pickleshare==0.7.4
prompt-toolkit==1.0.15
protobuf==3.3.0
ptyprocess==0.5.2
pycodestyle==2.3.1
Pygments==2.2.0
python-dateutil==2.6.1
pyzmq==16.0.2
simplegeneric==0.8.1
six==1.10.0
tensorflow==1.0.0
tornado==4.5.1
traitlets==4.3.2
wcwidth==0.1.7
Werkzeug==0.12.2
```