Identify the language of a given word. Used a character level RNN to predict the language. Each character is first embedded into a vector space then the sequence of vectors is passed into two Gated Recurrent Units (GRU) [1]. Finally a fully connected neural retwork with softmax activation function predicts the language.
run the following command in terminal:
python predict.py 'thanks'
[1]: Cho, Kyunghyun, Bart Van Merriënboer, Dzmitry Bahdanau, and Yoshua Bengio. "On the properties of neural machine translation: Encoder-decoder approaches." arXiv preprint arXiv:1409.1259 (2014).