How can i create Word Embeddings files for my train data?

[talatccan]

Hi,

Im trying to train my own data. I prepared source, target, vocab_source, target_source but i stuck on to create embeddings file. How can i do it?

[thushv89]

Hi @talatccan ,

I think I need a more descriptive explanation than that of what you are trying to do.

[talatccan]

Im trying to build nmt model with my train data and i followed your nmt tutorial.

Im in the stage of Inputs Outputs Masks. In there it want embeddings file as following:

# Need to use pre-trained word embeddings
encoder_emb_layer = tf.convert_to_tensor(np.load('de-embeddings.npy'))
decoder_emb_layer = tf.convert_to_tensor(np.load('en-embeddings.npy'))

I couldnt figure out how can i create embeddings file with my own source and target data. How can i do that?