You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi @vedantpuri, I would like to work on this issue. For predicting emojis on the go we can save the weights and load up the model to efficiently predict emojis for a sentence.
I am not much adept with text preprocessing. I wanted to know how would we go about converting sentences to vectors so that they can be passed to model for prediction?
Thanks for your interest in this project. I would not be very available to work on this for a while but would be happy to see you contribute to it!
For converting sentences to vectors there are a variety of ways. I'd suggest looking up word2vec or glove or whatever embedding methodology used in this library and start from there. I feel a simple concatenation of word vectors should be a good representation of the embeddings of a sentence!
Make UI so that can predict an emoji for a sentence.
Involve other emojis as well for training
The text was updated successfully, but these errors were encountered: