I'm the new one to ML, currently writing search engine with different models (tried LSTM, BERT, sentence-transformers) to get sentence embeddings, there are about 2k documents, and don't know how to rate perfomance of each model, because output is the range for each query for each model, but I don't have any labels, so I'm a little bit confused. I'm thinking about creating BOW for every document and analyze query on it, but the point of search engine with nueral network is understanding the context, so I don't know how. Will be very grateful for any ideas! Thanks



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.