I am currently using Universal Sentence Encoder to embed certain sentences which I would then feed to a deep learning model to do some prediction, but just to test whether the universal sentence encoder is encoding my text data effectively for my context I need to do a semantic similarity test on the embeddings done by universal sentence encoder, an example can be found at :refer this link, can anyone please help me with how to get the numerical values of the cosine similarity(given under the build evaluation graph section of the given link).

I am a beginner to the tf and tf-hub, pardon me if I asked a very doubt.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.