Changing to the larger model again to achieve a higher performance, because the smaller model did not provide enough accuracy

We have gone back to the previous model that was being used (sentence similarity spanish) because the one added in the previous commit was too large and it lasted a lot of time to extract the features from a PDF document

We have changed the embeddings' model that was being used with one

We have modified the implementation to return the dataframe with

This new implementation allows to obtain the embeddings of

We have included a new model that copes with sentences in Spanish.

We have included the real LLM model that will extract the

Changing a typo in the Dockerfile because the application was not running (using ' instead of ").

We have updated the Dockerfile because it was incomplete and the application was not starting.

We have modified the port of the application to make it run

First upload to the repository.

initial commit