KennethEnevoldsen / scandinavian-embedding-benchmark

A Scandinavian Benchmark for sentence embeddings
https://kennethenevoldsen.github.io/scandinavian-embedding-benchmark/
MIT License
27 stars 3 forks source link

Ensure consistent names #153

Closed KennethEnevoldsen closed 8 months ago

KennethEnevoldsen commented 8 months ago

Now, all the names are as they appear on the results table. This means that running e.g. sentence-transformers/paraphrase-multilingual-mpnet-base-v2 will run the model from HF, while paraphrase-multilingual-mpnet-base-v2 runs the registered model. Another choice would be to rename the model on the table with their full name.

Also added to method for moving encoders to device.

Fixes #150 Fixed #147 (partly, decided on not adding it to sebmodel, but only to the encoder)