Different result
#16
by
djstrong
- opened
I the model card you have:
Query: what is snowflake?
0.20051965 The Data Cloud!
0.07660701 Mexico City of Course!
Query: Where can I get the best tacos?
0.24481852 Mexico City of Course!
0.15664819 The Data Cloud!
but I am getting:
Query: what is snowflake?
0.2747492 The Data Cloud!
0.19998044 Mexico City of Course!
Query: Where can I get the best tacos?
0.29974818 Mexico City of Course!
0.2344071 The Data Cloud!
Hello!
Interesting, back when I contributed the integration, it gave the original results:
from sentence_transformers import SentenceTransformer
model = SentenceTransformer("Snowflake/snowflake-arctic-embed-m", revision="ceb801b3fa8ffbf10e2809da664edb3775bcba8b")
queries = ['what is snowflake?', 'Where can I get the best tacos?']
documents = ['The Data Cloud!', 'Mexico City of Course!']
query_embeddings = model.encode(queries, prompt_name="query")
document_embeddings = model.encode(documents)
scores = query_embeddings @ document_embeddings.T
for query, query_scores in zip(queries, scores):
doc_score_pairs = list(zip(documents, query_scores))
doc_score_pairs = sorted(doc_score_pairs, key=lambda x: x[1], reverse=True)
# Output passages & scores
print("Query:", query)
for document, score in doc_score_pairs:
print(score, document)
Since commit ceb801b3fa8ffbf10e2809da664edb3775bcba8b something must have changed.
Oh, #6 updated the model, it was previously incorrect.
@spacemanidol
I'd recommend updating the README outputs to
Query: what is snowflake?
0.2747492 The Data Cloud!
0.19998045 Mexico City of Course!
Query: Where can I get the best tacos?
0.29974818 Mexico City of Course!
0.2344071 The Data Cloud!
to avoid this confusion.
- Tom Aarsen
Thanks for calling out. Fixed it.
spacemanidol
changed discussion status to
closed
Thanks. Results on MTEB are calculated after the correction? I am asking because snowflake models have much worse scores on my dataset in comparison to to other models and MTEB average.