Embedding Models
Collection
Some embedding models I've trained, finetuned, distilled, converted, or something else entirely • 15 items • Updated
A modified version of Snowflake/snowflake-arctic-embed-m-v2.0, without xformers, so it works on CPU.
from sentence_transformers import SentenceTransformer
import torch
device = torch.device("cpu")
model = SentenceTransformer("cnmoro/snowflake-arctic-embed-m-v2.0-cpu", device=device, trust_remote_code=True)
- Downloads last month
- 685
Model tree for cnmoro/snowflake-arctic-embed-m-v2.0-cpu
Spaces using cnmoro/snowflake-arctic-embed-m-v2.0-cpu 3
Collection including cnmoro/snowflake-arctic-embed-m-v2.0-cpu
Evaluation results
- accuracy on MTEB AmazonCounterfactualClassification (en-ext)test set self-reported66.687
- f1 on MTEB AmazonCounterfactualClassification (en-ext)test set self-reported55.037
- f1_weighted on MTEB AmazonCounterfactualClassification (en-ext)test set self-reported73.074
- ap on MTEB AmazonCounterfactualClassification (en-ext)test set self-reported18.077
- ap_weighted on MTEB AmazonCounterfactualClassification (en-ext)test set self-reported18.077
- main_score on MTEB AmazonCounterfactualClassification (en-ext)test set self-reported66.687
- accuracy on MTEB AmazonCounterfactualClassification (en)test set self-reported66.194
- f1 on MTEB AmazonCounterfactualClassification (en)test set self-reported60.854
- f1_weighted on MTEB AmazonCounterfactualClassification (en)test set self-reported69.573
- ap on MTEB AmazonCounterfactualClassification (en)test set self-reported30.279
- ap_weighted on MTEB AmazonCounterfactualClassification (en)test set self-reported30.279
- main_score on MTEB AmazonCounterfactualClassification (en)test set self-reported66.194
- accuracy on MTEB AmazonPolarityClassification (default)test set self-reported70.359
- f1 on MTEB AmazonPolarityClassification (default)test set self-reported70.041
- f1_weighted on MTEB AmazonPolarityClassification (default)test set self-reported70.041
- ap on MTEB AmazonPolarityClassification (default)test set self-reported64.819
- ap_weighted on MTEB AmazonPolarityClassification (default)test set self-reported64.819
- main_score on MTEB AmazonPolarityClassification (default)test set self-reported70.359
- accuracy on MTEB AmazonReviewsClassification (en)test set self-reported33.766
- f1 on MTEB AmazonReviewsClassification (en)test set self-reported33.366