使用方法

from sentence_transformers import SentenceTransformer

sentences = ["sentence1", "sentence2"]
model = SentenceTransformer('IYun-large-zh')
embeddings_1 = model.encode(sentences, normalize_embeddings=True)
embeddings_2 = model.encode(sentences, normalize_embeddings=True)
similarity = embeddings_1 @ embeddings_2.T
print(similarity)

Downloads last month: 0

Unable to determine this model's library. Check the docs .

Evaluation results

cos_sim_pearson on MTEB AFQMC
validation set self-reported

57.035
cos_sim_spearman on MTEB AFQMC
validation set self-reported

61.057
euclidean_pearson on MTEB AFQMC
validation set self-reported

59.929
euclidean_spearman on MTEB AFQMC
validation set self-reported

61.057
manhattan_pearson on MTEB AFQMC
validation set self-reported

59.911
manhattan_spearman on MTEB AFQMC
validation set self-reported

61.019
cos_sim_pearson on MTEB ATEC
test set self-reported

56.815
cos_sim_spearman on MTEB ATEC
test set self-reported

59.017
euclidean_pearson on MTEB ATEC
test set self-reported

63.444
euclidean_spearman on MTEB ATEC
test set self-reported

59.017
manhattan_pearson on MTEB ATEC
test set self-reported

63.417
manhattan_spearman on MTEB ATEC
test set self-reported

59.000
accuracy on MTEB AmazonReviewsClassification (zh)
test set self-reported

49.280
f1 on MTEB AmazonReviewsClassification (zh)
test set self-reported

46.844
cos_sim_pearson on MTEB BQ
test set self-reported

71.060
cos_sim_spearman on MTEB BQ
test set self-reported

72.631
euclidean_pearson on MTEB BQ
test set self-reported

71.339
euclidean_spearman on MTEB BQ
test set self-reported

72.631
manhattan_pearson on MTEB BQ
test set self-reported

71.315
manhattan_spearman on MTEB BQ
test set self-reported

72.609
v_measure on MTEB CLSClusteringP2P
test set self-reported

55.116
v_measure on MTEB CLSClusteringS2S
test set self-reported

45.056
map on MTEB CMedQAv1
test set self-reported

88.886
mrr on MTEB CMedQAv1
test set self-reported

90.941
map on MTEB CMedQAv2
test set self-reported

89.982
mrr on MTEB CMedQAv2
test set self-reported

92.061
map_at_1 on MTEB CmedqaRetrieval
self-reported

26.990
map_at_10 on MTEB CmedqaRetrieval
self-reported

40.187
map_at_100 on MTEB CmedqaRetrieval
self-reported

42.057
map_at_1000 on MTEB CmedqaRetrieval
self-reported

42.156
map_at_3 on MTEB CmedqaRetrieval
self-reported

35.704
map_at_5 on MTEB CmedqaRetrieval
self-reported

38.307
mrr_at_1 on MTEB CmedqaRetrieval
self-reported

40.835
mrr_at_10 on MTEB CmedqaRetrieval
self-reported

49.207
mrr_at_100 on MTEB CmedqaRetrieval
self-reported

50.164
mrr_at_1000 on MTEB CmedqaRetrieval
self-reported

50.200
mrr_at_3 on MTEB CmedqaRetrieval
self-reported

46.649
mrr_at_5 on MTEB CmedqaRetrieval
self-reported

48.082
ndcg_at_1 on MTEB CmedqaRetrieval
self-reported

40.835
ndcg_at_10 on MTEB CmedqaRetrieval
self-reported

46.976
ndcg_at_100 on MTEB CmedqaRetrieval
self-reported

54.162
ndcg_at_1000 on MTEB CmedqaRetrieval
self-reported

55.840
ndcg_at_3 on MTEB CmedqaRetrieval
self-reported

41.417
ndcg_at_5 on MTEB CmedqaRetrieval
self-reported

43.865
precision_at_1 on MTEB CmedqaRetrieval
self-reported

40.835
precision_at_10 on MTEB CmedqaRetrieval
self-reported

10.403
precision_at_100 on MTEB CmedqaRetrieval
self-reported

1.622
precision_at_1000 on MTEB CmedqaRetrieval
self-reported

0.184
precision_at_3 on MTEB CmedqaRetrieval
self-reported

23.473
precision_at_5 on MTEB CmedqaRetrieval
self-reported

17.094
recall_at_1 on MTEB CmedqaRetrieval
self-reported

26.990
recall_at_10 on MTEB CmedqaRetrieval
self-reported

57.949
recall_at_100 on MTEB CmedqaRetrieval
self-reported

87.578
recall_at_1000 on MTEB CmedqaRetrieval
self-reported

98.741
recall_at_3 on MTEB CmedqaRetrieval
self-reported

41.244
recall_at_5 on MTEB CmedqaRetrieval
self-reported

48.727
cos_sim_accuracy on MTEB Cmnli
validation set self-reported

85.075
cos_sim_ap on MTEB Cmnli
validation set self-reported

92.050
cos_sim_f1 on MTEB Cmnli
validation set self-reported

85.864
cos_sim_precision on MTEB Cmnli
validation set self-reported

82.000
cos_sim_recall on MTEB Cmnli
validation set self-reported

90.110
dot_accuracy on MTEB Cmnli
validation set self-reported

85.075
dot_ap on MTEB Cmnli
validation set self-reported

92.056
dot_f1 on MTEB Cmnli
validation set self-reported

85.864
dot_precision on MTEB Cmnli
validation set self-reported

82.000
dot_recall on MTEB Cmnli
validation set self-reported

90.110
euclidean_accuracy on MTEB Cmnli
validation set self-reported

85.075
euclidean_ap on MTEB Cmnli
validation set self-reported

92.050
euclidean_f1 on MTEB Cmnli
validation set self-reported

85.864
euclidean_precision on MTEB Cmnli
validation set self-reported

82.000
euclidean_recall on MTEB Cmnli
validation set self-reported

90.110
manhattan_accuracy on MTEB Cmnli
validation set self-reported

85.135
manhattan_ap on MTEB Cmnli
validation set self-reported

92.029
manhattan_f1 on MTEB Cmnli
validation set self-reported

85.877
manhattan_precision on MTEB Cmnli
validation set self-reported

82.297
manhattan_recall on MTEB Cmnli
validation set self-reported

89.783
max_accuracy on MTEB Cmnli
validation set self-reported

85.135
max_ap on MTEB Cmnli
validation set self-reported

92.056
max_f1 on MTEB Cmnli
validation set self-reported

85.877

View on Papers With Code