metadata

tags:
  - mteb
model-index:
  - name: alime-reranker-large-zh
    results:
      - task:
          type: Reranking
        dataset:
          type: C-MTEB/CMedQAv1-reranking
          name: MTEB CMedQAv1
          config: default
          split: test
          revision: None
        metrics:
          - type: map
            value: 82.32176162633382
          - type: mrr
            value: 84.91440476190478
      - task:
          type: Reranking
        dataset:
          type: C-MTEB/CMedQAv2-reranking
          name: MTEB CMedQAv2
          config: default
          split: test
          revision: None
        metrics:
          - type: map
            value: 84.08586457179406
          - type: mrr
            value: 86.9011507936508
      - task:
          type: Reranking
        dataset:
          type: C-MTEB/Mmarco-reranking
          name: MTEB MMarcoReranking
          config: default
          split: dev
          revision: None
        metrics:
          - type: map
            value: 35.497382125464284
          - type: mrr
            value: 35.29206349206349
      - task:
          type: Reranking
        dataset:
          type: C-MTEB/T2Reranking
          name: MTEB T2Reranking
          config: default
          split: dev
          revision: None
        metrics:
          - type: map
            value: 68.25849742148222
          - type: mrr
            value: 78.64202157956387

alime-reranker-large-zh

The alime reranker model.

Usage


from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

pairs = [["西湖在哪？", "西湖风景名胜区位于浙江省杭州市"],["今天天气不错","你吓死我了"]]

if torch.cuda.is_available():
    device = torch.device("cuda")
else:
    device = torch.device("cpu")

tokenizer = AutoTokenizer.from_pretrained("Pristinenlp/alime-reranker-large-zh")
model = AutoModelForSequenceClassification.from_pretrained("Pristinenlp/alime-reranker-large-zh").to(device)

inputs = tokenizer(pairs, padding=True, truncation=True, return_tensors='pt', max_length=512).to(device)
scores = model(**inputs, return_dict=True).logits.view(-1, ).float()
print(scores.tolist())