from sentence_transformers import SentenceTransformer, CrossEncoder, util
cross_encoder = CrossEncoder("./m3/model")
corpus = ["西安","太原","北京","海南"]
top_res=cross_encoder.rank(query="陕西省的省会城市是哪个", documents=corpus, top_k=3, return_documents=True)
print(top_res)
Some weights of XLMRobertaForSequenceClassification were not initialized from the model checkpoint at ./m3/model and are newly initialized: ['classifier.dense.bias', 'classifier.dense.weight', 'classifier.out_proj.bias', 'classifier.out_proj.weight']
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
[{'corpus_id': 0, 'score': 0.5507802, 'text': '西安'}, {'corpus_id': 2, 'score': 0.54930556, 'text': '北京'}, {'corpus_id': 1, 'score': 0.5357058, 'text': '太原'}]
[{'corpus_id': 1, 'score': 0.5277521, 'text': '太原'}, {'corpus_id': 2, 'score': 0.5160099, 'text': '北京'}, {'corpus_id': 0, 'score': 0.4968127, 'text': '西安'}]