update

README.md

@@ -14,12 +14,37 @@ ko-reranker는 [BAAI/bge-reranker-larger](https://huggingface.co/BAAI/bge-rerank
 
 - - -
 
-## 0. Usage
+## 0. Features
 - #### <span style="#FF69B4;"> Reranker는 임베딩 모델과 달리 질문과 문서를 입력으로 사용하며 임베딩 대신 유사도를 직접 출력합니다.</span>
 - #### <span style="#FF69B4;"> Reranker에 질문과 구절을 입력하면 연관성 점수를 얻을 수 있습니다.</span>
 - #### <span style="#FF69B4;"> Reranker는 CrossEntropy loss를 기반으로 최적화되므로 관련성 점수가 특정 범위에 국한되지 않습니다.</span>
 
-## 1. Backgound
+## 1.Usage
+
+- Local
+'''
+    def exp_normalize(x):
+      b = x.max()
+      y = np.exp(x - b)
+      return y / y.sum()
+    
+    from transformers import AutoModelForSequenceClassification, AutoTokenizer
+
+    tokenizer = AutoTokenizer.from_pretrained(model_path)
+    model = AutoModelForSequenceClassification.from_pretrained(model_path)
+    model.eval()
+
+    pairs = [["나는 너를 싫어해", "나는 너를 사랑해"], \
+             ["나는 너를 좋아해", "너에 대한 나의 감정은 사랑 일 수도 있어"]]
+
+    with torch.no_grad():
+        inputs = tokenizer(pairs, padding=True, truncation=True, return_tensors='pt', max_length=512)
+        scores = model(**inputs, return_dict=True).logits.view(-1, ).float()
+        scores = exp_normalize(scores.numpy())
+        print (f'first: {scores[0]}, second: {scores[1]}')
+'''
+
+## 2. Backgound
 - #### <span style="#FF69B4;"> **컨택스트 순서가 정확도에 영향 준다**([Lost in Middel, *Liu et al., 2023*](https://arxiv.org/pdf/2307.03172.pdf)) </span>
 
 - #### <span style="#FF69B4;"> [Reranker 사용해야 하는 이유](https://www.pinecone.io/learn/series/rag/rerankers/)</span>
@@ -31,7 +56,7 @@ ko-reranker는 [BAAI/bge-reranker-larger](https://huggingface.co/BAAI/bge-rerank
 
 - - -
 
-## 2. Reranker models
+## 3. Reranker models
 
 - #### <span style="#FF69B4;"> [Cohere] [Reranker](https://txt.cohere.com/rerank/)</span>
 - #### <span style="#FF69B4;"> [BAAI] [bge-reranker-large](https://huggingface.co/BAAI/bge-reranker-large)</span>
@@ -39,7 +64,7 @@ ko-reranker는 [BAAI/bge-reranker-larger](https://huggingface.co/BAAI/bge-rerank
 
 - - -
 
-## 3. Dataset
+## 4. Dataset
 
 - #### <span style="#FF69B4;"> [msmarco-triplets](https://github.com/microsoft/MSMARCO-Passage-Ranking) </span>
     - (Question, Answer, Negative)-Triplets from MS MARCO Passages dataset, 499,184 samples
@@ -48,7 +73,7 @@ ko-reranker는 [BAAI/bge-reranker-larger](https://huggingface.co/BAAI/bge-rerank
     
 - - -
 
-## 4. Performance
+## 5. Performance
 | Model                     | has-right-in-contexts | mrr (mean reciprocal rank) |
 |:---------------------------|:-----------------:|:--------------------------:|
 | without-reranker (default)| 0.93 | 0.80 |
@@ -76,22 +101,22 @@ ko-reranker는 [BAAI/bge-reranker-larger](https://huggingface.co/BAAI/bge-rerank
 
 - - -
 
-## 5. Acknowledgement
+## 6. Acknowledgement
 - <span style="#FF69B4;"> Part of the code is developed based on [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding/tree/master?tab=readme-ov-file) and [KoSimCSE-SageMaker](https://github.com/daekeun-ml/KoSimCSE-SageMaker/tree/7de6eefef8f1a646c664d0888319d17480a3ebe5).</span>
 
 - - -
 
-## 6. Citation
+## 7. Citation
 - <span style="#FF69B4;"> If you find this repository useful, please consider giving a like ⭐ and citation</span>
 
 - - -
 
-## 7. Contributors:
+## 8. Contributors:
 - <span style="#FF69B4;"> **Dongjin Jang, Ph.D.** (AWS AI/ML Specislist Solutions Architect) | [Mail](mailto:dongjinj@amazon.com) | [Linkedin](https://www.linkedin.com/in/dongjin-jang-kr/) | [Git](https://github.com/dongjin-ml) | </span>
 
 - - -
 
-## 8. License
+## 9. License
 - <span style="#FF69B4;"> FlagEmbedding is licensed under the [MIT License](https://github.com/aws-samples/aws-ai-ml-workshop-kr/blob/master/LICENSE). </span>