update readme.md

README.md

@@ -21,13 +21,18 @@ tags:
 
 ✨ The output score can be transformed into a [0, 1] range using a sigmoid function, providing a clear and interpretable measure of relevance.
 
+## Arabic RAG Pipeline 
+
+
+![Arabic RAG Pipeline](https://i.ibb.co/z4Fc3Kd/Screenshot-2024-11-28-at-10-17-39-AM.png)
+
 
 
 ## Usage 
 ### Using sentence-transformers
 
 ```
-pip installsentence-transformers
+pip install sentence-transformers
 ```
 ```python
 from sentence_transformers import CrossEncoder
@@ -62,7 +67,31 @@ for i, (candidate, score) in enumerate(ranked_candidates, 1):
     print(f"Score: {score}\n")
 ```
 ## Evaluation 
+### Dataset
+
+Size: 3000 samples.
+
+### Structure:
+🔸 Query: A string representing the user's question.
+
+🔸 Candidate Document: A candidate passage to answer the query.
+
+🔸 Relevance Label: Binary label (1 for relevant, 0 for irrelevant).
+
+### Evaluation Process
+
+🔸 Query Grouping: Queries are grouped to evaluate the model's ability to rank candidate documents correctly for each query.
+
+🔸 Model Prediction: Each model predicts relevance scores for all candidate documents corresponding to a query.
+
+🔸 Metrics Calculation: Metrics are computed to measure how well the model ranks relevant documents higher than irrelevant ones.
 
+| Model                                     | MRR              | MAP              | nDCG@10          |
+|-------------------------------------------|------------------|------------------|------------------|
+| cross-encoder/ms-marco-MiniLM-L-6-v2      | 0.6313333333333334 | 0.6313333333333334 | 0.725444959171438 |
+| cross-encoder/ms-marco-MiniLM-L-12-v2     | 0.6643333333333332 | 0.6643333333333332 | 0.7500407855785803 |
+| BAAI/bge-reranker-v2-m3                   | 0.9023333333333332 | 0.9023333333333332 | 0.9274971489500038 |
+| Omartificial-Intelligence-Space/ARA-Reranker-V1 | 0.9335           | 0.9335           | 0.9507001860964314 |