Sentence Similarity
Transformers
PyTorch
ONNX
sentence-transformers
Arabic
bert
feature-extraction
miniDense
passage-retrieval
knowledge-distillation
middle-training
text-embeddings-inference
Instructions to use prithivida/miniDense_arabic_v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use prithivida/miniDense_arabic_v1 with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("prithivida/miniDense_arabic_v1") model = AutoModel.from_pretrained("prithivida/miniDense_arabic_v1") - sentence-transformers
How to use prithivida/miniDense_arabic_v1 with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("prithivida/miniDense_arabic_v1") sentences = [ "هذا شخص سعيد", "هذا كلب سعيد", "هذا شخص سعيد جدا", "اليوم هو يوم مشمس" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Notebooks
- Google Colab
- Kaggle
update hybrid numbers
Browse files
README.md
CHANGED
|
@@ -174,7 +174,7 @@ The below numbers are with mDPR model, but miniDense_arabic_v1 should give a eve
|
|
| 174 |
|
| 175 |
| Language | ISO | nDCG@10 BM25 | nDCG@10 mDPR | nDCG@10 Hybrid |
|
| 176 |
|-----------|-----|--------------|--------------|----------------|
|
| 177 |
-
| **Arabic** | **
|
| 178 |
|
| 179 |
*Note: MIRACL paper shows a different (higher) value for BM25 Arabic, So we are taking that value from BGE-M3 paper, rest all are form the MIRACL paper.*
|
| 180 |
|
|
|
|
| 174 |
|
| 175 |
| Language | ISO | nDCG@10 BM25 | nDCG@10 mDPR | nDCG@10 Hybrid |
|
| 176 |
|-----------|-----|--------------|--------------|----------------|
|
| 177 |
+
| **Arabic** | **ar** | **0.395** | **0.499** | **0.67.3** |
|
| 178 |
|
| 179 |
*Note: MIRACL paper shows a different (higher) value for BM25 Arabic, So we are taking that value from BGE-M3 paper, rest all are form the MIRACL paper.*
|
| 180 |
|