Sentence Similarity
sentence-transformers
ONNX
Safetensors
Transformers
Transformers.js
English
bert
feature-extraction
text-embeddings-inference
information-retrieval
knowledge-distillation
Instructions to use MongoDB/mdbr-leaf-ir with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use MongoDB/mdbr-leaf-ir with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("MongoDB/mdbr-leaf-ir") sentences = [ "That is a happy person", "That is a happy dog", "That is a very happy person", "Today is a sunny day" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Transformers
How to use MongoDB/mdbr-leaf-ir with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("MongoDB/mdbr-leaf-ir") model = AutoModel.from_pretrained("MongoDB/mdbr-leaf-ir") - Transformers.js
How to use MongoDB/mdbr-leaf-ir with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('sentence-similarity', 'MongoDB/mdbr-leaf-ir'); - Inference
- Notebooks
- Google Colab
- Kaggle
Upload README.md
Browse files
README.md
CHANGED
|
@@ -55,20 +55,20 @@ The table below shows the average BEIR benchmark scores (nDCG@10) for `mdbr-leaf
|
|
| 55 |
|
| 56 |
`mdbr-leaf-ir` ranks #1 on the BEIR public leaderboard, and when run in asymmetric "**(asym.)**" mode as described [here](#asymmetric-retrieval-setup), the results improve even further.
|
| 57 |
|
| 58 |
-
| Model | Size
|
| 59 |
-
|------------------------------------|------|----------------------|
|
| 60 |
-
|
|
| 61 |
-
|
|
| 62 |
-
|
|
| 63 |
-
|
|
| 64 |
-
|
|
| 65 |
-
|
|
| 66 |
-
|
|
| 67 |
-
|
|
| 68 |
-
|
|
| 69 |
-
|
|
| 70 |
-
|
| 71 |
-
|
| 72 |
|
| 73 |
|
| 74 |
# Quickstart
|
|
|
|
| 55 |
|
| 56 |
`mdbr-leaf-ir` ranks #1 on the BEIR public leaderboard, and when run in asymmetric "**(asym.)**" mode as described [here](#asymmetric-retrieval-setup), the results improve even further.
|
| 57 |
|
| 58 |
+
| Model | Size | BEIR Avg. (nDCG@10) |
|
| 59 |
+
|------------------------------------|---------|----------------------|
|
| 60 |
+
| Google text-embedding-005 | Unknown | 55.81 |
|
| 61 |
+
| OpenAI text-embedding-3-large | Unknown | 55.43 |
|
| 62 |
+
| **mdbr-leaf-ir (asym.)** | 23M | **54.03** |
|
| 63 |
+
| **mdbr-leaf-ir** | 23M | **53.55** |
|
| 64 |
+
| snowflake-arctic-embed-s | 32M | 51.98 |
|
| 65 |
+
| bge-small-en-v1.5 | 33M | 51.65 |
|
| 66 |
+
| granite-embedding-small-english-r2 | 47M | 50.87 |
|
| 67 |
+
| snowflake-arctic-embed-xs | 23M | 50.15 |
|
| 68 |
+
| e5-small-v2 | 33M | 49.04 |
|
| 69 |
+
| SPLADE++ | 110M | 48.88 |
|
| 70 |
+
| MiniLM-L6-v2 | 23M | 41.95 |
|
| 71 |
+
| BM25 | – | 41.14 |
|
| 72 |
|
| 73 |
|
| 74 |
# Quickstart
|