up

README.md

@@ -1,34 +1,61 @@
 # Cross-Encoder for MS Marco
 
-This model uses [TinyBERT](https://github.com/huawei-noah/Pretrained-Language-Model/tree/master/TinyBERT), a tiny BERT model with only 6 layers. The base model is: General_TinyBERT_v2(6layer-768dim)
+This model was trained on the [MS Marco Passage Ranking](https://github.com/microsoft/MSMARCO-Passage-Ranking) task.
 
-It was trained on [MS Marco Passage Ranking](https://github.com/microsoft/MSMARCO-Passage-Ranking) task.
+The model can be used for Information Retrieval: Given a query, encode the query will all possible passages (e.g. retrieved with ElasticSearch). Then sort the passages in a decreasing order. See [SBERT.net Retrieve & Re-rank](https://www.sbert.net/examples/applications/retrieve_rerank/README.html) for more details. The training code is available here: [SBERT.net Training MS Marco](https://github.com/UKPLab/sentence-transformers/tree/master/examples/training/ms_marco)
 
-The model can be used for Information Retrieval: Given a query, encode the query will all possible passages (e.g. retrieved with ElasticSearch). Then sort the passages in a decreasing order. See [SBERT.net Information Retrieval](https://github.com/UKPLab/sentence-transformers/tree/master/examples/applications/information-retrieval) for more details. The training code is available here: [SBERT.net Training MS Marco](https://github.com/UKPLab/sentence-transformers/tree/master/examples/training/ms_marco)
 
-## Usage and Performance
+## Usage with Transformers
 
-Pre-trained models can be used like this:
+```python
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
+
+model = AutoModelForSequenceClassification.from_pretrained('model_name')
+tokenizer = AutoTokenizer.from_pretrained('model_name')
+
+features = tokenizer(['How many people live in Berlin?', 'How many people live in Berlin?'], ['Berlin has a population of 3,520,031 registered inhabitants in an area of 891.82 square kilometers.', 'New York City is famous for the Metropolitan Museum of Art.'],  padding=True, truncation=True, return_tensors="pt")
+
+model.eval()
+with torch.no_grad():
+    scores = model(**features).logits
+    print(scores)
 ```
+
+
+## Usage with SentenceTransformers
+
+The usage becomes easier when you have [SentenceTransformers](https://www.sbert.net/) installed. Then, you can use the pre-trained models like this:
+```python
 from sentence_transformers import CrossEncoder
 model = CrossEncoder('model_name', max_length=512)
 scores = model.predict([('Query', 'Paragraph1'), ('Query', 'Paragraph2') , ('Query', 'Paragraph3')])
 ```
 
+
+## Performance
 In the following table, we provide various pre-trained Cross-Encoders together with their performance on the [TREC Deep Learning 2019](https://microsoft.github.io/TREC-2019-Deep-Learning/) and the [MS Marco Passage Reranking](https://github.com/microsoft/MSMARCO-Passage-Ranking/) dataset. 
 
 
-| Model-Name        | NDCG@10 (TREC DL 19) | MRR@10 (MS Marco Dev)  | Docs / Sec (BertTokenizerFast) | Docs / Sec |
-| ------------- |:-------------| -----| --- | --- |
-| cross-encoder/ms-marco-TinyBERT-L-2  | 67.43 | 30.15  | 9000 | 780
-| cross-encoder/ms-marco-TinyBERT-L-4  | 68.09 | 34.50  | 2900 | 760
-| cross-encoder/ms-marco-TinyBERT-L-6 |  69.57 | 36.13  | 680 | 660
-| cross-encoder/ms-marco-electra-base | 71.99 | 36.41 | 340 | 340
-| *Other models* | | | |
-| nboost/pt-tinybert-msmarco | 63.63 | 28.80 | 2900 | 760
-| nboost/pt-bert-base-uncased-msmarco | 70.94 | 34.75 | 340 | 340|
-| nboost/pt-bert-large-msmarco | 73.36 | 36.48 | 100 | 100 |
-| Capreolus/electra-base-msmarco | 71.23 | | 340 | 340 |
-| amberoad/bert-multilingual-passage-reranking-msmarco | 68.40 | | 330 | 330 
+| Model-Name        | NDCG@10 (TREC DL 19) | MRR@10 (MS Marco Dev)  | Docs / Sec |
+| ------------- |:-------------| -----| --- | 
+| **Version 2 models** | | | 
+| cross-encoder/ms-marco-TinyBERT-L-2-v2 | 69.84 | 32.56 | 9000
+| cross-encoder/ms-marco-MiniLM-L-2-v2 | 71.01 | 34.85 | 4100
+| cross-encoder/ms-marco-MiniLM-L-4-v2 | 73.04 | 37.70 | 2500
+| cross-encoder/ms-marco-MiniLM-L-6-v2 | 74.30 | 39.01 | 1800
+| cross-encoder/ms-marco-MiniLM-L-12-v2 | 74.31 | 39.02 | 960
+| **Version 1 models** | | | 
+| cross-encoder/ms-marco-TinyBERT-L-2  | 67.43 | 30.15  | 9000
+| cross-encoder/ms-marco-TinyBERT-L-4  | 68.09 | 34.50  | 2900
+| cross-encoder/ms-marco-TinyBERT-L-6 |  69.57 | 36.13  | 680
+| cross-encoder/ms-marco-electra-base | 71.99 | 36.41 | 340
+| **Other models** | | | 
+| nboost/pt-tinybert-msmarco | 63.63 | 28.80 | 2900 
+| nboost/pt-bert-base-uncased-msmarco | 70.94 | 34.75 | 340 
+| nboost/pt-bert-large-msmarco | 73.36 | 36.48 | 100 
+| Capreolus/electra-base-msmarco | 71.23 | 36.89 | 340 
+| amberoad/bert-multilingual-passage-reranking-msmarco | 68.40 | 35.54 | 330 
+| sebastian-hofstaetter/distilbert-cat-margin_mse-T2-msmarco | 72.82 | 37.88 | 720
  
- Note: Runtime was computed on a V100 GPU. A bottleneck for smaller models is the standard Python tokenizer from Huggingface v3. Replacing it with the fast tokenizer based on Rust, the throughput is significantly improved:
+ Note: Runtime was computed on a V100 GPU.