Hugging Face's logo

rufimelo
/
Legal-BERTimbau-sts-base

Sentence Similarity
sentence-transformers
PyTorch
Transformers
Portuguese
bert
feature-extraction
Eval Results
Inference Endpoints
Model card Files Community
rufimelo commited on
Commit
51f00ab
•
1 Parent(s): 80a6c23

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +17 -16
README.md CHANGED
@@ -85,22 +85,23 @@ print(sentence_embeddings)
85
  ```
86
  ## Evaluation Results STS
87
 
88
- | Model| Assin | Assin2|stsb_multi_mt pt|
89
- | ---------------------------------------- | ---------- | ---------- |---------- |
90
- | Legal-BERTimbau-sts-base| 0.71457| 0.73545 | 0.72383|
91
- | Legal-BERTimbau-sts-base-ma| 0.74874 | 0.79532|0.82254 |
92
- | Legal-BERTimbau-sts-base-ma-v2| 0.75481 | 0.80262|0.82178|
93
- | Legal-BERTimbau-sts-large| 0.76629| 0.82357 | 0.79120|
94
- | Legal-BERTimbau-sts-large-v2| 0.76299 | 0.81121|0.81726 |
95
- | Legal-BERTimbau-sts-large-ma| 0.76195| 0.81622 | 0.82608|
96
- | Legal-BERTimbau-sts-large-ma-v2| 0.7836| 0.8462| 0.8261|
97
- | Legal-BERTimbau-sts-large-ma-v3| 0.7749| 0.8470| 0.8364|
98
- | ---------------------------------------- | ---------- |---------- |---------- |
99
- | BERTimbau base Fine-tuned for STS|0.78455 | 0.80626|0.82841|
100
- | BERTimbau large Fine-tuned for STS|0.78193 | 0.81758|0.83784|
101
- | ---------------------------------------- | ---------- |---------- |---------- |
102
- | paraphrase-multilingual-mpnet-base-v2| 0.71457| 0.79831 |0.83999 |
103
- | paraphrase-multilingual-mpnet-base-v2 Fine-tuned with assin(s)| 0.77641|0.79831 |0.84575 |
 
104
  ## Training
105
  rufimelo/Legal-BERTimbau-sts-base is based on Legal-BERTimbau-largewhich derives from [BERTimbau](https://huggingface.co/neuralmind/bert-base-portuguese-cased) base.
106
  It was trained for Semantic Textual Similarity, being submitted to a fine tuning stage with the [assin](https://huggingface.co/datasets/assin) and [assin2](https://huggingface.co/datasets/assin2) datasets.
 
85
  ```
86
  ## Evaluation Results STS
87
 
88
+
89
+ | Model| Assin | Assin2|stsb_multi_mt pt| avg|
90
+ | ---------------------------------------- | ---------- | ---------- |---------- |---------- |
91
+ | Legal-BERTimbau-sts-base| 0.71457| 0.73545 | 0.72383|0.72462|
92
+ | Legal-BERTimbau-sts-base-ma| 0.74874 | 0.79532|0.82254 |0.78886|
93
+ | Legal-BERTimbau-sts-base-ma-v2| 0.75481 | 0.80262|0.82178|0.79307|
94
+ | Legal-BERTimbau-sts-large| 0.76629| 0.82357 | 0.79120|0.79369|
95
+ | Legal-BERTimbau-sts-large-v2| 0.76299 | 0.81121|0.81726 |0.79715|
96
+ | Legal-BERTimbau-sts-large-ma| 0.76195| 0.81622 | 0.82608|0.80142|
97
+ | Legal-BERTimbau-sts-large-ma-v2| 0.7836| 0.8462| 0.8261| 0.81863|
98
+ | Legal-BERTimbau-sts-large-ma-v3| 0.7749| **0.8470**| 0.8364| **0.81943**|
99
+ | ---------------------------------------- | ---------- |---------- |---------- |---------- |
100
+ | BERTimbau base Fine-tuned for STS|**0.78455** | 0.80626|0.82841|0.80640|
101
+ | BERTimbau large Fine-tuned for STS|0.78193 | 0.81758|0.83784|0.81245|
102
+ | ---------------------------------------- | ---------- |---------- |---------- |---------- |
103
+ | paraphrase-multilingual-mpnet-base-v2| 0.71457| 0.79831 |0.83999 |0.78429|
104
+ | paraphrase-multilingual-mpnet-base-v2 Fine-tuned with assin(s)| 0.77641|0.79831 |**0.84575**|0.80682|
105
  ## Training
106
  rufimelo/Legal-BERTimbau-sts-base is based on Legal-BERTimbau-largewhich derives from [BERTimbau](https://huggingface.co/neuralmind/bert-base-portuguese-cased) base.
107
  It was trained for Semantic Textual Similarity, being submitted to a fine tuning stage with the [assin](https://huggingface.co/datasets/assin) and [assin2](https://huggingface.co/datasets/assin2) datasets.