Update README.md
Browse files
README.md
CHANGED
|
@@ -5,6 +5,15 @@ tags:
|
|
| 5 |
- sentence-transformers
|
| 6 |
- sentence-similarity
|
| 7 |
- feature-extraction
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
# SentenceTransformer
|
|
@@ -40,6 +49,42 @@ SentenceTransformer(
|
|
| 40 |
)
|
| 41 |
```
|
| 42 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
## Usage
|
| 44 |
|
| 45 |
### Direct Usage (Sentence Transformers)
|
|
|
|
| 5 |
- sentence-transformers
|
| 6 |
- sentence-similarity
|
| 7 |
- feature-extraction
|
| 8 |
+
license: mit
|
| 9 |
+
datasets:
|
| 10 |
+
- avemio/GRAG-Embedding-Triples-Hessian-AI
|
| 11 |
+
language:
|
| 12 |
+
- de
|
| 13 |
+
- en
|
| 14 |
+
base_model:
|
| 15 |
+
- avemio/GRAG-UAE-LARGE-V1-TRIPLES-HESSIAN-AI
|
| 16 |
+
- WhereIsAI/UAE-Large-V1
|
| 17 |
---
|
| 18 |
|
| 19 |
# SentenceTransformer
|
|
|
|
| 49 |
)
|
| 50 |
```
|
| 51 |
|
| 52 |
+
## Evaluation MTEB-Tasks
|
| 53 |
+
|
| 54 |
+
### Classification
|
| 55 |
+
- AmazonCounterfactualClassification
|
| 56 |
+
- AmazonReviewsClassification
|
| 57 |
+
- MassiveIntentClassification
|
| 58 |
+
- MassiveScenarioClassification
|
| 59 |
+
- MTOPDomainClassification
|
| 60 |
+
- MTOPIntentClassification
|
| 61 |
+
|
| 62 |
+
### Pair Classification
|
| 63 |
+
- FalseFriendsGermanEnglish
|
| 64 |
+
- PawsXPairClassification
|
| 65 |
+
|
| 66 |
+
### Retrieval
|
| 67 |
+
- GermanQuAD-Retrieval
|
| 68 |
+
- GermanDPR
|
| 69 |
+
|
| 70 |
+
### STS (Semantic Textual Similarity)
|
| 71 |
+
- GermanSTSBenchmark
|
| 72 |
+
|
| 73 |
+
| TASK | UAE | GRAG-UAE | Merged-UAE | GRAG vs. UAE | Merged vs. UAE |
|
| 74 |
+
|-------------------------------------|-------|----------|------------|--------------|----------------|
|
| 75 |
+
| AmazonCounterfactualClassification | 0.5650 | 0.5449 | 0.5401 | -2.01% | -2.48% |
|
| 76 |
+
| AmazonReviewsClassification | 0.2738 | 0.2745 | **0.2782** | 0.08% | 0.44% |
|
| 77 |
+
| FalseFriendsGermanEnglish | 0.4808 | 0.4777 | 0.4703 | -0.32% | -1.05% |
|
| 78 |
+
| GermanQuAD-Retrieval | 0.7811 | 0.8353 | **0.8628** | 5.42% | 8.18% |
|
| 79 |
+
| GermanSTSBenchmark | 0.6421 | 0.6568 | **0.6754** | 1.47% | 3.33% |
|
| 80 |
+
| MassiveIntentClassification | 0.5139 | 0.4884 | 0.4714 | -2.55% | -4.25% |
|
| 81 |
+
| MassiveScenarioClassification | 0.6062 | 0.5837 | **0.6111** | -2.25% | 0.49% |
|
| 82 |
+
| GermanDPR | 0.6750 | 0.7210 | **0.7507** | 4.60% | 7.57% |
|
| 83 |
+
| MTOPDomainClassification | 0.7625 | 0.7450 | **0.7686** | -1.75% | 0.61% |
|
| 84 |
+
| MTOPIntentClassification | 0.4994 | 0.4516 | 0.4413 | -4.77% | -5.80% |
|
| 85 |
+
| PawsXPairClassification | 0.5452 | 0.5077 | 0.5162 | -3.76% | -2.90% |
|
| 86 |
+
|
| 87 |
+
|
| 88 |
## Usage
|
| 89 |
|
| 90 |
### Direct Usage (Sentence Transformers)
|