Update README.md
Browse files
README.md
CHANGED
@@ -7,21 +7,21 @@ tags:
|
|
7 |
- feature-extraction
|
8 |
license: mit
|
9 |
datasets:
|
10 |
-
- avemio/
|
11 |
language:
|
12 |
- de
|
13 |
- en
|
14 |
base_model:
|
15 |
-
- avemio/
|
16 |
- WhereIsAI/UAE-Large-V1
|
17 |
base_model_relation: merge
|
18 |
---
|
19 |
|
20 |
-
<img src="https://www.
|
21 |
|
22 |
-
#
|
23 |
|
24 |
-
This is a [sentence-transformers](https://www.SBERT.net) model trained on this [Dataset](https://huggingface.co/datasets/avemio/
|
25 |
It was merged with the Base-Model [WhereIsAI/UAE-Large-V1](https://huggingface.co/WhereIsAI/UAE-Large-V1) again to maintain performance on other languages again.
|
26 |
|
27 |
## Model Details
|
@@ -73,7 +73,7 @@ SentenceTransformer(
|
|
73 |
### STS (Semantic Textual Similarity)
|
74 |
- GermanSTSBenchmark
|
75 |
|
76 |
-
| TASK | [UAE](https://huggingface.co/WhereIsAI/UAE-Large-V1/) | [
|
77 |
|-------------------------------------|-------|----------|------------|--------------|----------------|
|
78 |
| AmazonCounterfactualClassification | **0.5650** | 0.5449 | 0.5401 | -2.01% | -2.48% |
|
79 |
| AmazonReviewsClassification | 0.2738 | 0.2745 | **0.2782** | 0.08% | 0.44% |
|
@@ -88,20 +88,20 @@ SentenceTransformer(
|
|
88 |
| PawsXPairClassification | **0.5452** | 0.5077 | 0.5162 | -3.76% | -2.90% |
|
89 |
|
90 |
|
91 |
-
## Evaluation on
|
92 |
|
93 |
Accuracy is calculated by evaluating if the relevant context is the highest ranking embedding of the whole context array.
|
94 |
-
See Eval-Dataset and Evaluation Code [here](https://huggingface.co/datasets/avemio/
|
95 |
|
96 |
| Model Name | Accuracy |
|
97 |
|-------------------------------------------------|-----------|
|
98 |
| [bge-m3](https://huggingface.co/BAAI/bge-m3 ) | 0.8806 |
|
99 |
| [UAE-Large-V1](https://huggingface.co/WhereIsAI/UAE-Large-V1) | 0.8393 |
|
100 |
-
| [
|
101 |
-
| [
|
102 |
-
| [
|
103 |
-
| [
|
104 |
-
| [
|
105 |
|
106 |
|
107 |
## Usage
|
|
|
7 |
- feature-extraction
|
8 |
license: mit
|
9 |
datasets:
|
10 |
+
- avemio/German-RAG-EMBEDDING-TRIPLES-HESSIAN-AI
|
11 |
language:
|
12 |
- de
|
13 |
- en
|
14 |
base_model:
|
15 |
+
- avemio/German-RAG-UAE-LARGE-V1-TRIPLES-HESSIAN-AI
|
16 |
- WhereIsAI/UAE-Large-V1
|
17 |
base_model_relation: merge
|
18 |
---
|
19 |
|
20 |
+
<img src="https://www.German-RAG.ai/wp-content/uploads/2024/12/German-RAG-ICON-TO-WORDLOGO-Animation_Loop-small-ezgif.com-video-to-gif-converter.gif" alt="German-RAG Logo" width="400" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
|
21 |
|
22 |
+
# German-RAG-UAE-LARGE-V1-TRIPLES-MERGED-HESSIAN-AI
|
23 |
|
24 |
+
This is a [sentence-transformers](https://www.SBERT.net) model trained on this [Dataset](https://huggingface.co/datasets/avemio/German-RAG-Embedding-Triples-Hessian-AI) with roughly 300k Triple-Samples. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
|
25 |
It was merged with the Base-Model [WhereIsAI/UAE-Large-V1](https://huggingface.co/WhereIsAI/UAE-Large-V1) again to maintain performance on other languages again.
|
26 |
|
27 |
## Model Details
|
|
|
73 |
### STS (Semantic Textual Similarity)
|
74 |
- GermanSTSBenchmark
|
75 |
|
76 |
+
| TASK | [UAE](https://huggingface.co/WhereIsAI/UAE-Large-V1/) | [German-RAG-UAE](https://huggingface.co/avemio/German-RAG-UAE-LARGE-V1-TRIPLES-HESSIAN-AI/) | Merged-UAE | German-RAG vs. UAE | Merged vs. UAE |
|
77 |
|-------------------------------------|-------|----------|------------|--------------|----------------|
|
78 |
| AmazonCounterfactualClassification | **0.5650** | 0.5449 | 0.5401 | -2.01% | -2.48% |
|
79 |
| AmazonReviewsClassification | 0.2738 | 0.2745 | **0.2782** | 0.08% | 0.44% |
|
|
|
88 |
| PawsXPairClassification | **0.5452** | 0.5077 | 0.5162 | -3.76% | -2.90% |
|
89 |
|
90 |
|
91 |
+
## Evaluation on German-RAG-EMBEDDING-BENCHMARK
|
92 |
|
93 |
Accuracy is calculated by evaluating if the relevant context is the highest ranking embedding of the whole context array.
|
94 |
+
See Eval-Dataset and Evaluation Code [here](https://huggingface.co/datasets/avemio/German-RAG-EMBEDDING-BENCHMARK)
|
95 |
|
96 |
| Model Name | Accuracy |
|
97 |
|-------------------------------------------------|-----------|
|
98 |
| [bge-m3](https://huggingface.co/BAAI/bge-m3 ) | 0.8806 |
|
99 |
| [UAE-Large-V1](https://huggingface.co/WhereIsAI/UAE-Large-V1) | 0.8393 |
|
100 |
+
| [German-RAG-BGE-M3-TRIPLES-HESSIAN-AI](https://huggingface.co/avemio/German-RAG-BGE-M3-TRIPLES-HESSIAN-AI) | 0.8857 |
|
101 |
+
| [German-RAG-BGE-M3-TRIPLES-MERGED-HESSIAN-AI](https://huggingface.co/avemio/German-RAG-BGE-M3-TRIPLES-MERGED-HESSIAN-AI) | **0.8866** |
|
102 |
+
| [German-RAG-BGE-M3-MERGED-x-SNOWFLAKE-ARCTIC-HESSIAN-AI](https://huggingface.co/avemio/German-RAG-BGE-M3-MERGED-x-SNOWFLAKE-ARCTIC-HESSIAN-AI) | **0.8866** |
|
103 |
+
| [German-RAG-UAE-LARGE-V1-TRIPLES-HESSIAN-AI](https://huggingface.co/avemio/German-RAG-UAE-LARGE-V1-TRIPLES-HESSIAN-AI) | 0.8763 |
|
104 |
+
| [German-RAG-UAE-LARGE-V1-TRIPLES-MERGED-HESSIAN-AI](https://huggingface.co/avemio/German-RAG-UAE-LARGE-V1-TRIPLES-MERGED-HESSIAN-AI) | 0.8771 |
|
105 |
|
106 |
|
107 |
## Usage
|