Sentence Similarity
sentence-transformers
Safetensors
English
bert
feature-extraction
historical-text
early-modern-english
economic-history
history-of-science
text-embeddings-inference
Instructions to use juliuskoschnick/SteamBERTh with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use juliuskoschnick/SteamBERTh with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("juliuskoschnick/SteamBERTh") 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] - Notebooks
- Google Colab
- Kaggle
Model card: training data includes encyclopedias and dictionaries
Browse files
README.md
CHANGED
|
@@ -67,8 +67,8 @@ are strongly anisotropic. Contrastive fine-tuning on the texts of interest addre
|
|
| 67 |
|
| 68 |
- **Base model:** [emanjavacas/MacBERTh](https://huggingface.co/emanjavacas/MacBERTh), mean pooling over tokens.
|
| 69 |
- **Data:** 21,760 texts: all titles of the English Short Title Catalogue (ESTC) in *applied physics, astronomy,
|
| 70 |
-
chemistry, mathematics, technical instructions in trades and agriculture,
|
| 71 |
-
and the short descriptions of English patents from Woodcroft (1854); patent descriptions with fewer than three
|
| 72 |
words are dropped.
|
| 73 |
- **Objective:** unsupervised SimCSE (Gao, Yao and Chen 2021): each text is paired with itself, dropout produces two
|
| 74 |
different views, and the other texts of the batch serve as negatives (multiple negatives ranking loss, scale 15).
|
|
|
|
| 67 |
|
| 68 |
- **Base model:** [emanjavacas/MacBERTh](https://huggingface.co/emanjavacas/MacBERTh), mean pooling over tokens.
|
| 69 |
- **Data:** 21,760 texts: all titles of the English Short Title Catalogue (ESTC) in *applied physics, astronomy,
|
| 70 |
+
chemistry, mathematics, encyclopedias and dictionaries, technical instructions in trades and agriculture,
|
| 71 |
+
navigation,* and *scientific instruments*, and the short descriptions of English patents from Woodcroft (1854); patent descriptions with fewer than three
|
| 72 |
words are dropped.
|
| 73 |
- **Objective:** unsupervised SimCSE (Gao, Yao and Chen 2021): each text is paired with itself, dropout produces two
|
| 74 |
different views, and the other texts of the batch serve as negatives (multiple negatives ranking loss, scale 15).
|