|
|
--- |
|
|
library_name: model2vec |
|
|
license: mit |
|
|
model_name: cnmoro/custom-model2vec-tokenlearn-medium |
|
|
tags: |
|
|
- embeddings |
|
|
- static-embeddings |
|
|
- sentence-transformers |
|
|
language: |
|
|
- pt |
|
|
- en |
|
|
--- |
|
|
A custom model2vec model, trained using a modified version of the [tokenlearn](https://github.com/MinishLab/tokenlearn) library. |
|
|
|
|
|
Base model is nomic-ai/nomic-embed-text-v2-moe. |
|
|
|
|
|
The output dimension is 256, and the vocabulary size is 249.999 |
|
|
|
|
|
The training process used a mix of English (10%) and Portuguese (90%) texts. |
|
|
|
|
|
```python |
|
|
from model2vec import StaticModel |
|
|
|
|
|
# Load a pretrained Sentence Transformer model |
|
|
model = StaticModel.from_pretrained("cnmoro/custom-model2vec-tokenlearn-medium") |
|
|
|
|
|
# Compute text embeddings |
|
|
embeddings = model.encode(["Example sentence"]) |
|
|
``` |
