76 lines
2.9 KiB
Markdown
76 lines
2.9 KiB
Markdown
# Fabelous-Albert-Uncased
|
|
|
|
**Fabelous-Albert-Uncased** is a bilingual ALBERT model pretrained on German, English, and code. This uncased model is a Masked Language Model (MLM) and can be fine-tuned for a variety of tasks, including but not limited to:
|
|
|
|
- Named Entity Recognition (NER)
|
|
- Binary Classification
|
|
- Text Completion
|
|
|
|
The model has been designed for efficiency and compatibility, requiring the use of the `FastTokenizer` for optimal performance.
|
|
|
|
|
|
## Features
|
|
|
|
### 1. **Bilingual Support**
|
|
- Trained on English and German text, enabling seamless bilingual tasks.
|
|
|
|
### 2. **Code Understanding**
|
|
- Incorporates code in its training data, making it suitable for programming-related NLP tasks.
|
|
|
|
### 3. **Uncased**
|
|
- Treats words as case-insensitive, which simplifies preprocessing steps and generalizes better for certain tasks.
|
|
|
|
### 4. **Fine-Tuning Ready**
|
|
- Easily fine-tune for tasks such as text classification, named entity recognition, and more.
|
|
|
|
|
|
## Downloading the Model
|
|
|
|
You can download the `fabelous-albert-uncased` model using the following link:
|
|
|
|
[Download Fabelous-Albert-Uncased Model](https://gitea.fabelous.app/Fabel/Fabelous-albert-uncased/releases/download/latest/fabelous-albert-uncased.zip)
|
|
|
|
### Installation Instructions
|
|
|
|
1. Click the link above to download a ZIP file containing the model files.
|
|
2. Extract the ZIP file into your desired directory.
|
|
3. Load the model in your Python project using the `transformers` library.
|
|
|
|
## Usage Example
|
|
|
|
Below is a sample code snippet to demonstrate how to use the `fabelous-albert-uncased` model for a masked language modeling task:
|
|
|
|
```python
|
|
from transformers import pipeline
|
|
|
|
# Load the pipeline with the Fabelous-Albert-Uncased model
|
|
unmasker = pipeline('fill-mask', model='fabelous-albert-uncased')
|
|
|
|
# Perform masked language modeling
|
|
output = unmasker("Hello I'm a [MASK] model.")
|
|
print(output)
|
|
```
|
|
|
|
|
|
|
|
## Future Enhancements: New Model Announcement
|
|
|
|
We are thrilled to announce that a new version of the model is currently under development! The upcoming model will:
|
|
|
|
- **Quadruple the Training Size**: With four times more data, expect significantly improved performance across diverse tasks.
|
|
- **Cased Version**: In addition to the uncased version, a cased model will be introduced, preserving capitalization for more nuanced language understanding.
|
|
- **Extended Language Support**: Support for multiple additional languages beyond English and German.
|
|
- **Slow- and FastTokenizer Support**: Support for both tokenizer Versions.
|
|
|
|
Stay tuned for updates as we prepare to release this enhanced model in the near future.
|
|
|
|
|
|
## License
|
|
|
|
This model is released under the [Creative Commons Attribution 4.0 International Licence](https://creativecommons.org/licenses/by/4.0/).
|
|
|
|
## Feedback and Support
|
|
|
|
If you encounter any issues or have questions, feel free to reach out through the project's [Gitea Issues page](https://gitea.fabelous.app/Fabel/Fabelous-albert-uncased/issues) or contact our support team at support@fabelous.app.
|
|
|