inital commit with explanation and release window

.gitignore

@@ -153,6 +153,9 @@ dmypy.json
 # Cython debug symbols
 cython_debug/
 
+# Models
+fabelous-albert-uncased
+
 # PyCharm
 #  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
 #  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore

README.md

@@ -1,3 +1,76 @@
-# Fabelous-albert-uncased
+# 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.
 
-This is a bilingual ALBERT model that has been pretrained on German, English, and code. It is a Masked Language Model (MLM) and can be fine-tuned for various tasks, such as Named Entity Recognition (NER) or binary classification. The model is uncased and is exclusively compatible with the FastTokenizer.

example.py

@@ -0,0 +1,5 @@
+from transformers import pipeline
+
+
+unmasker = pipeline('fill-mask', model='fabelous-albert-uncased')
+print(unmasker("Hello I'm a [MASK] model."))