|
--- |
|
title: Gemma Italian Camoscio Language Model |
|
tags: |
|
- italian-language-generation |
|
- camoscio-dataset |
|
- gemma-2b |
|
- autotrain |
|
datasets: |
|
- camoscio |
|
library_name: transformers |
|
model: theoracle/gemma_italian_camoscio |
|
license: other |
|
--- |
|
|
|
## Overview |
|
|
|
`theoracle/gemma_italian_camoscio` is a cutting-edge model specifically designed for Italian language generation. Leveraging the comprehensive Camoscio dataset, this model enhances the Gemma 2B architecture's capabilities in producing high-quality, contextually accurate Italian text. Developed with AutoTrain, it excels in various Italian text generation tasks, including but not limited to creative writing, article generation, and conversational responses. |
|
|
|
## Key Features |
|
|
|
- **Italian Language Focus**: Tailored to understand and generate Italian text, capturing the language's nuances and complexities. |
|
- **Camoscio Dataset Training**: Utilizes the rich Camoscio dataset, ensuring the model is well-versed in a wide range of Italian language styles and contexts. |
|
- **Gemma 2B Architecture**: Built on the powerful Gemma 2B framework, known for its efficiency and effectiveness in language generation tasks. |
|
- **AutoTrain Enhanced**: Benefits from AutoTrain's optimization, making the model both robust and versatile in handling Italian text generation. |
|
|
|
## Usage |
|
|
|
Here's how to use this model for generating Italian text: |
|
|
|
```python |
|
from transformers import AutoModelForCausalLM, AutoTokenizer |
|
|
|
model_path = "theoracle/gemma_italian_camoscio" |
|
tokenizer = AutoTokenizer.from_pretrained(model_path) |
|
model = AutoModelForCausalLM.from_pretrained( |
|
model_path, |
|
device_map="auto", |
|
torch_dtype='auto' |
|
).eval() |
|
|
|
# Example: Generating Italian text |
|
prompt = "Inizia la storia con una giornata soleggiata in Sicilia, dove" |
|
|
|
# Tokenize and generate text |
|
encoding = tokenizer(prompt, return_tensors='pt', padding=True, truncation=True, max_length=500, add_special_tokens=True) |
|
input_ids = encoding['input_ids'] |
|
attention_mask = encoding['attention_mask'] |
|
|
|
output_ids = model.generate( |
|
input_ids.to('cuda'), |
|
attention_mask=attention_mask.to('cuda'), |
|
max_new_tokens=300, |
|
pad_token_id=tokenizer.eos_token_id |
|
) |
|
|
|
generated_text = tokenizer.decode(output_ids[0], skip_special_tokens=True) |
|
print(generated_text) |
|
``` |
|
|
|
## Application Scenarios |
|
|
|
This model is ideal for: |
|
- Content creators looking to produce Italian-language articles, stories, or scripts. |
|
- Developers creating conversational AI applications in Italian. |
|
- Educators and language learners seeking tools for Italian language practice and improvement. |
|
|
|
## Training and Technology |
|
|
|
The `theoracle/gemma_italian_camoscio` model is trained using the AutoTrain platform for optimal performance, ensuring that it is well-suited for a broad spectrum of Italian text generation tasks. The Camoscio dataset provides a solid foundation, offering diverse and extensive coverage of the Italian language, which, combined with the Gemma 2B architecture, enables the model to generate coherent, nuanced, and contextually relevant Italian text. |
|
|
|
## License |
|
|
|
This model is available under an "other" license, facilitating its use in a wide array of applications. Users are encouraged to review the license terms to ensure compliance with their project requirements and intended use cases. |
