|
--- |
|
language: |
|
- en |
|
- fr |
|
license: mit |
|
datasets: |
|
- UMA-IA/VELA-Engine-v1 |
|
base_model: mistralai/Mistral-7B-v0.1 |
|
tags: |
|
- aerospace |
|
- aeronautics |
|
- engineering |
|
- technical-QA |
|
pipeline_tag: text-generation |
|
--- |
|
|
|
|
|
## Model Details |
|
|
|
**Model Name:** UMA-IA/CENTAURUS-Engine-v1 |
|
**Authors:** |
|
- **Youri LALAIN**, Engineering student at French Engineering School ECE |
|
- **Lilian RAGE**, Engineering student at French Engineering School ECE |
|
|
|
**Base Model:** [Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) |
|
**Fine-tuned Dataset:** [UMA-IA/VELA-Engine-v1](https://huggingface.co/datasets/UMA-IA/UMA_Dataset_Engine_Aero_LLM) |
|
**License:** Apache 2.0 |
|
|
|
## Model Description |
|
|
|
# Mistral-7B Fine-tuné sur les moteurs aérospatiaux |
|
|
|
UMA-IA/CENTAURUS-Engine-v1 is a specialized fine-tuned version of Mistral-7B designed to provide accurate and detailed answers to technical questions related to aerospace and aeronautical engines. The model leverages the UMA-IA/UMA_Dataset_Engine_Aero_LLM to enhance its understanding of complex engineering principles, propulsion systems, and aerospace technologies. |
|
|
|
## Capabilities |
|
- Technical Q&A on aerospace and aeronautical engines |
|
- Analysis and explanations of propulsion system components |
|
- Assistance in understanding aerospace engineering concepts |
|
|
|
## Use Cases |
|
- Aerospace research and engineering support |
|
- Educational purposes for students and professionals |
|
- Assisting in aerospace-related R&D projects |
|
|
|
## Training Details |
|
This model was fine-tuned on UMA-IA/VELA-Engine-v1, a curated dataset focusing on aerospace engines, propulsion systems, and general aeronautical engineering. The fine-tuning process was performed using supervised learning to adapt Mistral-7B to technical discussions. |
|
|
|
|
|
## How to Use |
|
You can load the model using Hugging Face's `transformers` library: |
|
|
|
```python |
|
from transformers import AutoModelForCausalLM, AutoTokenizer |
|
|
|
model_name = "UMA-IA/CENTAURUS-Engine-v1" |
|
|
|
model = AutoModelForCausalLM.from_pretrained(model_name) |
|
tokenizer = AutoTokenizer.from_pretrained(model_name) |
|
|
|
input_text = "Explain the working principle of a turbofan engine." |
|
inputs = tokenizer(input_text, return_tensors="pt") |
|
outputs = model.generate(**inputs) |
|
print(tokenizer.decode(outputs[0], skip_special_tokens=True)) |
