|
--- |
|
license: apache-2.0 |
|
--- |
|
|
|
|
|
# Agentic Transformer System for Hugging Face |
|
|
|
This repository contains the code, configurations, and scripts to create a modular AI system with multiple specialized agents for text classification, sentiment analysis, summarization, code generation, and more. |
|
|
|
## Project Overview |
|
|
|
This system uses a mixture of specialized agents to handle diverse tasks. Each agent is fine-tuned for its specific role and is managed by an orchestration layer that ensures smooth communication and data flow between agents. |
|
|
|
## Models and Datasets |
|
|
|
### Pre-trained Models |
|
|
|
- **Text Classification Agent:** BERT Base Uncased ([Hugging Face Model](https://huggingface.co/bert-base-uncased)) |
|
- **Sentiment Analysis Agent:** RoBERTa ([Hugging Face Model](https://huggingface.co/roberta-base)) |
|
- **Summarization Agent:** BART ([Hugging Face Model](https://huggingface.co/facebook/bart-large)) |
|
- **Code Generation Agent:** CodeBERT ([Hugging Face Model](https://huggingface.co/microsoft/codebert-base)) |
|
- **Web Search Agent:** DistilBERT ([Hugging Face Model](https://huggingface.co/distilbert-base-uncased)) |
|
|
|
### Datasets |
|
|
|
- **Training and Validation Dataset:** IMDB Movie Reviews ([Hugging Face Dataset](https://huggingface.co/datasets/imdb)) |
|
- **Evaluation Dataset:** SST-2 ([Hugging Face Dataset](https://huggingface.co/datasets/glue/viewer/sst2)) |
|
|
|
## Installation |
|
|
|
To install the required dependencies, run: |
|
|
|
```bash |
|
pip install -r requirements.txt |
|
|
