https://github.com/rupanjana15/beforreal-backend

A FastAPI-based sarcasm detection API that uses a fine-tuned BERT model to classify text as sarcastic or not, with a sarcasm probability score.
https://github.com/rupanjana15/beforreal-backend

api bert docker fastapi machine-learning nlp python sarcasm-detection text-classification torch transformers

Last synced: 2 months ago
JSON representation

A FastAPI-based sarcasm detection API that uses a fine-tuned BERT model to classify text as sarcastic or not, with a sarcasm probability score.

• Host: GitHub
• URL: https://github.com/rupanjana15/beforreal-backend
• Owner: rupanjana15
• License: mit
• Created: 2024-11-12T14:05:22.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-11-12T14:26:40.000Z (over 1 year ago)
• Last Synced: 2025-01-23T13:33:57.094Z (over 1 year ago)
• Topics: api, bert, docker, fastapi, machine-learning, nlp, python, sarcasm-detection, text-classification, torch, transformers
• Language: Python
• Homepage: https://beforreal.api-cloud.one/docs
• Size: 47.9 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

          
# beforreal backend 🧐

![FastAPI](https://img.shields.io/badge/FastAPI-0.95.2-green)

![Docker](https://img.shields.io/badge/Docker-Supported-blue)

![Python](https://img.shields.io/badge/Python-3.8%2B-yellow)

![License](https://img.shields.io/badge/License-MIT-brightgreen)

A FastAPI-based sarcasm detection API using a fine-tuned BERT model to predict sarcasm in text.

## Setup

### 1. Download Model and Tokenizer

Get `model.pt` and `tokenizer.pkl` from the [Releases](https://github.com/yourusername/sarcasm-detection-api/releases) section, and place them in the `src/` directory.

### 2. Run Locally

```bash

pip install -r requirements.txt

uvicorn src.app:app --host 0.0.0.0 --port 8000

```

### 3. Run with Docker

```bash

docker build -t sarcasm-detection-api .

docker run -p 8000:8000 sarcasm-detection-api

```

## Usage

**Endpoint:** `POST /predict`

**Payload:**

```json

{

  "sentence": "I just love waiting in traffic all day!"

}

```

**Response:**

```json

{

  "prediction": "Sarcastic",

  "sarcasm_score": 0.87,

  "sarcasm": true

}

```

**Example:**

```bash

curl -X POST "http://localhost:8000/predict" -H "Content-Type: application/json" -d '{"sentence": "Your input here"}'

```

## Testing

```bash

pytest test_api.py

```

## License

[MIT License](./LICENSE)