https://github.com/saba-gul/fastapi-sentiment-analysis-and-text-summarization-api
This repository contains code to create a simple API using FastAPI for performing sentiment analysis, text summarization, and emotion classification using pre-trained models from the Hugging Face Transformers library.
https://github.com/saba-gul/fastapi-sentiment-analysis-and-text-summarization-api
Last synced: 8 months ago
JSON representation
This repository contains code to create a simple API using FastAPI for performing sentiment analysis, text summarization, and emotion classification using pre-trained models from the Hugging Face Transformers library.
- Host: GitHub
- URL: https://github.com/saba-gul/fastapi-sentiment-analysis-and-text-summarization-api
- Owner: Saba-Gul
- Created: 2024-08-13T02:37:32.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2024-08-14T03:45:42.000Z (about 1 year ago)
- Last Synced: 2025-01-13T16:28:16.471Z (9 months ago)
- Language: Jupyter Notebook
- Size: 24.4 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# FastAPI Sentiment Analysis and Text Summarization API
This repository contains code to create a simple API using FastAPI for performing sentiment analysis, text summarization, and emotion classification using pre-trained models from the Hugging Face Transformers library.
## Introduction
FastAPI is a modern, fast (high-performance) web framework for building APIs with Python 3.6+ based on standard Python type hints. This project demonstrates how to create API endpoints for sentiment analysis, text summarization, and emotion classification using FastAPI.
## Features
- **Sentiment Analysis:** Classifies the sentiment of the input text as positive, negative, or neutral.
- **Text Summarization:** Summarizes long pieces of text into concise summaries.
- **Emotion Classification:** Identifies the emotion behind the summarized text.
- **Custom LLM Endpoint:** Example of integrating OpenAI's GPT-4 for custom language model responses.
## Installation
To install the required dependencies, run:
```bash
pip install fastapi nest-asyncio pyngrok uvicorn transformers openai
```
## Usage
1. **Run the FastAPI server:**
- The server runs on `http://localhost:8000` by default.
- Use `ngrok` to expose the server to the internet for public access.
2. **Using the API:**
- After running the server, you can access the API endpoints to perform sentiment analysis, text summarization, emotion classification, and more.
## API Endpoints
### Home
- **URL:** `/`
- **Method:** `GET`
- **Description:** Returns a welcome message.
### Sentiment Analysis
- **URL:** `/sentiment/{input_text}`
- **Method:** `GET`
- **Description:** Returns the sentiment analysis of the provided input text.
### Text Summarization and Emotion Classification
- **URL:** `/ai/{input_text}`
- **Method:** `GET`
- **Description:** Summarizes the input text and then classifies the emotion of the summarized text.
### Custom LLM Endpoint
- **URL:** `/chat/{prompt}`
- **Method:** `GET`
- **Description:** Uses OpenAI's GPT-4 to generate a response based on the provided prompt.
## Running on Google Colab
This code can be executed in a Google Colab environment. Here's a brief overview:
- **Installing Dependencies:** Colab requires installing FastAPI, `nest-asyncio`, `pyngrok`, and `uvicorn`.
- **Running the Server:** The server is run using `uvicorn`, and `ngrok` is used to expose it to the public.
- **Custom API Integration:** The code demonstrates how to use pre-trained models for NLP tasks and integrate a custom LLM using OpenAI's API.
Certainly! Here’s how you can test the endpoints using Postman and directly through `cURL`.
### Example Usage
#### Home Endpoint
**URL:** `/`
- **Method:** `GET`
- **Postman:**
- Set the request type to `GET`.
- Enter the URL provided by `ngrok` (e.g., `https:///`).
- Click on `Send`.
- **cURL:**
```bash
curl -X GET "https:///"
```
#### Test Endpoint
**URL:** `/test`
- **Method:** `GET`
- **Postman:**
- Set the request type to `GET`.
- Enter the URL provided by `ngrok` (e.g., `https:///test`).
- Click on `Send`.
- **cURL:**
```bash
curl -X GET "https:///test"
```
#### Sentiment Analysis
**URL:** `/sentiment/{user_input}`
- **Method:** `GET`
- **Example Input:** `I love FastAPI`
- **Postman:**
- Set the request type to `GET`.
- Enter the URL: `https:///sentiment/I%20love%20FastAPI`
- Click on `Send`.
- **cURL:**
```bash
curl -X GET "https:///sentiment/I%20love%20FastAPI"
```
#### Text Summarization & Emotion Classification
**URL:** `/ai/{input_text}`
- **Method:** `GET`
- **Example Input:** `FastAPI is a modern, fast (high-performance) web framework for building APIs with Python 3.6+ based on standard Python type hints.`
- **Postman:**
- Set the request type to `GET`.
- Enter the URL: `https:///ai/FastAPI%20is%20a%20modern,%20fast%20(high-performance)%20web%20framework%20for%20building%20APIs%20with%20Python%203.6+%20based%20on%20standard%20Python%20type%20hints.`
- Click on `Send`.
- **cURL:**
```bash
curl -X GET "https:///ai/FastAPI%20is%20a%20modern,%20fast%20(high-performance)%20web%20framework%20for%20building%20APIs%20with%20Python%203.6+%20based%20on%20standard%20Python%20type%20hints."
```
#### Custom GPT-3/4 Response
**URL:** `/gpt/{prompt}`
- **Method:** `GET`
- **Example Prompt:** `Tell me a joke.`
- **Postman:**
- Set the request type to `GET`.
- Enter the URL: `https:///gpt/Tell%20me%20a%20joke.`
- Click on `Send`.
- **cURL:**
```bash
curl -X GET "https:///gpt/Tell%20me%20a%20joke."
```
#### Multiplication Table
**URL:** `/{number}`
- **Method:** `GET`
- **Example Input:** `5`
- **Postman:**
- Set the request type to `GET`.
- Enter the URL: `https:///5`
- Click on `Send`.
- **cURL:**
```bash
curl -X GET "https:///5"
```
Replace `` with the actual `ngrok` URL provided when you start your FastAPI server.