Gender Prediction API

git clone https://github.com/Nocturnailed-Community/Serving-Gender-Prediction.git

Install Library dependencies

torch
transformers
Pillow
fastapi
uvicorn
python-multipart

pip install -r requirements.txt

Running the Server

After setting up the environment, you can run the FastAPI server locally using Uvicorn.

Start the server:

Run the following command to start the FastAPI app on `http://127.0.0.1:8000`:

```bash
uvicorn app:app --reload
```

Alternatively, if you're using a different entry point for the FastAPI app:

```bash
uvicorn main:app --reload
```

This will reload the server automatically upon code changes.

Testing the API

Once the FastAPI server is running, you can test the API by sending HTTP requests to it. You can use Hoppscotch or Postman to make requests.

Using Hoppscotch to Test the API

1. Go to Hoppscotch.io in your web browser.

2. Select the method: Choose POST from the dropdown menu next to the URL input.

3. Enter the API URL: In the URL bar, type:

   http://127.0.0.1:8000/predict
   

4. Set the request body: In the body section, select the multipart/form-data option and enter the following:

   {
       "file": "Upload File"
   }
   

5. Send the request: Click on the Send button.

6. Check the response: You will receive a response from the server, such as:

   {
       "label": "Male",
       "image_base64": "<base64_encoded_image>"
   }
   

This will show the predicted intent for the input text.

Deployment

Instructions for running the model in docker

docker build -t gender-prediction .
docker run -d -p 8000:8000 gender-prediction

The instruction looks at running containers

docker ps

Push Docker Hub

docker build . -t your_username_docker/gender-prediction
docker login
docker push your_username_docker/gender-prediction

Docker Hub Repository

The Docker image for the gender prediction API is available on Docker Hub. You can pull the image using the following command:

docker pull ikhwan17/gender-prediction

Source

• HuggingFace: https://huggingface.co/rizvandwiki/gender-classification
• DockerHub: https://hub.docker.com/r/ikhwan17/gender-prediction