RKSHT Harassment Detection Model

Model Description

This is a custom Vision Transformer (ViT) model fine-tuned for detecting instances of harassment in public and workplace environments. The model is built on google/vit-base-patch16-224-in21k and trained on a dataset tailored for harassment detection, classifying images into 'harassment' or 'non-harassment' categories.

Intended Use

This model is designed for use in applications requiring harassment detection through visual data, including:

• Workplace and public safety monitoring
• Real-time CCTV surveillance
• Automated alert systems

Model accuracy

The RKSHT model has been fine-tuned with high accuracy for distinguishing harassment behavior.

How to Use

Here’s an example of how to use the RKSHT Harassment Detection model for image classification:

import torch
from transformers import ViTForImageClassification, ViTFeatureExtractor
from PIL import Image

# Load the model and feature extractor
model = ViTForImageClassification.from_pretrained('Binarybardakshat/RKSHT')
feature_extractor = ViTFeatureExtractor.from_pretrained('Binarybardakshat/RKSHT')

# Load an image
image = Image.open('image.jpg')

# Preprocess the image
inputs = feature_extractor(images=image, return_tensors="pt")

# Perform inference
with torch.no_grad():
    outputs = model(**inputs)
    logits = outputs.logits
    predicted_class_idx = logits.argmax(-1).item()

# Print the predicted class
print("Predicted class:", model.config.id2label[predicted_class_idx])