Spaces:
Sleeping
Sleeping
File size: 1,662 Bytes
edb3fc9 b38b7b0 edb3fc9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 |
import gradio as gr
import torch
from transformers import AutoTokenizer
from PIL import Image
from torchvision import transforms
# Load model and tokenizer
model = load_model(model_weights.pth)
model.eval()
text_tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
# Image transform pipeline
image_transform = transforms.Compose([
transforms.Resize((224, 224)),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
])
# Prediction function
def predict(image: Image.Image, text: str) -> str:
# Process text input
text_inputs = text_tokenizer(
text,
return_tensors="pt",
padding="max_length",
truncation=True,
max_length=512
)
# Process image input
image_input = image_transform(image).unsqueeze(0) # Add batch dimension
# Model inference
with torch.no_grad():
classification_output = model(
pixel_values=image_input,
input_ids=text_inputs["input_ids"],
attention_mask=text_inputs["attention_mask"]
)
predicted_class = torch.sigmoid(classification_output).round().item()
return "Biased" if predicted_class == 1 else "Unbiased"
# Gradio Interface
interface = gr.Interface(
fn=predict,
inputs=[
gr.Image(type="pil", label="Upload Image"),
gr.Textbox(lines=2, placeholder="Enter text for classification...", label="Input Text")
],
outputs=gr.Label(label="Prediction"),
title="Multimodal Bias Classifier",
description="Upload an image and provide a text to classify it as 'Biased' or 'Unbiased'."
)
interface.launch()
|