app.py

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch
import gradio as gr

model_name = "ljz512187207/Social_Media_Fake_News_Detection"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)
device = 'cuda' if torch.cuda.is_available() else 'cpu'
model.to(device)

def predict_fake(title, text):
    print("Title:", title)
    print("Text:", text)
    input_str = "<title>" + title + "<content>" + text + "<end>"
    input_ids = tokenizer.encode_plus(input_str, max_length=512, padding="max_length", truncation=True, return_tensors="pt")
    with torch.no_grad():
        output = model(input_ids["input_ids"].to(device), attention_mask=input_ids["attention_mask"].to(device))
    print("Output:", output)
    probabilities = torch.nn.Softmax(dim=-1)(output.logits)[0]
    print("Probabilities:", probabilities)
    return {
        "Fake": probabilities[0].item(),
        "Real": probabilities[1].item()
    }

iface = gr.Interface(
    fn=predict_fake,
    inputs=[
        gr.Textbox(lines=1, label="Title"),
        gr.Textbox(lines=6, label="Text")
    ],
    outputs="label"
)

iface.launch(share=True)