app.py

import torch
from transformers import BertTokenizer
from torch.nn.functional import softmax
from google.colab import drive
import gradio as gr

drive.mount('/content/drive')

# Set the correct path for the model within the Hugging Face Space
model = torch.load('/content/drive/My Drive/Emotion/emotion_model.pth')
tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")


model.eval()  # Set the model to evaluation mode
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)

def predict_emotions(text):
    inputs = tokenizer.encode_plus(text, return_tensors="pt", max_length=512, truncation=True, padding='max_length')
    input_ids = inputs['input_ids'].to(device)
    attention_mask = inputs['attention_mask'].to(device)

    with torch.no_grad():
        outputs = model(input_ids=input_ids, attention_mask=attention_mask)
    probabilities = softmax(outputs.logits, dim=-1).squeeze()
    emotions = ['Sadness', 'Joy', 'Love', 'Anger', 'Fear', 'Surprise']
    response = ", ".join(f"{emotion}: {prob * 100:.2f}%" for emotion, prob in zip(emotions, probabilities))
    return response

iface = gr.Interface(fn=predict_emotions, inputs="text", outputs="label")
iface.launch()