from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
import gradio as gr


model = AutoModelForSeq2SeqLM.from_pretrained("PRAli22/flan-t5-base-imdb-text-classification")
tokenizer = AutoTokenizer.from_pretrained("PRAli22/flan-t5-base-imdb-text-classification")



def summarize(text):
    inputs = tokenizer.encode_plus(text, padding='max_length', max_length=512, return_tensors='pt')
    summarized_ids = model.generate(inputs['input_ids'], attention_mask=inputs['attention_mask'], 
                                    max_length=150, num_beams=4, early_stopping=True)

    return tokenizer.decode(summarized_ids[0], skip_special_tokens=True)
    
                    
    


css_code='body{background-image:url("https://media.istockphoto.com/id/1256252051/vector/people-using-online-translation-app.jpg?s=612x612&w=0&k=20&c=aa6ykHXnSwqKu31fFR6r6Y1bYMS5FMAU9yHqwwylA94=");}'

demo = gr.Interface(
    fn=summarize,
    inputs=
        gr.Textbox(label="text", placeholder="Enter the text "),
    
    outputs=gr.Textbox(label="sentiment"),
    title="Sentiment Classifier",
    description= "This is Sentiment Classifier, it takes a text  in English as inputs and returns the sentiment",
    css = css_code
)

demo.launch()