import gradio as gr
import torch
from transformers import GPT2Tokenizer, GPT2LMHeadModel, TextDataset, DataCollatorForLanguageModeling
from transformers import Trainer, TrainingArguments
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split

__checkpoint = "gpt2"
__tokenizer = GPT2Tokenizer.from_pretrained(__checkpoint)
__model = GPT2LMHeadModel.from_pretrained(__checkpoint)
__model_output_path = "paavansundar/Medical_QNA_GPT2"

#prepare data
def prepareData():
    df=pd.read_csv("paavansundar/medquadqna/MedQuAD.csv")
    
# Create a Data collator object
data_collator = DataCollatorForLanguageModeling(tokenizer=__tokenizer, mlm=False, return_tensors="pt")
def queryGPT(question):
    return generate_response(__model, __tokenizer, question)

def generate_response(model,tokenizer, prompt, max_length=200):
    
    prepareData()
    input_ids = tokenizer.encode(prompt, return_tensors="pt")      # 'pt' for returning pytorch tensor

    # Create the attention mask and pad token id
    attention_mask = torch.ones_like(input_ids)
    pad_token_id = tokenizer.eos_token_id

    output = model.generate(
        input_ids,
        max_length=max_length,
        num_return_sequences=1,
        attention_mask=attention_mask,
        pad_token_id=pad_token_id
    )

    return tokenizer.decode(output[0], skip_special_tokens=True)
    
with gr.Blocks() as demo:

    txt_input = gr.Textbox(label="Input Question", lines=2)
    txt_output = gr.Textbox(value="", label="Answer")
    btn = gr.Button(value="Submit")
    btn.click(queryGPT, inputs=[txt_input], outputs=[txt_output])
if __name__ == "__main__":
    demo.launch()