import re
import os
import requests
import gradio as gr
from datasets import load_dataset
from PIL import Image
from io import BytesIO
import torch
from torch import autocast

from transformers import pipeline, set_seed
from diffusers import DiffusionPipeline, StableDiffusionPipeline


# Config
DEVICE = "cuda"


# GPT2
def get_gpt2_pipeline():
  generator = pipeline('text-generation', model='gpt2')
  set_seed(42)

  # generator("Hello world, I'm vizard,", max_length=50, num_return_sequences=3)
  
  return generator
  
# Text Summarizer
def get_text_summarizer_pipeline():
  summarizer = pipeline("summarization", model="facebook/bart-large-cnn")

  # generator("Hello world, I'm vizard,", max_length=50, num_return_sequences=3)
  
  return summarizer 

# SD v1.4
def get_stable_diffusion_v14_pipeline():
  model_id = "CompVis/stable-diffusion-v1-4"
  pipe = StableDiffusionPipeline.from_pretrained(model_id, use_auth_token=True)
  # pipeline = StableDiffusionPipeline.from_pretrained(model_id, use_auth_token=True, revision="fp16", torch_dtype=torch.float16)
  pipe = pipe.to(DEVICE)
  torch.backends.cudnn.benchmark = True
  return pipe

# SD v1.5
def get_stable_diffusion_v15_pipeline():
  model_id = "runwayml/stable-diffusion-v1-5"
  pipe = DiffusionPipeline.from_pretrained(mode_id)
  pipe = pipe.to(DEVICE)
  return pipe

def get_image(url):
  response = requests.get(url)
  image = Image.open(BytesIO(response.content)).convert("RGB")
  resized_image = image.resize((768, 512))
  return resized_image

# main
def main():
  prompt = "Hello world, I'm vizard,"
  
  gpt2_pipe = get_gpt2_pipeline()
  def greet(prompt):
    return gpt2_pipe(prompt, max_length=1000, num_return_sequences=3)
  
  with gr.Blocks() as ui:
    with gr.Row():
        with gr.Column():
            gpt_int = gr.Interface(
              fn=greet,
              inputs=gr.Textbox(lines=2, placeholder="Enter some text here..."),
              outputs="text",
              title="GPT2",
              description="OneDesc",
            )
    with gr.Row():
        with gr.Column():
            gpt_int2 = gr.Interface(
              fn=greet,
              inputs=gr.Textbox(lines=2, placeholder="Enter some text here..."),
              outputs="text",
              title="GPT2",
              description="OneDesc",
            )
    gr.Examples(['one.png', 'two.png', 'three.jpeg'])
    
  
  # ui = gr.Interface.from_pipeline(
  #    get_text_summarizer_pipeline(),
  #    title="OneTitle",
  #    description="OneDesc",
  #    examples=['one.png', 'two.png', 'three.jpeg'],
  #  )
  ui.launch(enable_queue=True)
  
  # pipe = pipeline(task="image-classification", model="microsoft/dit-base-finetuned-rvlcdip")
  
  #gr.Interface.from_pipeline(
  #  pipe,
  #  title="OneTitle",
  #  description="OneDescription",
  #  examples=['one.png', 'two.png', 'three.jpeg'],
  # ).launch()
  
  # pipe2 = get_stable_diffusion_v15_pipeline()
  # images = pipe2(prompt).images

main()