base/app.py

import gradio as gr
from text_to_video import model_t2v_fun,setup_seed
from omegaconf import OmegaConf
import torch
import imageio
import os
import cv2
import torchvision
config_path = "./base/configs/sample.yaml"
args = OmegaConf.load("./base/configs/sample.yaml")
device = "cuda" if torch.cuda.is_available() else "cpu"
# ------- get model ---------------
model_t2V = model_t2v_fun(args)
model_t2V.to(device)
if device == "cuda":
    model_t2V.enable_xformers_memory_efficient_attention()

# model_t2V.enable_xformers_memory_efficient_attention()
css = """
h1 {
  text-align: center;
}
#component-0 {
  max-width: 730px;
  margin: auto;
}
"""

def infer(prompt, seed_inp, ddim_steps):


    setup_seed(seed_inp)
    videos = model_t2V(prompt, video_length=16, height = 320, width= 512, num_inference_steps=ddim_steps, guidance_scale=7).video
    print(videos[0].shape)
    if not os.path.exists(args.output_folder):
        os.mkdir(args.output_folder)
    torchvision.io.write_video(args.output_folder + prompt.replace(' ', '_') + '-.mp4', videos[0], fps=8)
    # imageio.mimwrite(args.output_folder + prompt.replace(' ', '_') + '.mp4', videos[0], fps=8)
    # video = cv2.VideoCapture(args.output_folder + prompt.replace(' ', '_') + '.mp4')
    # video = imageio.get_reader(args.output_folder + prompt.replace(' ', '_') + '.mp4',  'ffmpeg')


    # video = model_t2V(prompt, seed_inp, ddim_steps)

    return args.output_folder + prompt.replace(' ', '_') + '-.mp4'

print(1)

# def clean():
#     return gr.Image.update(value=None, visible=False), gr.Video.update(value=None)
def clean():
    return gr.Video.update(value=None)

title = """
    <div style="text-align: center; max-width: 700px; margin: 0 auto;">
        <div
        style="
            display: inline-flex;
            align-items: center;
            gap: 0.8rem;
            font-size: 1.75rem;
        "
        >
        <h1 style="font-weight: 900; margin-bottom: 7px; margin-top: 5px;">
            Intern·Vchitect (Text-to-Video)
        </h1>
        </div>
        <p style="margin-bottom: 10px; font-size: 94%">
        Apply Intern·Vchitect to generate a video 
        </p>
    </div>
"""

# print(1)
with gr.Blocks(css='style.css') as demo:
    gr.Markdown("<font color=red size=10><center>LaVie</center></font>")
    with gr.Row(elem_id="col-container"):
 
        with gr.Column():
                
            prompt = gr.Textbox(value="a teddy bear walking on the street", label="Prompt", placeholder="enter prompt", show_label=True, elem_id="prompt-in", min_width=200, lines=2)
            
            ddim_steps = gr.Slider(label='Steps', minimum=50, maximum=300, value=50, step=1)
            seed_inp = gr.Slider(label="Seed", minimum=0, maximum=2147483647, step=1, value=400, elem_id="seed-in")
            # with gr.Row():
            #     # control_task = gr.Dropdown(label="Task", choices=["Text-2-video", "Image-2-video"], value="Text-2-video", multiselect=False, elem_id="controltask-in")
            #     ddim_steps = gr.Slider(label='Steps', minimum=50, maximum=300, value=250, step=1)
            #     seed_inp = gr.Slider(label="Seed", minimum=0, maximum=2147483647, step=1, value=123456, elem_id="seed-in")
            
            # ddim_steps = gr.Slider(label='Steps', minimum=50, maximum=300, value=250, step=1)
        with gr.Column():
            submit_btn = gr.Button("Generate video")
            clean_btn = gr.Button("Clean video")
        # submit_btn = gr.Button("Generate video", size='sm')
            # video_out = gr.Video(label="Video result", elem_id="video-output", height=320, width=512)
            video_out = gr.Video(label="Video result", elem_id="video-output")
        # with gr.Row():
        #     video_out = gr.Video(label="Video result", elem_id="video-output", height=320, width=512)
        #     submit_btn = gr.Button("Generate video", size='sm')
        

            # video_out = gr.Video(label="Video result", elem_id="video-output", height=320, width=512)
        inputs = [prompt, seed_inp, ddim_steps]
        outputs = [video_out]

        
    # control_task.change(change_task_options, inputs=[control_task], outputs=[canny_opt, hough_opt, normal_opt], queue=False)
    # submit_btn.click(clean, inputs=[], outputs=[video_out], queue=False)
    clean_btn.click(clean, inputs=[], outputs=[video_out], queue=False)
    submit_btn.click(infer, inputs, outputs)
    # share_button.click(None, [], [], _js=share_js)

    print(2)
demo.queue(max_size=12).launch(server_name="0.0.0.0", server_port=7860)