Update app.py

app.py

@@ -1,46 +1,32 @@
-import torch
-from diffusers import AnimateDiffPipeline, LCMScheduler, MotionAdapter
-from diffusers.utils import export_to_gif
-from peft import PeftModel  # Import the PEFT library
+from huggingface_hub import snapshot_download
+from modelscope.pipelines import pipeline
+from modelscope.outputs import OutputKeys
+import pathlib
 import gradio as gr
 
-# Load AnimateLCM for video generation
-adapter = MotionAdapter.from_pretrained("Binarybardakshat/RCNA_MINI")
-pipe = AnimateDiffPipeline.from_pretrained("emilianJR/epiCRealism", motion_adapter=adapter, torch_dtype=torch.float16)
-pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config, beta_schedule="linear")
+# Download the model weights and prepare the model directory
+model_dir = pathlib.Path('weights')
+snapshot_download('damo-vilab/modelscope-damo-text-to-video-synthesis',
+                   repo_type='model', local_dir=model_dir)
 
-# Load LoRA weights using PEFT
-pipe = PeftModel.from_pretrained(pipe, "Binarybardakshat/RCNA_MINI", adapter_name="lcm-lora")  # PEFT for LoRA
+# Initialize the text-to-video synthesis pipeline
+pipe = pipeline('text-to-video-synthesis', model_dir.as_posix())
 
-pipe.set_adapters(["lcm-lora"], [0.8])
-pipe.enable_vae_slicing()
-pipe.enable_model_cpu_offload()
+# Define a function that takes a text prompt and generates a video
+def generate_video(text_prompt):
+    test_text = {'text': text_prompt}
+    output_video_path = pipe(test_text)[OutputKeys.OUTPUT_VIDEO]
+    return output_video_path
 
-# Function to generate video from user prompt
-def generate_video(prompt):
-    # Generate video using RCNA MINI
-    output = pipe(
-        prompt=prompt,
-        negative_prompt="bad quality, worse quality, low resolution",
-        num_frames=16,
-        guidance_scale=2.0,
-        num_inference_steps=6,
-        generator=torch.Generator("cpu").manual_seed(0),
-    )
-    frames = output.frames[0]
-    export_to_gif(frames, "animatelcm.gif")
-    
-    # Return the generated video
-    return "animatelcm.gif"
-
-# Create a Gradio interface
-interface = gr.Interface(
+# Set up Gradio interface
+demo = gr.Interface(
     fn=generate_video,
-    inputs="text",  # Accept text prompt
-    outputs="file",  # Output the generated GIF file
-    title="AnimateLCM Video Generator",
-    description="Generate videos with a simple prompt using AnimateLCM and RCNA MINI",
+    inputs=gr.Textbox(label="Enter a text prompt", placeholder="Describe the scene..."),
+    outputs=gr.Video(label="Generated Video"),
+    title="Text-to-Video Generator",
+    description="Enter a text description, and the model will generate a video based on your input.",
 )
 
-# Launch the Gradio app
-interface.launch()
+# Launch the Gradio interface
+if __name__ == "__main__":
+    demo.launch()