Update requirements.txt

requirements.txt

@@ -1,17 +1,58 @@
---extra-index-url https://download.pytorch.org/whl/cu113
-torch
-torchvision==0.13.1+cu113
-diffusers
-#git+https://github.com/huggingface/diffusers.git
-transformers
-#git+https://github.com/huggingface/transformers
-scipy
-ftfy
-psutil
-accelerate==0.12.0
-#OmegaConf
-#pytorch_lightning
-#triton==2.0.0.dev20220701
-triton
-#https://github.com/apolinario/xformers/releases/download/0.0.3/xformers-0.0.14.dev0-cp38-cp38-linux_x86_64.whl
-https://github.com/camenduru/stable-diffusion-webui-colab/releases/download/0.0.15/xformers-0.0.15.dev0+4c06c79.d20221205-cp38-cp38-linux_x86_64.whl
+import torch
+from diffusers import StableDiffusionPipeline, DPMSolverMultistepScheduler
+from PIL import Image
+import gradio as gr
+import random
+import utils
+
+class Model:
+    def __init__(self, name, path="", prefix=""):
+        self.name = name
+        self.path = path
+        self.prefix = prefix
+
+class DiffusionApp:
+    def __init__(self, models):
+        self.models = models
+        self.current_model = self.models[0]
+        self.current_steps = 25
+        self.is_colab = utils.is_google_colab()
+        self.device = "cuda" if torch.cuda.is_available() else "cpu"
+        self.pipe = self.initialize_pipe(self.current_model.path)
+
+    def initialize_pipe(self, model_path):
+        pipe = StableDiffusionPipeline.from_pretrained(
+            model_path,
+            torch_dtype=torch.float16,
+            scheduler=DPMSolverMultistepScheduler.from_pretrained(model_path, subfolder="scheduler")
+        )
+        pipe = pipe.to(self.device)
+        if self.device == "cuda":
+            pipe.enable_xformers_memory_efficient_attention()
+        return pipe
+
+    def update_model(self, model_name):
+        for model in self.models:
+            if model.name == model_name:
+                self.current_model = model
+                self.pipe = self.initialize_pipe(self.current_model.path)
+
+    def inference(self, model_name, prompt, guidance, steps, n_images=1, width=512, height=512, seed=0, img=None, strength=0.5, neg_prompt=""):
+        self.update_model(model_name)
+        # Rest of your inference code here...
+
+# Initialize your models
+models = [
+    Model("Vivid Watercolors", "Evel/VividWatercolors", "watercolor style"),
+    Model("Loving Vincent (Van Gogh)", "dallinmackay/Van-Gogh-diffusion", "lvngvncnt "),
+    # Add the rest of your models here...
+]
+
+# Initialize your app
+app = DiffusionApp(models)
+
+# Use your app in your Gradio interface
+# For example:
+inputs = [gr.Dropdown(label="Model", choices=[m.name for m in app.models], value=app.current_model.name)]
+outputs = [gr.Image(height=512)]
+gr.Interface(app.inference, inputs, outputs).launch()