hello

app.py

@@ -1,6 +1,7 @@
 import gradio as gr
 import torch
 import clip
+from PIL import Image, ImageEnhance
 
 device = "cuda" if torch.cuda.is_available() else "cpu"
 model, preprocess = clip.load("ViT-B/32", device=device)
@@ -10,6 +11,9 @@ def predict(image):
     labels = "Japanese, Chinese, Roman, Greek, Etruscan, Scandinavian, Celtic, Medieval, Victorian, Neoclassic, Romanticism, Art Nouveau, Art deco"
     labels = labels.split(',')
 
+    converter = ImageEnhance.Color(image)
+    image = converter.enhance(0.5)
+    image = image.convert("L")
     image = preprocess(image).unsqueeze(0).to(device)
     text = clip.tokenize([f"a character of origin {c}" for c in labels]).to(device)
 

requirements.txt

@@ -1,3 +1,4 @@
 git+https://github.com/openai/CLIP
 torch
-Jinja2
+Jinja2
+Pillow