Update app.py

app.py

@@ -2,47 +2,13 @@ import gradio as gr
 from PIL import Image
 import imagehash
 import hashlib
-import torch
-import torchvision.transforms as transforms
-from torchvision import models
-import numpy as np
 
 # -------------------------
 # MD5 HASH FUNCTION
 # -------------------------
-def get_md5(file):
-    md5 = hashlib.md5()
-    md5.update(file)
-    return md5.hexdigest()
-
-# -------------------------
-# dHash FUNCTION
-# -------------------------
-def get_dhash(image):
-    return imagehash.dhash(image)
-
-# -------------------------
-# LOAD SIMPLE MODEL (OPTIONAL - lightweight instead of full ViT)
-# -------------------------
-model = models.resnet18(pretrained=True)
-model.eval()
-
-transform = transforms.Compose([
-    transforms.Resize((224, 224)),
-    transforms.ToTensor()
-])
-
-def get_features(image):
-    img = transform(image).unsqueeze(0)
-    with torch.no_grad():
-        features = model(img)
-    return features.numpy()
-
-# -------------------------
-# SIMILARITY FUNCTION
-# -------------------------
-def cosine_similarity(a, b):
-    return np.dot(a, b.T) / (np.linalg.norm(a) * np.linalg.norm(b))
+def get_md5(file_path):
+    with open(file_path, "rb") as f:
+        return hashlib.md5(f.read()).hexdigest()
 
 # -------------------------
 # MAIN FUNCTION
@@ -53,45 +19,32 @@ def find_duplicates(files):
 
     md5_map = {}
     dhash_map = {}
-    features_list = []
     results = []
 
-    images = []
-
+    # Process images
     for file in files:
-        img = Image.open(file.name).convert("RGB")
-        images.append((file.name, img))
+        file_path = file.name
 
-        # MD5
-        file_bytes = open(file.name, "rb").read()
-        md5 = get_md5(file_bytes)
+        # Load image
+        img = Image.open(file_path).convert("RGB")
 
+        # MD5 (Exact duplicates)
+        md5 = get_md5(file_path)
         if md5 in md5_map:
-            results.append(f"Exact Duplicate: {file.name} == {md5_map[md5]}")
+            results.append(f"Exact Duplicate: {file_path} == {md5_map[md5]}")
         else:
-            md5_map[md5] = file.name
+            md5_map[md5] = file_path
 
-        # dHash
-        dh = get_dhash(img)
-        dhash_map[file.name] = dh
+        # dHash (Similar images)
+        dhash_map[file_path] = imagehash.dhash(img)
 
-        # Features
-        features = get_features(img)
-        features_list.append((file.name, features))
-
-    # dHash comparison
-    names = list(dhash_map.keys())
-    for i in range(len(names)):
-        for j in range(i+1, len(names)):
-            if dhash_map[names[i]] - dhash_map[names[j]] < 5:
-                results.append(f"Similar (dHash): {names[i]} ~ {names[j]}")
-
-    # Feature similarity
-    for i in range(len(features_list)):
-        for j in range(i+1, len(features_list)):
-            sim = cosine_similarity(features_list[i][1], features_list[j][1])
-            if sim > 0.8:
-                results.append(f"Near Duplicate (AI): {features_list[i][0]} ~ {features_list[j][0]}")
+    # Compare dHash
+    file_names = list(dhash_map.keys())
+    for i in range(len(file_names)):
+        for j in range(i + 1, len(file_names)):
+            diff = dhash_map[file_names[i]] - dhash_map[file_names[j]]
+            if diff < 5:
+                results.append(f"Similar Image: {file_names[i]} ~ {file_names[j]}")
 
     if not results:
         return "No duplicates found"
@@ -106,7 +59,10 @@ interface = gr.Interface(
     inputs=gr.File(file_count="multiple", label="Upload Images"),
     outputs="text",
     title="Image Duplicate Finder",
-    description="Upload images to find exact and near duplicates using MD5, dHash, and AI"
+    description="Upload images to detect exact and similar duplicates using MD5 and dHash"
 )
 
-interface.launch()
+# -------------------------
+# LAUNCH (IMPORTANT FOR HF)
+# -------------------------
+interface.launch(server_name="0.0.0.0", server_port=7860, ssr_mode=False)