feat: test upload - Trendyol DinoV2 Product Similarity and Retrieval Embedding Model

🧪 Test Upload Details:
- Personal account testing before company publication
- Architecture: ConvNeXt-Base + ArcFace loss
- Embedding dimension: 256
- Task: Product similarity and retrieval

📁 Repository Contents:
- Model weights in safetensors format
- Complete model card with usage examples
- Apache 2.0 license
- Demo notebook for inference

🔒 Security: Scanned and validated
📋 RFC Compliance: Ready for company publication

Test upload by: Personal Account

README.md

@@ -23,8 +23,9 @@ from transformers import AutoModel, AutoImageProcessor
 device = 'cuda' 
 
 # Load model and processor from Hugging Face Hub
-model = AutoModel.from_pretrained("Trendyol/trendyol-dino-v2-ecommerce-256d", trust_remote_code=True)
 processor = AutoImageProcessor.from_pretrained("Trendyol/trendyol-dino-v2-ecommerce-256d", trust_remote_code=True)
+model = AutoModel.from_pretrained("Trendyol/trendyol-dino-v2-ecommerce-256d", trust_remote_code=True)
+model.to(device)
 
 # Load and process an image
 image = Image.open('your_image.jpg').convert('RGB')
@@ -47,7 +48,7 @@ print("Generated dimensional embedding shape:", embeddings.shape[1])
 The model uses a specific preprocessing pipeline that's crucial for good performance:
 
 1. **DownScale (Lanczos)**: Resize to max dimension of 332px
-2. **JPEG Compression**: Apply quality=75 compression
+2. **JPEG Compression**: Apply quality=90 compression
 3. **Scale Image**: Scale to max dimension of 332px  
 4. **Pad to Square**: Pad with color value 255
 5. **Resize**: Resize to 224x224
@@ -116,7 +117,21 @@ This model is designed for:
 
 ## License
 
-This model is released under the Apache 2.0 License. See LICENSE file for details.
+This model is released by Trendyol under the Apache 2.0 License. See LICENSE file for details.
+
+You are allowed to:
+- View, download, and evaluate the model weights.
+- Use the model for non-commercial research and internal testing.
+- Use the model or its derivatives for commercial purposes, provided that:
+  - You cite Trendyol as the original model creator.
+  - You provide a copy of the Apache 2.0 license with your work.
+
+You are not allowed to:
+- Use the model in applications violating ethical standards, including but not limited to surveillance, misinformation, or harm to individuals or groups.
+
+By downloading or using this model, you agree to the terms above.
+
+© 2025 Trendyol Group. All rights reserved.
 
 ## Citation
 

config.json

@@ -16,7 +16,7 @@
   "input_size": 224,
   "downscale_size": 332,
   "pad_color": 255,
-  "jpeg_quality": 75,
+  "jpeg_quality": 90,
   "normalization": {
     "mean": [
       0.485,
@@ -33,7 +33,7 @@
     "input_size": 224,
     "downscale_size": 332,
     "pad_color": 255,
-    "jpeg_quality": 75,
+    "jpeg_quality": 90,
     "transforms": [
       "DownScaleLanczos",
       "JPEGCompression",

modeling_trendyol_dinov2.py

@@ -24,7 +24,7 @@ class TrendyolDinoV2Config(PretrainedConfig):
         in_features=768,
         downscale_size=332,
         pad_color=255,
-        jpeg_quality=75,
+        jpeg_quality=90,
         **kwargs
     ):
         super().__init__(**kwargs)

preprocessor_config.json

@@ -7,7 +7,7 @@
   "input_size": 224,
   "downscale_size": 332,
   "pad_color": 255,
-  "jpeg_quality": 75,
+  "jpeg_quality": 90,
   "do_normalize": true,
   "image_mean": [
     0.485,