resnet50-cars

ResNet50 - Stanford Cars Classification

This repository contains a pre-trained ResNet50 model, fine-tuned to classify vehicle makes, models, and years using the Stanford Cars dataset. This model is maintained and provided by ZEDEDA.

---

📊 Model Details

• Architecture: ResNet50
• Framework: ONNX
• Precision: FP32 (Single Precision)
• Model Size: 94 MB
• Task: Image Classification
• Number of Classes: 196

---

📈 Performance

The model was evaluated on the Stanford Cars validation split and achieved the following performance metrics:

• Top-1 Accuracy: 89.39%

---

🗂️ Dataset Information

This model was trained on the Stanford Cars Dataset.

• The dataset contains 16,185 images of 196 classes of cars.
• The data is split into 8,144 training images and 8,041 testing images.
• Classes are typically at the level of Make, Model, Year (e.g., 2012 Tesla Model S or 2012 BMW M3 coupe).

---

🚀 Usage

Because the model is saved in the ONNX format, it can be easily deployed using onnxruntime in various environments.

Prerequisites

pip install onnxruntime numpy Pillow