GhostNet 100
| Property | Value |
|---|---|
| Parameter Count | 5.2M |
| Model Type | Image Classification |
| License | Apache 2.0 |
| Paper | GhostNet: More Features from Cheap Operations |
| Input Size | 224 x 224 |
What is ghostnet_100.in1k?
GhostNet is an innovative lightweight neural network architecture designed for efficient image classification. It introduces the concept of "ghost" feature generation to reduce computational costs while maintaining high accuracy. Trained on ImageNet-1k, this model achieves competitive performance with just 5.2M parameters and 0.1 GMACs.
Implementation Details
The model leverages a unique architecture that generates more features using cheap operations. It operates on 224x224 pixel images and produces 3.5M activations during inference. The implementation is available through the timm library, making it easily accessible for both inference and feature extraction tasks.
- Efficient ghost module architecture
- Trained on ImageNet-1k dataset
- Supports feature map extraction at multiple scales
- Provides image embedding capabilities
Core Capabilities
- Image classification with 1000 classes
- Feature extraction at various network depths
- Embedding generation for downstream tasks
- Flexible interface through timm library
Frequently Asked Questions
Q: What makes this model unique?
GhostNet's uniqueness lies in its ghost module design, which generates more feature maps using lightweight operations. This approach significantly reduces computational costs while maintaining model effectiveness.
Q: What are the recommended use cases?
The model is ideal for resource-constrained environments where efficient image classification is needed. It's particularly suitable for mobile devices, edge computing, and applications requiring real-time inference with limited computational resources.
