Vit Food101 Int8

🚀 Vision Transformer (ViT)

This is a Vision Transformer (ViT) model that has been quantized and exported to the OpenVINO Intermediate Representation (IR). It is useful for image - classification tasks, especially in the context of food images.

🚀 Quick Start

This ViT model, fine - tuned on the Food - 101 dataset, has been statically quantized and exported to the OpenVINO IR using optimum.

✨ Features

• Quantization: The model is statically quantized to optimize performance.
• OpenVINO IR Export: It is exported to the OpenVINO Intermediate Representation, enabling efficient inference on Intel hardware.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

from transformers import pipeline, AutoFeatureExtractor
from optimum.intel.openvino import OVModelForImageClassification

model_id = "echarlaix/vit-food101-int8"
model = OVModelForImageClassification.from_pretrained(model_id)
feature_extractor = AutoFeatureExtractor.from_pretrained(model_id)
pipe = pipeline("image-classification", model=model, feature_extractor=feature_extractor)
outputs = pipe("http://farm2.staticflickr.com/1375/1394861946_171ea43524_z.jpg")

📚 Documentation

Model Details

Model Description: This ViT model fine - tuned on Food - 101 was statically quantized and exported to the OpenVINO IR using optimum.

Information Table

📄 License

This project is licensed under the Apache - 2.0 license.