Image Classification
Image classification assigns labels to entire images, identifying what they primarily depict.
Hands-on Example: Classifying Images
from transformers import pipeline
from PIL import Image
import requests
from io import BytesIO
import matplotlib.pyplot as plt
# Initialize the image classification pipeline
image_classifier = pipeline("image-classification")
# Load images from URLs
image_urls = [
"https://upload.wikimedia.org/wikipedia/commons/thumb/d/d9/Collage_of_Nine_Dogs.jpg/800px-Collage_of_Nine_Dogs.jpg",
"https://upload.wikimedia.org/wikipedia/commons/thumb/4/4d/Cat_November_2010-1a.jpg/767px-Cat_November_2010-1a.jpg",
"https://upload.wikimedia.org/wikipedia/commons/thumb/9/91/F-15E_Strike_Eagle.jpg/800px-F-15E_Strike_Eagle.jpg"
]
# Classify each image
for url in image_urls:
# Load image
response = requests.get(url)
image = Image.open(BytesIO(response.content))
# Classify
results = image_classifier(image)
# Display top 3 predictions
print(f"Image: {url.split('/')[-1]}")
for result in results[:3]:
print(f"• {result['label']}: {result['score']:.4f}")
print("-" * 50)
The image classification pipeline predicts what’s depicted in the image, providing labels and confidence scores.
Try It Yourself:
- Classify images from your own collection by loading them from disk:
image = Image.open("path/to/image.jpg"). - Try different pre-trained models like
google/vit-base-patch16-224ormicrosoft/resnet-50by specifying them in the pipeline. - See how the model performs on ambiguous images or images with multiple subjects.