A data scientist describes a simple yet effective custom embedding technique for image defect detection, outperforming a complex deep learning model on the KolektorSDD2 dataset.
A tutorial on fine-tuning a Vision Transformer (ViT) model for satellite image classification using Hugging Face Transformers and Keras.
A team built a handwritten sign digitizer for Hack Zurich 2016, creating a custom dataset and training a random forest image classifier in one day.
A tutorial on implementing image classification in a React Native app using TensorFlow.js and the MobileNet pre-trained model.
A tutorial on building a React Native app that uses Google's Vision API to classify images as 'hotdog' or 'not hotdog', inspired by HBO's Silicon Valley.
Explores why modern neural networks succeed where older ones failed, emphasizing the critical role of massive computational power and data size.
