作者：禅与计算机程序设计艺术
1.简介
Breast cancer is a type of cancer that develops in the lining of the male breast. It mainly affects women and usually manifests as a dark sclerotic pattern. The early detection of breast cancer can prevent its harmful effects and reduce the risk of further complications such as cystitis, ulcers or blood clots. Therefore, accurate diagnosis of breast cancer at an early stage could lead to improved survival rate and improve life quality of patients. In this article, we will use deep learning algorithms with convolutional neural networks (CNN) to classify whether a human has breast cancer or not based on various features such as skin pixels, nipple size, and cell shape. We will also explore how CNN works under the hood by visualizing the filters learned by the model during training. Finally, we will evaluate our models using different evaluation metrics like accuracy, precision, recall, F1-score, ROC curve, AUC score and confusion