图像数据处理（0）

斯坦福李飞飞-深度学习计算机视觉笔记

说明：http://study.163.com/course/courseMain.htm?courseId=1003223001

Image Classification pipeline

1，Images are represented as 3D arrays of numbers,with integers between [0,255]，比如说300 x 100 x 3(3 for 3 color channels RGB)
2，Challenges：Viewpoint Variation【视角不同】，Illumination【照明】，Deformation【变形】，Occlusion，Background clutter【光影等干扰】，Intraclass variation【同类内的差异】
3，Data-driven approach:
- Collect a dataset of images and labels
- Use Machine Learning to train an image classifier
- Evaluate the classifier on a withheld set of test images
4，cross-validation
We saw that the choice of distance and the value of k are hyperparameters that are tuned using a validation set,or through cross-validation if the size of the data is small。

Loss functions and Optimization

Loss functions

Backpropagation and Neural Networks

backpropagation = recursive application of the chain rule along a
computational graph to compute the gradients of all
inputs/parameters/intermediates
推导：反向传导算法

Training Neural Networks

Activation Functions