cs231n 笔记

自己随便记的，可读性较差。

Lecture2：Image Classification pipeline

def classify_image(image):

      # some magic here?

       return class_label

Data-Driven Approach

1. Collect a dataset of images and labels
2. Use Machine Learning to train a classifier
3. Evaluate the classifier on new images

Example Dataset:CIFAR10

       10 classes

       50 000 training images

       10 000 testing images

Distance Metric to compare images

       L1 distance:曼哈顿

K-Nearest Neighbors

       Instead of copying label from nearest neighbor, take majority vote from K closest points.

Distance Metric

L1(Manhattan)distance曼哈顿距离 L2(Euclidean)distance欧式距离

Hyperparameters

       What is the best value of k to use?

       What is the best distance to use?

       These are hyperparameters: choices about the algorithm that we set rather than learn.

 

作业一：knn

参考：https://blog.csdn.net/qq_28448117/article/details/79399959

           https://www.cnblogs.com/danscarlett/p/9402469.html

遇到的问题：Python中Import Error: no module named 'past'错误以及解决方法

解决办法：https://blog.csdn.net/qq_31282773/article/details/78672584

转载于:https://www.cnblogs.com/woshizhizhang/p/11338422.html