![](https://img-blog.csdnimg.cn/20201014180756927.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
Machine Learning
iamzhangzhuping
这个作者很懒,什么都没留下…
展开
-
ROC曲线和PR(Precision-Recall)曲线的联系
http://www.chmod764sean.com/computer-tec/data-mining-machine-learning/501.html在机器学习中,ROC(Receiver Operator Characteristic)曲线被广泛应用于二分类问题中来评估分类器的可信度,但是当处理一些高度不均衡的数据集时,PR曲线能表现出更多的信息,发现更多的问题。1.R转载 2016-02-22 13:22:02 · 4271 阅读 · 0 评论 -
Faster-rcnn
最后更新日期:2016年4月29日本教程主要基于python版本的faster R-CNN,因为python layer的使用,这个版本会比matlab的版本速度慢10%,但是准确率应该是差不多的。目前已经实现的有两种方式:Alternative training Approximate joint training 推荐使用第二种,因为第二种使用的显存更小,而且训练会更快,同时准确率差不多甚至原创 2016-05-08 12:33:24 · 2910 阅读 · 0 评论 -
sigmoid cross entorpy loss
1.Cross Entropy Error The mathematics behind cross entropy (CE) error and its relationship to NN training are very complex, but, fortunately, the results are remarkably simple to understand and implem原创 2016-05-07 09:58:34 · 2329 阅读 · 0 评论 -
Multinomial Logistic Loss and Cross Entropy Loss are the same
Multinomial Logistic Loss and Cross Entropy Loss are the same转载 2016-05-20 10:06:37 · 2840 阅读 · 0 评论 -
What is zero-shot learning?
Zero-shot learning is being able to solve a task despite not having received any training examples of that task. For a concrete example, imagine recognizing a category of object in photos without ever原创 2016-07-13 14:38:38 · 2457 阅读 · 0 评论