LeetCode Weekly Contest 114
Time: Dec 09th, 2018 @ 10:30 AM - 12:00 AM (GMT+8)
这是发表在 ICML 2018 上的一篇关于 MIL 的文章,文章内容其实挺简单,但是讲的很好,其中MIL的思想可以借鉴。
文章的开源代码在 AttentionDeepMIL 。
What is MURA?
MURA (musculoskeletal radiographs) is a large dataset of bone X-rays. Algorithms are tasked with determining whether an X-ray study is normal or abnormal.
这是斯坦福大学机器学习组的骨骼X光深度学习比赛(Bone X-Ray Deep Learning Competition),这是它的官网。
数据集是七类人体部位的X光,分为normal和abnormal两种,需要训练一个模型,能够完成这样一个二分类的诊断任务。
Baseline是一个169层的DenseNet,我们用PyTorch实现了它。
答案的提交是通过CodaLab这个网站,官方给出了这样一个提交教程:
Most datasets have a high number of dimensions along which the data is distributed. It is key to understand how to visualize high-dimensional datasets. This can be achieved using techniques known as dimensionality reduction. This post will focus on two techniques: PCA and t-SNE.
Firstly, let's get some high-dimensional data to work with.
偶然间在知乎上看到了一篇文章: 我眼中一个好的Pythoneer应该具备的品质,于是想思考一下这些问题的答案。
I'd like to talk about something about EM algorithm in my understanding.
This post is mainly based on Richard Xu's machine learning course.
Recurrent neural networks have connections that have loops, adding feedback and memory to the networks over time.
Python 的 Decorator 是一个非常有用的工具,比如Flask的路由(router),Decorator的目的是对已有的模块做一些修饰工作,使用方法就是在方法名前加上 '@XXX' 注解来为这个方法装饰一些东西。