论文地址：https://openaccess.thecvf.com/content_ICCV_2019/html/Zheng_DeepHuman_3D_Human_Reconstruction_From_a_Single_Image_ICCV_2019_paper.html

作者：Zerong Zheng, Tao Yu, Yixuan Wei, Qionghai Dai, Yebin Liu

发表： ICCV2019

Code： https://github.com/ZhengZerong/DeepHuman

论文地址：https://link.springer.com/article/10.1631%2FFITEE.1910001

作者：潘云鹤

发表： Frontiers of Information Technology & Electronic Engineering

链接： https://link.springer.com/article/10.1631%2FFITEE.1910001

Heeeeyyyyy!!! I customized a theme called Evangelion for Typora Markdown Editor, and it was accepted! Do check it on the Typora - Theme official page:

And the theme's homepage is here: https://github.com/viridityzhu/Evangelion-typora-theme

11月平均每日阅读43分钟，最高一天阅读4小时，总共21个小时。这意味着，这30天里，我大约有一整天的人生活在书里的世界。幸甚至哉！

这些时间最大头还是花在了《奇鸟行状录》里（不太值），其次应该是钱钟书的《宋诗选注》和黑格尔的《小逻辑》里。看上去好奇怪的组合，不过不然。这其实很符合我的阅读习惯，一般是不同类型的书同时阅读，每次依当天的状态和心情去择一本翻开。实际上，想读《小逻辑》的时候，和想读《奇鸟行状录》的时候，整个人的状态是完全不一样的。据我观察，当我精神状态极佳、心情大好的时候，最想读哲学类的书；情感泛滥（高涨或者低落）的时候，最想读诗；感到疲乏的时候，想读小说来消遣。我倒也根据这么一点小习性，大言不惭地自觉“随性”，除了读书随心所欲以外，生活中大大小小的事上，心态也大抵类似。（主要是小事，大事实际上还是有点优柔寡断的，还想改呢。）

不过呢，这个月挑来读的这几本书，跟我之前的阅读惯性相比还是比较突兀的。一方面，小说上，我已经挺久没读日本小说了。虽然以前也喜欢村上春树，但这次是带着一些“怀旧”的心情去读他的。根据我的规划，在这个人生阶段，我是想读一些陀氏和海明威的，再往后也想涉猎一些不太熟悉但十分严肃的作家。

我认为隐喻该是为了更清楚、更深刻的表达而存在的，而这本书里的大量隐喻根本无益于表达。

互联网冲浪对于我来说最浪漫的事情之一，就是顺着一篇搜来的技术文档，随手点开博客主页时，发现博客中记满了朴素但真挚的生活感想。有哲思，有小故事，有少年得意的成就记叙，也有人到中年的感怀伤时……每个人的博客都是独一无二的，网站名字和独特装饰极力彰显着个性；但所有博客又有一个共同的特点，就是毫无保留的文字和期待被造访的心情。这就好像在漠然的大城市里，却有一些不起眼的小屋子，真诚地敞着所有的门窗，等待像我这样的，可能会来，也可能不会来的旅人。

In IT5003, I am studying data structures and algorithms. The language we learn with is Python. So I take notes about what I learned, and how to use them in Python. I struct my notes based on data structures, and all the ADTs that are implemented by the data structures are discussed within. I focus on their usage, key design, and sometimes problems. In the future maybe I will append all the implementation codes.

• ADT: Abstract data type, from the point of view of a user, defined by its behavior.
• Data Structure: from the point of view of an implementer, concrete representations of data.

Linear Data Structures

Searching and Sorting

Searching algorithms

Search things in a list:

• if the list doesn't change frequently, better to sort first, then use binary-search everytime.

Function

Function Decorator

1
2
3
4
5
6
7
8
9
10
11

def track(fn):
    def h(x):
        print('the fn {} has x {}'.format(fn, x))
        return fn(x)
    return h

@track
def foo(x):
    return x ** x

print(foo(3))

decorator equals to: foo = track(foo)

Iterable/Sequences

Strings, ranges, lists and tuples.

“An iterable which supports efficient element access using integer indices via the __getitem__ special method and defines a __len__ method that returns the length of the sequence.”

• use one index to fetch item in a list, the index can't out of the range; but if use slice method, then you can do it out of the range, because the slice will autolly find the greediest indexs.

• assigning a list to another list, like a = b, makes a and b refers to the same list. And a list is mutable. So if we change a, then b will also be affected. It's called alias. Do avoid this.

• Tuple also has order. Just it's immutable.

• Fxxxk, Tuple and Set are different things! A set is like {1, 2, 3}, that is what I thought as not ordered and not duplicated. And a set has intersection() union() difference() symmetric_difference().

  To delete items in a set:

  1 2 3

  set1.discard(6) # If the item doesn't exist, no exception set1.remove(6) # Raise an exception if doesn't exist set1.pop() # randomly delete one item.

• you can insert an interable in a list's slice, though they have different length!

  1 2 3 4

  >>> a = [1, 2, 3, 4, 5] >>> a[3:3] = [6, 7] >>> a [1, 2, 3, 6, 7, 4, 5]

• list() takes an iterable, and makes it a list;

  str() shows you the whole object's look.

• range() is also an iterable. It can be slicing, and len()

• a, b = b, a: the b, a on the right firstly is packed into tuple: (b, a), then it is unpacked and individually assigned to each var on the left