vim
实用的vim配置
作为一个工具控，平日里就喜欢配置一些好用的工具。俗话说，工欲善其事，必先利其器，而vim可以说就是这里面的尚方宝剑了。
话说我身边使用vim的人大致分为两种：第一类是由于特殊原因（比如在服务器）无法使用图形界面，所以被迫使用了这样的可以在命令汗工作的编辑工具；第二类是主动去学习这样
...
option-framework
Option framework in reinforcement learning参考自Learning Options in Reinforcement Learningoption 是在强化学习中的特殊的结构，它被设计在一些顶层任务中加入一些小的子目标的任务。同时一个option可以看成是一个
...
DFP_code
Code Structure
common
util.py
tf_ops.py
defaults.py
Simulator
doom_simulator.py
multi_doom_simulator.py
Agent
agent.py
future_predictor_agent_bas
...
OCR
OCR 文字光学识别问题描述
文字检测
文字识别
文字检测(localization)特点
IOU不是一个好的criterion, 检测到一部分文字也行
various fonts, colors, languages and bks etc.
perspective transformation
...
pickle
pickle 模块的简易使用序列化存储1234import pickledata = ...f = open('file.pkl','wb')pickle.dump(data, f)
序列化读取123import picklepkl_file = open('file.pkl','rb')data
...
tmux
TmuxDownload and InstallreferTmux
ShortcutsSessions
:new new session
s list sessions
$ name session
Windows
c create window
w list window
n next wind
...
RL7 Policy Gradient
reference: UCL Course on RL
lecture 7 Policy GradientIntroductionPolicy-Based Reinforcement Learning
In the last lecture we approximated the value
...
NLP3 More Word Vectors
referencecs224n
lecture 3 More Word VectorsStochastic gradients with word vectors
But in each window, we only have at most $2m+1$ words, so $\nabla_\t
...
NLP2 Word Vectors
referencecs224n
lecture 2 Word VectorsWord meaningDefinition: meaning
the idea that is represented by a word, phrase, etc.
the idea that a person wan
...
NLP1 Introduction to NLP and DL
referencecs224n
lecture 1 Introduction to NLP and DLNatural Language Processing (NLP)NLP is a field at the intersection of cs, ai and linguisticsGoal:
...