papering - 博客园

2017年10月11日

Optimizer in SQL - Catalyst Optimizer in Spark SQL

摘要： SELECT sum(v) FROM ( SELECT score.id, 100+80+score.math_score+ score.english_score AS v FROM people JOIN score WHERE people.id = score.id AND people.a 阅读全文

posted @ 2017-10-11 20:26 papering 阅读(215) 评论(0) 推荐(0)

社交网络营销对象选择

摘要：阅读全文

posted @ 2017-10-11 19:17 papering 阅读(144) 评论(0) 推荐(0)

Louvain Modularity Fast unfolding of communities in large networks

摘要： Louvain Modularity Fast unfolding of communities in large networks https://arxiv.org/pdf/0803.0476.pdf 【一些算法的思路都很简单；但是数学翻译是严谨的】【0-全局模块度,原始模块度，全局目标函阅读全文

posted @ 2017-10-11 17:34 papering 阅读(501) 评论(0) 推荐(0)

PCA tries to preserve linear structure, MDS tries to preserve global geometry, and t-SNE tries to preserve topology (neighborhood structure)

摘要： https://colah.github.io/posts/2014-10-Visualizing-MNIST/ 阅读全文

posted @ 2017-10-11 16:37 papering 阅读(170) 评论(0) 推荐(0)

维度建模高效性

摘要：阅读全文

posted @ 2017-10-11 13:07 papering 阅读(155) 评论(0) 推荐(0)

chunk

摘要：【为了寻找chunk 文件块】为了寻找某个文件的文件块，存在另一个称为master node 主节点或name node 名字节点的小文件为该文件所用。阅读全文

posted @ 2017-10-11 00:59 papering 阅读(392) 评论(0) 推荐(0)

机架故障

摘要：单节点故障单机架故障阅读全文

posted @ 2017-10-11 00:47 papering 阅读(222) 评论(0) 推荐(0)

分布式底层上层

摘要：分布式底层的一方面：软件栈下层的文件系统分布式上层的一种形式：SQL的某种实现阅读全文

posted @ 2017-10-11 00:42 papering 阅读(243) 评论(0) 推荐(0)

power law 幂定律

摘要： y=cx∧a 卖品销量排名与销量阅读全文

posted @ 2017-10-11 00:23 papering 阅读(179) 评论(0) 推荐(0)

2017年10月10日

visio_action

摘要： bug 沟通效率不能为负值！！ https://blogs.office.com/en-us/2012/11/05/containers-and-callouts-in-visio/?eu=true visio-document——subprocess callouts 【start/end dec 阅读全文

posted @ 2017-10-10 17:35 papering 阅读(249) 评论(0) 推荐(0)

Asynchronous_method_invocation 异步方法调用让步 yielding

摘要： zh.wikipedia.org/wiki/同步【同步不同事件发生时间一致】同步（英语：Synchronization），指在一个系统中所发生的事件（event），之间进行协调，在时间上出现一致性与统一化的现象。在系统中进行同步，也被称为及时（in time）、同步化的（synchronous 阅读全文

posted @ 2017-10-10 01:08 papering 阅读(359) 评论(0) 推荐(0)

Python解释器是单线程应用 IO 密集型计算密集型 GIL global interpreter lock 线程状态和全局解释器锁引用计数内存管理纤程安全原子

摘要： Frequently Asked Questions — PyPy documentation https://doc.pypy.org/en/latest/faq.html#does-pypy-have-a-gil-why PEP 703 – Making the Global Interpret 阅读全文

posted @ 2017-10-10 00:42 papering 阅读(240) 评论(0) 推荐(0)

2017年10月9日

going

摘要：阅读全文

posted @ 2017-10-09 18:57 papering 阅读(173) 评论(0) 推荐(0)

Chernoff-Hoeffding inequality -- Chernoff bounds, and some applications

摘要： https://www.cs.utah.edu/~jeffp/teaching/cs5955/L3-Chern-Hoeff.pdf 【大数据-通过随机过程降维】 When dealing with modern big data sets, a very common theme is reduc 阅读全文

posted @ 2017-10-09 15:15 papering 阅读(480) 评论(0) 推荐(0)

Jaccard Similarity and Shingling

摘要： https://www.cs.utah.edu/~jeffp/teaching/cs5955/L4-Jaccard+Shingle.pdf https://www.cs.utah.edu/~jeffp/teaching/cs5955/L5-Minhash.pdf 【可测空间 convert the 阅读全文

posted @ 2017-10-09 14:23 papering 阅读(374) 评论(0) 推荐(0)

Locality-sensitive hashing Pr[m(Si) = m(Sj )] = E[JSˆ (Si, Sj )] = JS(Si, Sj )

摘要： A hash function that maps names to integers from 0 to 15. There is a collision between keys "John Smith" and "Sandra Dee".、、 A minimal perfect hash fu 阅读全文

posted @ 2017-10-09 11:34 papering 阅读(530) 评论(0) 推荐(0)

2017年10月7日

Software-defined networking

摘要： Software-defined networking administrators to programmatically initialize, control, change, and manage network behavior dynamically via open interface 阅读全文

posted @ 2017-10-07 20:15 papering 阅读(221) 评论(0) 推荐(0)

多项式回归仍是线性回归

摘要：阅读全文

posted @ 2017-10-07 09:34 papering 阅读(236) 评论(0) 推荐(0)

2017年10月5日

Aspect Oriented Programming

摘要： Aspect Oriented Programming 阅读全文

posted @ 2017-10-05 18:37 papering 阅读(171) 评论(0) 推荐(0)

2017年10月3日

ExpandRank

摘要： ExpandRank 阅读全文

posted @ 2017-10-03 18:55 papering 阅读(304) 评论(0) 推荐(0)

2017年9月30日

flash 搜索算法

摘要： flash 阅读全文

posted @ 2017-09-30 21:19 papering 阅读(285) 评论(0) 推荐(0)

小波分析

摘要： https://en.wikipedia.org/wiki/Wavelet For example, a wavelet could be created to have a frequency of Middle C and a short duration of roughly a 32nd n 阅读全文

posted @ 2017-09-30 19:14 papering 阅读(256) 评论(0) 推荐(0)

Generalised Policy Iteration With Monte-Carlo Evaluation

摘要： http://www0.cs.ucl.ac.uk/staff/d.silver/web/Teaching_files/control.pdf 阅读全文

posted @ 2017-09-30 19:08 papering 阅读(182) 评论(0) 推荐(0)

Learning an Optimal Policy: Model-free Methods

摘要： http://www.mit.edu/~9.54/fall14/slides/Reinforcement%20Learning%202-Model%20Free.pdf 【基于所有、单个样本】阅读全文

posted @ 2017-09-30 18:36 papering 阅读(184) 评论(0) 推荐(0)

To discount or not to discount in reinforcement learning: A case study comparing R learning and Q learning

摘要： https://www.cs.cmu.edu/afs/cs/project/jair/pub/volume4/kaelbling96a-html/node26.html 【平均-打折奖励】 Schwartz [106] examined the problem of adapting Q-learn 阅读全文

posted @ 2017-09-30 18:28 papering 阅读(225) 评论(0) 推荐(0)

adaptive heuristic critic 自适应启发评价强化学习

摘要： https://www.cs.cmu.edu/afs/cs/project/jair/pub/volume4/kaelbling96a-html/node24.html 【旧知-新知强化学习：对新知、旧知的综合】 The adaptive heuristic critic algorithm is 阅读全文

posted @ 2017-09-30 17:34 papering 阅读(635) 评论(0) 推荐(0)

Q-learning

摘要： https://www.cs.cmu.edu/afs/cs/project/jair/pub/volume4/kaelbling96a-html/node25.html https://www.cs.cmu.edu/afs/cs/project/jair/pub/volume4/kaelbling9 阅读全文

posted @ 2017-09-30 17:29 papering 阅读(202) 评论(0) 推荐(0)

transition_matrix Markov chain

摘要：阅读全文

posted @ 2017-09-30 17:09 papering 阅读(323) 评论(0) 推荐(0)

On Using Very Large Target Vocabulary for Neural Machine Translation Candidate Sampling Sampled Softmax

摘要：【softmax分类器的加速器】 https://www.tensorflow.org/api_docs/python/tf/nn/sampled_softmax_loss This is a faster way to train a softmax classifier over a huge 阅读全文

posted @ 2017-09-30 10:36 papering 阅读(698) 评论(0) 推荐(0)

Brotli

摘要： https://engineering.linkedin.com/blog/2017/05/boosting-site-speed-using-brotli-compression?utm_source=mybridge&utm_medium=web&utm_campaign=read_more 阅读全文

posted @ 2017-09-30 01:08 papering 阅读(204) 评论(0) 推荐(0)

数据到达etl系统的时间晚于事务发生的时间

摘要：迟到阅读全文

posted @ 2017-09-30 00:56 papering 阅读(136) 评论(0) 推荐(0)

事实表维度表从数据仓库角度看数据分析实质

摘要：阅读全文

posted @ 2017-09-30 00:39 papering 阅读(169) 评论(0) 推荐(0)

2017年9月29日

小波分析蒙特卡洛树

摘要：小波蒙特卡洛阅读全文

posted @ 2017-09-29 23:26 papering 阅读(179) 评论(0) 推荐(0)

CNN延拓至复数域

摘要：阅读全文

posted @ 2017-09-29 22:40 papering 阅读(665) 评论(2) 推荐(0)

squared-error loss is much more repaidly updated than mean-absolute-deviation when searching for splits

摘要：平方差损失能较绝对值差损失更快地更新阅读全文

posted @ 2017-09-29 20:37 papering 阅读(324) 评论(0) 推荐(0)

Greedy Function Approximation:A Gradient Boosting Machine

摘要： https://statweb.stanford.edu/~jhf/ftp/trebst.pdf page10 90% to 95% of the observations were often deleted without sacrificing accuracy of theestimates 阅读全文

posted @ 2017-09-29 20:03 papering 阅读(460) 评论(0) 推荐(0)

ridge regression 无惩罚，导致预测结果空间过大而无实用价值

摘要：【 biased regression methods to reduce variance 通过偏回归来减小方差】 https://onlinecourses.science.psu.edu/stat857/node/137 Introducing biased regression method 阅读全文

posted @ 2017-09-29 17:45 papering 阅读(298) 评论(0) 推荐(0)

7 Types of Regression Techniques

摘要： https://www.analyticsvidhya.com/blog/2015/08/comprehensive-guide-regression/ Linear Regression Logistic Regression Polynomial Regression Stepwise Regr 阅读全文

posted @ 2017-09-29 16:39 papering 阅读(346) 评论(0) 推荐(0)

lasso the moon

摘要：阅读全文

posted @ 2017-09-29 16:35 papering 阅读(138) 评论(0) 推荐(0)

Regularization: how complicated the model is? Regularization, measures complexity of model 使预测准确、平稳 predictive stable

摘要： http://www.kdd.org/kdd2016/papers/files/rfp0697-chenAemb.pdf https://homes.cs.washington.edu/~tqchen/pdf/BoostedTree.pdf 【Training Loss measures how w 阅读全文

posted @ 2017-09-29 16:16 papering 阅读(242) 评论(0) 推荐(0)