Log-Sum-Exp Trick to Prevent Numerical Underflow

摘要: 引自:http://wittawat.com/log-sum_exp_underflow.htmlMultiplying a series of termsp1p2⋯pnwhere0≤pi≤1can easily result in a numerical underflow. This is es... 阅读全文
posted @ 2015-03-15 22:54 Jia 阅读(784) 评论(0) 推荐(0) 编辑

Parameters estimation

摘要: The observed data will be regarded as realizations of random variables X1, X2,...,Xn,whose joint distribution depends on an unknown parameter θ.An estimate of θ will be a function of X1, X2,...,Xnand will hence be a random variable with a probability distribution called its samplingdistribution. 阅读全文
posted @ 2012-04-25 12:48 Jia 阅读(177) 评论(0) 推荐(0) 编辑

Stars and Cubes -- <Two> Fact Table Features

摘要: Every fact table represents a business process by capturing measurements that describe it.The level of detail at which the fact table records information is referred to as its grain.It is important to identify the grain of each fact table clearly and avoid situations of mixedgrain. Fact tables do no 阅读全文
posted @ 2012-02-27 19:11 Jia 阅读(295) 评论(0) 推荐(0) 编辑

Stars and Cubes -- <One> Dimension Table Features

摘要: Surrogate Keys and Natural KeysIn a star schema, each dimension table is assigned a surrogate key. This key is not a carryover from an operational system. It is created especially for the data warehouse or data mart. Surrogate keys are usually integers, generated and managed as part of the extract, 阅读全文
posted @ 2012-02-27 18:13 Jia 阅读(230) 评论(0) 推荐(0) 编辑

Data Warehouse Architectures -- <Two> Kimball’s Dimensional Data Warehouse

摘要: Like the Corporate Information Factory, this architecture begins by assuming a separation of the operational and analytic systems. As before, operational systems appear on the far left of the diagram. Again, these may incorporate data stores that are relational and nonrelational, and are likely to b 阅读全文
posted @ 2012-02-27 17:48 Jia 阅读(640) 评论(0) 推荐(0) 编辑

Data Warehouse Architectures -- <One> Inmon’s Corporate Information Factory

摘要: A highly simplified depiction of the Corporate Information Factory appears in Figure.To understand this architecture, start by looking at the left side of the diagram. There, you will find the operational systems, or transaction systems, that support the business. The data stores associated with the 阅读全文
posted @ 2012-02-27 16:46 Jia 阅读(684) 评论(0) 推荐(0) 编辑

Fundamental -- <Four> Using Star Schema

摘要: Understanding the basic usage pattern of the star schema allows the dimensional designer to make intelligent choices.Querying FactsMost queries against a star schema follow a consistent pattern. One or more facts are requested, along with the dimensional attributes that provide the desired context. 阅读全文
posted @ 2012-02-27 16:06 Jia 阅读(183) 评论(0) 推荐(0) 编辑

Fundamental -- <Three> The Star Schema

摘要: Dimension TablesIn a star schema, a dimension table contains columns representing dimensions. As discussed, these columns will provide context for facts.The dimension tables serve to provide the rich context needed for the study of facts. Inqueries and reports, the dimensions will be used to specify 阅读全文
posted @ 2012-02-27 15:40 Jia 阅读(209) 评论(0) 推荐(0) 编辑

Fundamental -- <Two> Dimensional Design

摘要: Information systems fall into two major categories: those that support the execution of business processes and those that support the analysis of business processes. The principles of dimensional design have evolved as a direct response to the unique requirements of analytic systems.The core of ever 阅读全文
posted @ 2012-02-27 15:15 Jia 阅读(183) 评论(0) 推荐(0) 编辑

Fundamental -- <One>

摘要: Notes for Star Schema--The complete reference1. Powerful and effective techniques for the design of analyticdatabases. These techniques unleash business analytics in a very simple way: they model themeasurement of business processes.2.The dimensional model of a business process is made up of two com 阅读全文
posted @ 2012-02-27 14:17 Jia 阅读(144) 评论(0) 推荐(0) 编辑