- Deep Learning Essentials
- Wei Di Anurag Bhardwaj Jianing Wei
- 191字
- 2021-06-30 19:17:46
Data operations
In this section, we will look at some of the most common transformations applied on matrices.
- Matrix transpose: Matrix transpose is a matrix transform that simply mirrors the matrix along its main diagonal. Mathematically it is defined as follows:


- Matrix multiplication: Matrix multiplication is one of the most fundamental operations that can be applied to any two matrices. A matrix, A, of shape Ar x Ac can be multiplied by another matrix, B, of shape Br x Bc if and only if Ac = Br. The resultant matrix, C, is the shape Ar x Bc .The multiplication operation is defined as follows:

Matrix multiplication generally has very useful properties. For example, matrix multiplication is distributive:

Matrix multiplication is also associative:

Matrix multiplication also has a very simple form for its transpose:

Matrix multiplication is not commutative, which means A x B ≠ B x A. However, the dot products between two vectors is commutative:

推薦閱讀
- 大學(xué)計(jì)算機(jī)信息技術(shù)導(dǎo)論
- 人工免疫算法改進(jìn)及其應(yīng)用
- 蕩胸生層云:C語(yǔ)言開發(fā)修行實(shí)錄
- Java實(shí)用組件集
- 精通Excel VBA
- 大數(shù)據(jù)挑戰(zhàn)與NoSQL數(shù)據(jù)庫(kù)技術(shù)
- Creo Parametric 1.0中文版從入門到精通
- 單片機(jī)技術(shù)項(xiàng)目化原理與實(shí)訓(xùn)
- 簡(jiǎn)明學(xué)中文版Photoshop
- 計(jì)算機(jī)硬件技術(shù)基礎(chǔ)學(xué)習(xí)指導(dǎo)與練習(xí)
- JSP通用范例開發(fā)金典
- 新世紀(jì)Photoshop CS6中文版應(yīng)用教程
- 數(shù)據(jù)結(jié)構(gòu)與實(shí)訓(xùn)
- 網(wǎng)絡(luò)規(guī)劃與設(shè)計(jì)
- 微計(jì)算機(jī)原理及應(yīng)用