Deep learning: Tensorflow2.0 load and identify classic dataset MINIST

I. implementation ideas1. Load MNIST data set to get training set and test set2. Convert training set and test set to DataSet object3. Break up the data sequence Avoid reading data in the same order every time, make the model remember some characteristics of the training set, and reduce the generalizati ...

Posted on Thu, 20 Feb 2020 01:21:58 -0800 by mescal

Object oriented C language -- class, field, method

Object oriented C language -- class, field, method Things in reality are abstracted into classes The basic elements of Class are: field - variable, method - function Example: Person (class): _Field: age,name _Method: GetInfo,IsOlderThan,SayHello 1, Defining fields and methods in a class Define a p ...

Posted on Thu, 20 Feb 2020 00:28:58 -0800 by sunder

Python learning notes: Tsak03 - over fitting, under fitting and Solutions

Training error and generalization error Before explaining the above phenomena, we need to distinguish between training error and generalization error. Generally speaking, the former refers to the error of the model in the training data set, and the latter refers to the expectation of the error of the ...

Posted on Wed, 19 Feb 2020 05:13:00 -0800 by luddeb

VGG paper summary

VGG paper summary 1, Thesis translation abstract (1) Introduction (2) ConvNet configuration 1. architecture 2. configuration 3. discussion (3) Classification framework 1. training 2. test 3. Implementation details (4) Classification experiment 1. Single scale evaluation 2. Multi scale evaluatio ...

Posted on Tue, 18 Feb 2020 21:26:44 -0800 by websteve

Talk about the drawing process of View in an abstract way

Preface: Many times, when we look at the source code, we can understand its principle when we look at it, but it's easy to forget it soon after. This is because we don't leave an impression, know what we see, but feel unable to say it. This is the reason why we didn't summarize it; At present, many peo ...

Posted on Tue, 18 Feb 2020 20:29:50 -0800 by jolly

Sort Quick

Sorting is the operation of increasing or decreasing a list of data by the size of one or more of its keywords. I usually refer to sorting in ascending order and in situ Quick Sort, or Quick Row, is described below Quick Row Principle: Select a number from the interval to be sorted as a pivot Partition: Traverses through the entire interva ...

Posted on Tue, 18 Feb 2020 11:50:29 -0800 by nivaspm

Deep understanding of JDK8 HashMap

The author wrote in the last article, " Deep understanding of JDK7 HashMap HashMap's implementation principle in JDK7 is explained in detail, mainly around its put, get, resize, transfer methods. This paper will continue to analyze the specific implementation of HashMap in JDK8. Starting with put, ...

Posted on Mon, 17 Feb 2020 16:32:48 -0800 by rrhody

From nothing to useful Python to create a programming language of its own 1

Preface If you can compile the principle, understand the lexical analysis algorithm and syntax analysis algorithm, you can use any language to do such a thing. The reason why you use Python is that Python has the longest and most handy programming language. The so-called performance is not in the scope of this paper, the main focus of this pap ...

Posted on Mon, 17 Feb 2020 01:37:06 -0800 by ram4nd

ACM plan UVA - 1103 accident messages (breadth first)

Brief title A bitmap in hexadecimal is given. It is required to recognize the pictographs in the graph according to the binary system. There are 6 kinds of pictographs (below). All the hieroglyphs identified will be sorted according to their corresponding letters and then output. Corresponding letters ...

Posted on Sun, 16 Feb 2020 21:50:12 -0800 by jmarais

Deep learning task 3 over fitting, under fitting / gradient disappearing, gradient explosion / cyclic neural network advanced

1, Over fitting, under fitting and their solutions ·Model selection, over fitting and under fitting Training error: the error of the model in the training data set Generalization error: the expectation of the model error on any test sample, which is often approximated by the error on the test ...

Posted on Sun, 16 Feb 2020 21:29:38 -0800 by JonnyThunder