- "hidden dangers are more dangerous than open fire, prevention is better than disaster relief, and the responsibility is heavier than Mount Tai"
This paper introduces the security test process of sending SMS verification code interface in detail, including ideas, some test codes and test results. This test website ...
Posted by consolestrat on Mon, 07 Mar 2022 17:36:17 +0100
In 2018, due to the development of a large water conservancy Province Hydrological data website After the revision, the hydrological data of the province are changed into small pictures with a height of 13px. For example, the three pictures below represent the station name, upstream water level and downstream water level respectively: After d ...
Posted by rbudj on Fri, 04 Mar 2022 18:22:50 +0100
1, Common skills in transfer learning: fine tuning
Take the weights trained on the big data set as the initialization weight of the specific task (small data set), and retrain the network (modify the full connection layer output as needed); The training methods can be: 1. Fine tune all layers; 2. Fix the weights of the front l ...
Posted by daanoz on Mon, 21 Feb 2022 10:51:38 +0100
definitionLabel smoothing, like L1, L2 and dropout, is a regularization method in the field of machine learning. It is usually used for classification problems. The purpose is to prevent the model from predicting labels too confidently during training and improve the problem of poor generalization ability.backgroundFor the classification proble ...
Posted by runfastrick on Thu, 27 Jan 2022 14:28:53 +0100
*This sample program demonstrates how to use shape based matching to find slightly deformed objects, using the parameter "maximum deformation";
*Note that to find deformed objects, you can use parameters to apply shape based matching "maximum deformation" or - apply local deformable matching;
*In this example, the task is ...
Posted by 8ennett on Wed, 26 Jan 2022 12:15:09 +0100
maix asr (automatic speech recognition)
This document is in MaixPy 0.5.1_128 minimum_speech_with_ide_support firmware passed the test. Please ensure that the recording function / call model is available before use.
This is a speech recognition module based on acoustic model. When the user loads the words composed of Pinyin into the modul ...
Posted by anoopmail on Tue, 25 Jan 2022 04:26:56 +0100
Libraries and tools required and installation
The following libraries and tools will be used:
Testseract download address: https://digi.bib.uni-mannheim.de/tesseract/
1. Download the non dev exe file suitable for your version (32 or 64 bit), and then install it all the way.
Note: if you need to support multiple language ...
Posted by davidjmorin on Sat, 22 Jan 2022 14:48:04 +0100
Recently, I studied the knowledge of artificial intelligence image recognition, using the famous tensorflow tool. At the same time, in order to avoid repeated wheel building, I used the object detection api implemented by Google to realize image recognition. The download address is as follows: https://github.com/tensorflow/models
In fact, ther ...
Posted by homer09001 on Sat, 22 Jan 2022 09:45:37 +0100
DenseNet, a densely connected convolution network with Pytorch Note32
A summary of all notes:
Pytorch Note Happy Planet
Previous Resnet enhanced the information flow between the front and back layers through Shortcuts, alleviated the disappearance of gradients to some extent, and built the neural network very deeply. Furthermore ...
Posted by utpal on Mon, 10 Jan 2022 19:21:24 +0100
The original text is reproduced from "Liu Yue's technology blog" https://v3u.cn/a_id_201Some time ago, AnimeGAN, the industry's famous animation style transformation filter library, released the latest v2 version. For a time, there was no difference in the limelight. Tiktok two yuan, the largest user base in China is undoubtedly the j ...
Posted by knickerlas on Wed, 15 Dec 2021 22:05:28 +0100