Neural network learning -- image recognition using convolution neural network

Convolutional neural network (CNN) Convolutional neural network is especially suitable for processing such data as picture, video, audio, language and so on, which are related to each other's position. Convolutional neural network, like multi connected neural network, belongs to a category of nUTF-8...

Posted by jkraft10 on Sat, 15 May 2021 02:14:15 +0930

Solution of TensorFlow DLL file missing

preface: Recently, in the project of license plate recognition with TensorFlow, we found the following problem when training the model, that is, xxx.dll not found. (I am a Windows computer, maybe Windows computer has more problems, if there is a problem with the configuration of cuda environmenUTF-8...

Posted by buroy on Sat, 15 May 2021 06:14:56 +0930

Rtx3070 + ubuntu18.04 + cuda11.1 + cudnn8.0.4 + tensorflow 1.15.4 + pytorch1.7.0 + yolov3 environment configuration

Rtx3070 + ubuntu18.04 + cuda11.1 + cudnn8.0.4 + tensorflow 1.15.4 + pytorch1.7.0 + yolov3 environment configuration 1, VMware + Ubuntu 18.04 installation (1) Install virtual machine 16Pro virtual machine download 16Pro version cracking key 16 ZF3R0-FHED2-M80TY-8QYGC-NPKYFYF390-0HF8P-M81RQ-2DXQEUTF-8...

Posted by crackerjax on Wed, 19 May 2021 04:45:48 +0930

Detailed model introduction to the keras function API - (with comments)

Keras's functional model is Model, which is a generalized model with inputs and outputs. The tf.keras.Sequential model is a simple stack of layers and cannot represent any model.The Keras functional API allows you to build complex model topologies.For example: Multiple input model, Multiple ouUTF-8...

Posted by cockney on Sat, 22 May 2021 03:34:54 +0930

keras functional model

Keras's functional model is Model, which is a generalized model with inputs and outputs. The tf.keras.Sequential model is a simple stack of layers and cannot represent any model.The Keras functional API allows you to build complex model topologies.For example: Multiple input model, Multiple ouUTF-8...

Posted by tbare on Sat, 22 May 2021 04:54:18 +0930

NLP real combat 01: starting from simplicity, teaching you how to classify Chinese text (news text) step by step with Python

NLP real combat 01: starting from simplicity, teaching you how to classify Chinese text (news text) step by step with Python 1. Preface NLP natural language processing has always been a popular classification of artificial intelligence and data science. The theme of a certain society that bloggUTF-8...

Posted by tnylsej on Sun, 06 Jun 2021 05:40:49 +0930

Explain in detail the structure of Deeplab series: SPP, atrous revolution, ASPP, encoder decoder with atrous revolution

catalogue 1.SPP 2.Atrous Convolution 3.ASPP 4.Encoder-Decoder with Atrous Convolution 1.SPP Spatial Pyramid Pooling: Spatial Pyramid Pooling, arbitrary size input + fixed proportion pooling = > Fixed len gt h output. Such as the variant ROI pooling in fastrcnn. It takes different sizes of the sUTF-8...

Posted by pedroz on Mon, 21 Jun 2021 03:35:44 +0930

Introduction to machine learning for programmers

Artificial intelligence learning is inseparable from the verification of practice, it is recommended that you can learn more in practice Flyai AI competition service platform Take part in more training and competition, in order to improve their ability. FlyAI is a one-stop service platform thaUTF-8...

Posted by benyamin on Sat, 26 Jun 2021 07:25:14 +0930

On the final of Zhongxing holding the moon competition and its summary

preface In April, on the recommendation of my senior brother, I signed up for the Zhongxing holding month competition. At the beginning, it was just for the purpose of mixing up the qualification of a written test (because the written test is not required if the valid results are submitted). ThUTF-8...

Posted by little_webspinner on Thu, 01 Jul 2021 06:30:46 +0930

Introduction of conditional Gan (CGAN) and its application in image to image generation

1, Introduction 1.1 text generated image The figure below is a schematic diagram of traditional supervised learning text generation image The existing problems: more output, and the neural network can only give a fuzzy result, can not give a specific result and the effect they really want to gUTF-8...

Posted by aayatoos on Sun, 04 Jul 2021 03:20:13 +0930