Title : Object Detection and Image Classification using CNN on CIFAR 10.
---------------------------------------------
The below procedure explains how to run the given code :
#########################################################################
Pre requisites:
1. System with Anaconda software installed
Link: https://www.continuum.io/downloads
To open your workspace type the following and navigate to the folder where you have kept the code.
jupyter notebook
2. System with Matlab 2017a version installed.
#########################################################################
Code Structure:
AlexNet/
-- data/results/
-- stats_alexnet_testing.mat
-- stats_alexnet_validation.mat
-- Logs/out_train_alexnet_cifar10_875643.cph-m1.uncc.edu
-- AlexNet_Tester.m
-- AlexNet_Trainer.m
-- AlexNet_Validator.m
-- confusionmatStats.m
Custom_CNN/
-- data/results/
-- stats_testing.mat
-- stats_validation.mat
-- Logs/out_train_Validation_cifar10_873177.cph-m1.uncc.edu
-- CustomCNN_Tester.m
-- CustomCNN_Trainer.m
-- CustomCNN_Validator.m
-- confusionmatStats.m
Dataset/
-- Matlab/cifar-10-batches-mat/
-- Python/cifar-10-batches-py/
KNN/
--Logs/out_knn_test_knn_cifar_879323.cph-m1.uncc.edu
--KNN_Analysis.py
--KNN_Tester.py
NeuralNetworks/
-- Logs/out_nn_train_sgd_nn_cifar_882619.cph-m1.uncc.edu
--NeuralNetwork_Trainer_Tester.py
PCA_NeuralNetworks/
-- Logs/out_nn_train_pca_nn_cifar_882632.cph-m1.uncc.edu
--PCA_NeuralNetwork_Trainer_Tester.py
TrainedModels/
-- AlexNet/trained_alexnet_validation.mat
-- Custom_CNN/trained_cnn_validation.mat
README.txt
###########################################################################
To run the program:
1) For KNN, Neural Networks and PCA_Neural Networks (Traditional Methods)
use anaconda software.
2) Open Jupyter notebook in command prompt and run the program.
3) For AlexNet, Custom_CNN, Use Matlab2017a.
To run the program consider the pre trained models in TrainedModels folder and run the AlexNet_Tester.m or CustomCNN_Tester.m
This gives the testing results and saves the results in data/results folder.
To verify the results just load the mat file in the above data/results folder and get the respective Performance measure : FScore, Confusion Matrix, Precision etc.
There are a total of 10 performance measures calculated for each of the alexnet and custom cnn.
###########################################################################
matlab2017a代码-deeplearning-cifar10:使用卷积神经网络的CIFAR10目标检测
需积分: 45 96 浏览量
2021-05-28
13:41:44
上传
评论 3
收藏 338.3MB ZIP 举报
deeplearning-cifar10-master.zip （42个子文件） 
deeplearning-cifar10-master NeuralNetworks 
NeuralNetwork_Trainer_Tester.py 4KB Logs out_nn_train_sgd_nn_cifar_882619.cph-m1.uncc.edu 5KB Custom_CNN Custom_CNN_Trainer.m 2KB confusionmatStats.m 2KB Custom_CNN_Validator.m 1009B data results stats_testing.mat 1KB stats_validation.mat 1KB Custom_CNN_Tester.m 708B Logs out_train_Validation_cifar10_873177.cph-m1.uncc.edu 352KB KNN KNN_Analysis.py 3KB KNN_Tester.py 3KB Logs out_knn_test_knn_cifar_879323.cph-m1.uncc.edu 2KB AlexNet AlexNet_Tester.m 801B confusionmatStats.m 2KB data results stats_alexnet_validation.mat 1KB stats_alexnet_testing.mat 1KB AlexNet_Validator.m 979B Logs out_train_alexnet_cifar10_875643.cph-m1.uncc.edu 450KB AlexNet_Trainer.m 2KB README.md 2KB TrainedModels Custom_CNN trained_cnn_validation.mat 696KB AlexNet trained_alexnet_validation.mat 439KB PCA_NeuralNetwork PCA_NeuralNetwork_Trainer_Tester.py 4KB Logs out_nn_train_sgd_nn_cifar_882632.cph-m1.uncc.edu 7KB Dataset Python cifar-10-batches-py data_batch_3 29.6MB .ipynb_checkpoints ExtractData-checkpoint.ipynb 8KB ExtractData.ipynb 8KB data_batch_5 29.6MB batches.meta 158B data_batch_4 29.6MB
readme.html 88B data_batch_2 29.6MB test_batch 29.6MB data_batch_1 29.6MB Matlab cifar-10-batches-mat data_batch_2.mat 29.16MB data_batch_5.mat 29.16MB test_batch.mat 29.15MB data_batch_1.mat 29.15MB batches.meta.mat 299B readme.html 88B data_batch_4.mat 29.15MB data_batch_3.mat 29.16MB
评论0