In this project, we classify images from the CIFAR-10 dataset. The dataset consists of airplanes, dogs, cats, and other objects. Here, we preprocessed the data, then train a convolutional neural network on all the samples. We normalize the images, one-hot encode the labels, build a convolutional layer, max pool layer, and fully connected layer. At then end, we inspect their predictions on the sample images.
![Cifar10][image-cifar10]
You only need to install Conda and run the following commands:
conda env create -f ud-im-classification.yml
source activate ud-im-classification
You can open the solution by simply:
jupyter notebook dlnd_image_classification.ipynb
1.Install Anaconda
2.Download or clone this github repository
3.Launch jupyter notebok within the file containing image_classification.iypynb file
4.Run the cells to train and execute the model.
It contains images of the various objects and goal of the model is to identify the objects from this data set
