In machine learning designing the structure of the model and training the neural network are relatively small elements of a longer chain of activities. We usually start with understanding business requirements, collecting and curating data, dividing it into training, validation and test subsets, and finally serving data to the model....
Computer Vision is one of the most interesting and my favorite application area for artificial intelligence. A big challenge for image analysis algorithms is fast detection and classification of objects in real time. The problem of detecting objects is much more difficult than the classification that I have discussed many...
If I had to indicate one algorithm in machine learning that is both very simple and highly effective, then my choice would be the k-nearest neighbors (KNN). What's more, it's not only simple and efficient, but it works well in surprisingly many areas of application. In this post I decided...
In the previous three parts of the tutorial, we learned about convolutional networks in detail. We looked at the convolution operation, the convolutional network architecture, and the problem of overfitting. In the classification of the CIFAR-10 dataset we achieved 81% on the test set. To go further we would have...
Convolutional neural network is one of the most effective neural network architecture in the field of image classification. In the first part of the tutorial, we discussed the convolution operation and built a simple densely connected neural network, which we used to classify CIFAR-10 dataset, achieving accuracy of 47%. In...
Convolutional neural network provides one of the best classification results for images. In the previous post, you had the opportunity to learn what a convolution is and how to classify a CIFAR-10 dataset using a simple densly connected neural network. By the way, we have obtained accuracy of 47% on...
