Build your own Perceptron - Machine learning in Python
Machine learning is everywhere these days and thanks to google's DeepMind project deep neural networks in particular have gained a lot of attention. While artificial neural networks are already taking care of many complex tasks like filtering social media or beating Go champions, setting up a neural network yourself has never been easier.
Especially python has seen many amazing open source contributions in the field of machine learning that make it extremely easy to construct networks. The package we use here is the scikit-learn package.
In this video I show how to run the multilayer perceptron from scikit-learn to classify images. Specifically, we try to distinguish images showing dogs or frogs from the cifar-10 dataset. Our perceptron is moderately successful but is outperformed by
other types of artificial neural networks such as convolutional neural networks.
scikit-learn Package: http://scikit-learn.org/stable/
cifar-10 Dataset: https://www.cs.toronto.edu/~kriz/cifa...
numpy Package: http://www.numpy.org/
Python: https://www.python.org/
Смотрите видео Build your own Perceptron - Machine learning in Python онлайн без регистрации, длительностью часов минут секунд в хорошем качестве. Это видео добавил пользователь DMK Data 27 Январь 2018, не забудьте поделиться им ссылкой с друзьями и знакомыми, на нашем сайте его посмотрели 187 раз и оно понравилось 2 людям.