From the course: Full-Stack Deep Learning with Python

Unlock this course with a free trial

Join today to access over 24,400 courses taught by industry experts.

Set up the dataset and data loader

Set up the dataset and data loader - Python Tutorial

From the course: Full-Stack Deep Learning with Python

Start my 1-month free trial Buy for my team

Set up the dataset and data loader

“

- [Instructor] Now that we know how ML flow tracks runs within experiments, we are ready to train our dense neural network for image classification. Once again, A DNN is not the right model for this, but it'll give us a handle on how ML flow works and will work well as a preparatory step before convolutional neural network training. The first thing I'm going to to do is set up a dataset to access our emnistS data. A dataset is just an abstraction that represents your data and it typically provides access to individual data samples one at a time. This emnist dataset class inherits from the dataset base class in PyTorch. Within the init method, we initialize the images and labels as torch tensors. Observe that the init method takes in the data frame for our training or test data as an input argument. The Len Method returns the length of the dataset. That's just the length of the images get item allows us to access the data…

Contents