There are many ways in which you can feed data to the CNTK trainer. Which technique you should use depends on the size of the dataset and the format of the data. Let's take a look at how to work with smaller in-memory datasets first.
When you work with in-memory data in Python you will most likely use a framework such as Pandas or NumPy. These frameworks work with vectors and matrices of floating point or object data at their core and offer various levels of convenience when it comes to working with data.
Let's go over each of these libraries and explore how you can use data stored in these libraries to train your neural network.
