I have an h5 file which was created in Julia. I need to practice training but unsure of how to work with big-data.
feature_matrix = h5py.File('features.h5','r')
This file has dataset which has following architecture e.g.
Path: /feature1
Shape: (900000,)
Data type: float32
Path: /feature2
Shape: (900000,)
Data type: float32
Path: /label
Shape: (900000,)
Data type: float32
which is derived from code at https://www.titanwolf.org/Network/q/8a9ac510-a806-4504-8c61-1a8720717dab/y If I extract the features (70) into a matrix, it'll be too big to process. How should I combine features separately and keep label separately to train a model?