pip install hox
The code snippet down below shows mnist training script using the high-level utils package for handling the training loop. Alternatively, you can choose to construct your own training loop using the low-level functions provided in hox. An example demonstrating low-level Mnist training can be found in the examples/mnist folder.
from hox import *
import hox.utils as utils
#Create model
model = Model.create([Dense(784, 144, Relu()), Dense(144, 10, Sigmoid())])
#Upload mnist dataset
X, Y, x, y = utils.mnist()
#Train the model
utils.train(model, X, Y, epochs = 2, rate = 1, batch_size = 32)