How to create a dataset using PyBrain

Recipe Objective - How to create a dataset using PyBrain?

Datasets are data to be given to test, validate and train on networks.
To create a dataset, we need to use the pybrain dataset package: pybrain.datasets.
Pybrain supports dataset classes like SupervisedDataset, SequentialDataset, ClassificationDataSet. We are going to use SupervisedDataset to create our data set. The dataset to use depends on the machine learning task that the user is trying to implement. SupervisedDataset is the simplest, and we will use the same here.

Apply Machine Learning to Demand Forecasting Data Science Problems

For more related projects -

https://www.projectpro.io/projects/data-science-projects/tensorflow-projects
https://www.projectpro.io/projects/data-science-projects/keras-deep-learning-projects

Let's try to create an AND truth table. The inputs given are like a two-dimensional array, and we get 1 output. Here the input becomes the size, and the target becomes the output, which is 1. So the inputs that are used for our dataset are 2,1.

# importing supervised dataset
from pybrain.datasets import SupervisedDataSet

# Creating object
supervised_dataset = SupervisedDataSet(2, 1)

# AND table
and_table = [ [(0,0), (0,)], [(0,1), (0,)], [(1,0), (0,)], [(1,1), (1,)], ]

# Adding sample from and_table into supervised_dataset
for input, target in and_table:
  supervised_dataset.addSample(input, target)

print("Input=>\n", supervised_dataset['input'])
print("")
print("Target=>\n", supervised_dataset['target'])


Output -
Input=>
 [[0. 0.]
 [0. 1.]
 [1. 0.]
 [1. 1.]]

Target=>
 [[0.]
 [0.]
 [0.]
 [1.]]

In this way, we can create a dataset using PyBrain.