Artificial Intelligence By Example

Training a random sample of the training dataset

Before spending resources on AWS, we must prepare a model locally and make sure it works. AWS possesses efficient mini-batch algorithms. But to master them, a local baseline will guarantee good AWS job management.

The k-means_clustering_ch7_minibatch.py file provides a way to verify the mini-batch solution.

The program begins by loading the data in the following lines of code:

dataset = pd.read_csv('data.csv')print (dataset.head())print(dataset)

Now a mini-batch dataset name dataset1 will be randomly created using Monte Carlo's large data volume principle with a mini-batch size of 1,000. Many variations of the method's Monte Carlo approaches apply to machine learning. The estimator within the ...