diabetes

Basic Support for Diabetes 130-US hospitals for years 1999-2008 Data Set

https://archive.ics.uci.edu/ml/datasets/Diabetes+130-US+hospitals+for+years+1999-2008

Requirements

• Python 3.6 >=
• Numpy
• Pandas

data_loader

import data_loader or from . import data_loader Note: import statement depends on your source root

data_loader.load_dataset(file_path, valid_rate=0.1)

Input

• file_path: label file path e.g. 'dataset_diabetes/diabetic_data.csv'
• valid_rate: validation data rate (0 - 1), 0.1 by default

Output

• train_features: training feature 2D-array (numpy.ndarray, str and int columns)
• train_labels: training label array (numpy.ndarray)
• valid_features: validation feature 2D-array (numpy.ndarray, str and int columns)
• valid_labels: validation label array (numpy.ndarray)
• test_features: test feature 2D-array (numpy.ndarray, str and int columns)
• test_labels: test label array (numpy.ndarray)
• label_dict: label dictionary (key: str, value: int)