Medical Tabular Data
摘要
The most typical types of data in machine learning challenges are tabular data. A set of body mass index (BMI) values for a group of people is just a simple example of this type of data. We have the weight, age, and height values divided by the person to which they are assigned. Such data might be too simple to be used with a deep neural network or even most shallow methods, but we can use methods such as linear regression to see the correlation between these metrics. The tabular type of data consists of numbers and text values divided by features saved usually as a comma-separated values (CSV) or similar file format. Each value is separated by a semicolon or other sign, making it easy to load the data into a numpy or pandas matrix.