Stat

In this sample program, we will introduce how to use the statistical analysis function jubastat through the Jubatus Client.

By setting windows on time-series data for the statistical analysis, you can utilize the abnormal data detection, sensor monitoring and etc.

Abstract of sample program

In this sample, we will study the ‘price’, ‘weight’, ‘diameter’ of orange, apples and melon. Program trivial_stat is used for the statistical analysis, such as standard deviation and total value of the parameter, in each fruit.

At first, please download the artificially created fruit-data from (fruit.csv). It is used to training the statistical model in jubastat .

Next, we will call methods to get the total value and the standard deviation of each fruit.

Processing flow

The flow of development using Jubatus Client is following:

  1. Connection settings to jubastat
    Setting the HOST and RPC port of jubastat .
  2. Prepare the training data
    Read the fruit-data from the downloaded CSV file line by line.
  3. Data training (update the model)
    Use the push method to send the fruit-data to jubastat and training the model there.
  4. Output the result
    Use the methods to get total-value, standard deviation, etc. And output them.

Sample Program

Currently, we have no sample programs except Python. (We welcome your contribution!)