Machine Failure Prediction

1. Description

Model and predict machine failure based on temperatures and disk error counts.

2. Contents

This README should be part of a distribution containing the following files:

• compdata.txt -- Training data file.
• compdata_true_errors.txt -- Error results corresponding to compdata.txt.
• driver.py -- A sample driver file.
• failuremodel.py -- The main source code file.
• README.md -- This file.

3. Requirements

The API relies on scikit-learn (http://scikit-learn.org), which requires SciPy and NumPy. Assuming that these are installed, scikit-learn can be installed using pip:

pip install -U scikit-learn

4. Usage

To use the API, failuremodel must be imported and a PredictFail object created:

import failuremodel

pf = failuremodel.PredictFail()

Tests can then be run against the model using the predict() method as follows:

pf.predict("test01", 100, 0)

Two methods can be used to access the alerts. First, print_alerts() will pretty-print all alerts in the queue in the order in which they were generated. Alternatively, get_alert_queue() will return the AlertQueue, allowing manual manipulation. Note that neither method clears the queue. This can be done by calling clear_alerts():

pf.clear_alerts()

A sample driver file (driver.py) has been provided which demonstrates the above methods as well as manual manipulation of the AlertQueue.