In the context of modern digital infrastructure, physical servers constitute the backbone of data centers and enterprise IT architectures. The efficiency, reliability, and operational longevity of these servers are critical, directly influencing the overall business continuity and performance. This documentation report delves into two distinct approaches to predictive maintenance through machine learning (ML), as encapsulated in two models developed over the course of a project.
The project's essence is encapsulated in the "Project Proposal" by O’Neil Stefan Smith, which aims to leverage TensorFlow-based AI algorithms for enhancing predictive maintenance, thereby optimizing the operational longevity and efficiency of physical servers.
IniitialModel.ipynb & FinalModel.ipynb are to be run in Jupyter Notebook or Google Collab. pm_sensor_data_converted.csv is the csv file to be read by the files. pm_sensor_data_converted.csv will be ameneded as more data is collected . after Grading of project, ML_Final_project will allow contributors.