Automated ML Data Pipeline

• Collect, analyze, train and produce prediction models.
• Monitor and maintain the model accuracy.
• Predict a range of environmental or cost states.
• Deploy on-prem or in the cloud.

Input(s):

• Sensory data from HVAC equipment, environmental data (temperature, humidity, etc), energy cost data, etc.

Main feature(s):

• An automated process for ML-based predictive modules. The framework includes data extraction, training, model creation, approval, monitoring and validation components.

Output(s):

• Models for predicting temperature, energy cost, etc for multiple zones within a facility.

N/A

Organizations running datacenters or other commercial facilities.

N/A