awslabs / predictive-maintenance-using-machine-learning

Set up end-to-end demo architecture for predictive maintenance issues with Machine Learning using Amazon SageMaker
Apache License 2.0
102 stars 59 forks source link

Wrong explaintaion for inference result. #10

Open dashjim opened 2 years ago

dashjim commented 2 years ago

In the notebook _sagemaker_predictivemaintenance.ipynb there is one place which explained how to interpret the inference result. That said "The predictions are a fraction of MAX_RUL which is 130.0, therefore the Remaining Useful Life predictions can be obtained by multiplying the output with 130". This is wrong.

By checking the code in entry_point.py, we know that the label divided 300 in training, and thus we need to multiply 300 to get it back, instead of multiply 130.