device_prediction/inputsamples/predict* (input sample files to the model)
server/ceph_telemetry/requirements.txt (updated)
Add logic to generate prediction results in a time series format:
device_prediction/predict_device.py
db_create_device.sql (updated)
For each device (which has a sufficient amount of data) we generate a daily input (which goes up to 12 reports back). We then invoke the model with this input, and insert the result into device.prediction_result table.
Add Red Hat's device failure prediction model:
Add logic to generate prediction results in a time series format:
For each device (which has a sufficient amount of data) we generate a daily input (which goes up to 12 reports back). We then invoke the model with this input, and insert the result into device.prediction_result table.