Fault Diagnosis and forecast with univariate time series samples

Hello,

i will detect and make forecast about machine faults.
Therefore I have short time series from machine or sensor signal (e.c. Vibration Signal Input or Current, Position Difference)

The "Heart Disease Prediction", "Power Anomaly Detection" and "Forecasting" are very interresting, but the inputdata are only a oneliner for each date.
In my case I have a short time series (trace) with one ore more signal and hundres of data points to detect the actual machine state.

With this traces I will train and run the AI.

I've read some articles and they wrote to do this with Deep Convolutional Neural Network, but I don´t know how to start with dotnet ml.

Non vengono indicati file, test o punti di ingresso. Inizia esaminando gli esempi esistenti Heart Disease Prediction, Power Anomaly Detection e Forecasting, quindi determina come un nuovo esempio ML.NET rappresenterebbe le tracce di macchine o sensori e dimostrerebbe il rilevamento dei guasti e la previsione come comportamento completo.

Scritto dal modello di indicizzazione a partire dal testo della issue.