It's been over a year since I last posted about NeuralMET, my experiments in AI/ML as applied to aviation weather forecasting. Here is the original post that introduced the subject, and here is the previous update from last year. At that time, the results (in terms of prediction accuracy) were somewhat marginal.
Since that time, I have continued to gather METAR data every half-hour, and by now have accumulated well over a year's worth of records for each location in my basket. I have thus re-trained all the ML models on these larger datasets, and the corresponding prediction results are now quite encouraging (see here for the online live forecasts plus historical performance for a selection of locations).
Interestingly, I obtain the best accuracy using a combination of Deep Neural Nets and Random Forests. The particular combination depends on the variable in question and on the forecast horizon. These best combinations (chosen by trial-and-error) are reflected in the online live forecasts.