Contains some examples about Time Series Forecasting using TensorFlow Neural Networks.
In many of these examples I use a public dataset called AirPassengers. This dataset provides monthly totals of a US airline passengers from 1949 to 1960. This dataset was originally taken from an inbuilt dataset of R called AirPassengers. The copy of the dataset in the data folder was downloaded from Kaggle.
The notebooks use simple full connected Feed Forward Neural Networks. They describe how this kind of Neural Networks can be used in priciples with time series. I got a lot of experiences using these principles for forecasting purposes to support budgeting processes.
I show there also how to find good model configurations as well as to add additional features. As data quality is one of keys for success, one notebook shows what are the effects on missing data and outlines.