I previously illustrated two forecasting implementations with the same data set. This is the last one in these series using deep learning model. The dataset is about the temperature of major cities of the world but I only pulled Istanbul's data. Weather information is helpful for a lot of data science tasks like sales forecasting, logistics etc. and it is also one of the relevant data for time series analysis.
Briefly in this post, you will read about:
- Data preprocessing and basic EDA
- Keras data strutures
- Simple RNN model
- Model evaluation
- LSTM model and evaluation
- LSTM model tuning
- Summary and suggestions