Train MLP Neural Network on time series data?

Question

Newbie question here but I was curious to ask if an MLP Neural type network can be trained on time series data?

The dataset that I have is an electricity type data set from a building power meter and I can find I can train a decent NN model with including a lot of weather data and also a lot of one hot encoding dummy variables for time-of-week. (day, hour, month number, etc.)

I am experimenting in Python with the Tensorflow Keras library and I know the default during the training process randomly shuffles the data. Is this a No-No for a time series type problem where the random shuffle will take out the seasonality from the data? (stationary/non-stationary) The results shuffling the data really aren't that bad at a glance but not-randomly shuffling the data the results for MLP NN are poor, like the model doesn't train well.

I know some other times series forecast methods can include ARIMA, LSTM, etc. but I was curious to inquire if MLP can be used for these purposes too? What I ultimately need is a short term forecast method that can incorporate hourly weather forecast (from a web API) to forecast future hourly building electricity. Any tips greatly appreciated.

Thanks

Answer 1

So I posted this same question on machine learning mastery post about removing trends and seasonality difference transform for time series data. And Jason Brownlee responded to my questions:

Yes, you can use MLP, CNN and LSTM. It requires first converting the data to a supervised learning problem using a sliding window:

https://machinelearningmastery.com/time-series-forecasting-supervised-learning/

Then evaluating models in a way that respects the temporal ordering of obs, called walk-forward validation:

https://machinelearningmastery.com/backtest-machine-learning-models-time-series-forecasting/

You can see tens of tutorials and my book on this here:

https://machinelearningmastery.com/start-here/#deep_learning_time_series

MY REPLY TO THIS POST

Ben February 12, 2021 at 8:09 am #

Cool thanks for all the info. So if I used multivariate sliding window for MLP NN, is Ok when training the model that shuffle_data == True? or should I not shuffle training data…? Thanks so much!

Jason's REPLY TO MY QUESTION ABOVE

Jason Brownlee February 12, 2021 at 1:36 pm #

Yes, as long as all data in the training dataset is in the past compared to the test set.

Answer 2

Consider entity embeddings for categorical features in neural networks which is necessity for tabular data on NN.

Not mine: https://www.kaggle.com/abhishek/entity-embeddings-to-handle-categories