LMU - Time series forecasting. Help needed!

Hi everyone,

I am new to Nengo DL. I have been trying to see if it is possible to make time-series forecasts using an SNN.

looking through the forum I came across this post:

I tried to run the code provided in the article by user [happyjang7] I get a close enough result on the data provided. Also thanks to @tbekolay and @drasmuss I was able to improve my results using their recommended changes. but the issue is that the loss keeps on fluctuating across a wide range for some reason and also when I use my own time-series data it produces the same output for all test samples.

Inspired by this example I also tried to implement an LMU to forecast the time series. I have attached the code below:

LMU nengoDL.ipynb (149.5 KB)

but when I run the simulation on my own time series data(stock prices) then surprisingly again with this model I get the same prediction value(output) for all test samples and also get a huge loss.

I am not sure what I am doing wrong here, since I tried to tune the parameters but nothing seems to work. Would greatly appreciate it if someone could help. thank you.

Thank you

Hi @acnh371, and welcome to the Nengo forums. :smiley:

In addition to the thread you have posted, there is another (more recent) thread that may be of interest to you as well: link.

With respect to the notebook you posted, I found that for the dummy dataset, the model needed a large about of epochs (~5000) to reach a decent test result. I also reduced the model parameters (by making the units and order both 30) in order to make sure that the LMU network wasn’t overfitting the relatively small training data set.

As for the actual (stock price) data, I would caution that such a simplistic model (i.e. just the LMU network) may not be enough to produce decent results. Additionally, I don’t even know if forecasting the stock price is possible with the dataset you used.

Hi @xchoo,

Thank you for your reply.

For the actual data(stock data). I have implemented both LSTM and CNN in that dataset and they gave decent predictions. I read in here that LMUs are an alternative to LSTM therefore I wanted to see if they gave comparable results if I implemented an LMU.

Also, I get the same prediction value(output) for all test samples in the real test dataset. did you get the same? do you think normalizing the data in the range [0,1] would solve this?

Do you think I can change the model in any way to make it work with the stock market dataset? I know stock markets are not very predictable but I am doing this to compare SNN to ANN networks and would really appreciate some guidance on this.

Thank you