thank u for your tutorial i have followed but i have issue with generating future forecast, in testing the result is accurate but when it comes to future it generate the first value and repeat that. could u also provide tutorial on how to run future prediction after this stage?
You simply did what is in the official matlab tutorial with a background song. mentioned about seq2seq and seq2num but did not show to deploy it. what waste of eleven minutes
Hi, I’m working on predictive maintenance and I have a big dataset. The data is of dimension [1 X 20152320] and when I have followed your video, I got good results when I have considered a small subset. But while going for bigger data subsets, I am having problems while training. May I know what parameters in trainingOptions should I play with?
Just replace the split function with this code: def split(data): print(len(data)) half = int(len(data)/2) train = data[0:half-2] test = data[half+1:len(data)-1] print(len(train), len(test)) train=np.array(np.split(train,len(train)/7)) test=np.array(np.split(test,len(test)/7)) return train,test
Hi, Do you have any idea how to this problem to a multivariate problem that takes multiple inputs? And what part to be changed to predict for any time step in future? Thanks in advance
Hi! I have some sequences generated from some initial conditions, what model should I use to have a sequence generated from some initial condition based on the data I have? seq2seq models usually predict the following data of a series but don't generate sequence from initial conditions.
which version of matlab are you using ? im currently using lstm for rainfall, and got stuck at plot (data) ... the error i got is error using plot and invalid data argument. Can you detect where is my mistakes?
It would have been good if you showed the predicted vs actual consumption during the four years. It will be even more interesting to predict the consumption for 2011 year without having an actual data to compare with.
Hi. Thanks for the video I implemented the same with your code. However, it returns an error in Path.m file Undefined function or variable 'start'. Error in Path (line 6) [~,move] = max(q(start,:)); Error in Running (line 33) path = Path(q,Goal); As I checked, there is no declaration for 'start' variable. Do you know why this error happens?
Hiii How to handle persistent model problem. While doing time series analysis i get the output which seems to be one time step ahead of the actual series. How to rectify this problem?? This thing i am getting with several ML, DL, and as well as with statistical algos. Please do reply??
When I train the net using 80% of the data, then can I predict the all range of the data and compare the original data with forecast data in all data point???