I am making a neural network for the weather forecasts. I trained it on a random input data at two input nodes and trained it, but I do not know how we can obtain results from Neural Network for unknown data. The inputs which I used are shown below. I tried obtaining results by using net( x) where x is input data, but it says that 'Number of inputs do not match net.numInputs.
Better way is that you upload your input data using excel files.and in your code,the number of columns about inputs should be same as number of columns in your excel file.
Have a look at this article - https://www.linkedin.com/pulse/artificial-neural-networks-optimization-using-genetic-ahmed-gad/
The number of inputs in the neural net should match when you are submitting a new input for testing. Otherwise, a mismatch in the number will not be accepted. This should be taken care at the time of constructing the neural network.
Also, the entire data set, should be classified into training data and test data. After training the neural network, you must have tested t check how far your network have been trained (I mean the accuracy). If the accuracy is in the expected lines (That is above the desired accuracy) then it is now ready to use.
I suggest you follow B.K. Tripathy's suggestion and comments. Dimensions compatibility must be respected, otherwise it will not execute.
Article Weather Forecasting Model using Artificial Neural Network
Article Dynamic fine-tuning stacked auto-encoder neural network for ...
B.K. Tripathy Sir I have two excel files containing 3 rows and 15 columns. I want to use both of these files as input (because they have different kinds of data). Do I need to make two input nodes for this purpose?
Have a look at
https://www.mathworks.com/matlabcentral/answers/262773-how-to-use-the-already-trained-neural-network
as I see in your matlab code, you trained your network with a 3D data( with 3 feature ) and you tested it with a 2 dimension data! So your network won't work! Use the same size as you you used in training!
Also you recall predict function with " {" !!
but you must use "("
So this is your the second error!
See matlab help script. It has helpful examples!
good luck...
try to make the dimension the same in both sides (training and testing)
- Badges
- Science topic
- Similar topics
- Organizational Psychology
- Training