Hi Tobias,
To train a neural network with your provided data format, you need to ensure that the input layer can properly handle the cell array structure and the dimensions of your data. Given that your input data consists of 4 cells, each containing a 101x3 matrix, and your target data consists of 4 cells, each containing a 101x1 vector, you can use a sequence input layer in MATLAB.
Assuming the input data and target data are present in cell array formats, you can use a sequence input layer followed by a LSTM or other layers suitable for time-series prediction. Assuming 3 features for each operating point, you can use a ‘sequenceInputLayer(3)’ to create a sequence input layer with an input size of 3. For more information regarding the sequence input layer, you can refer to the following documentation: https://www.mathworks.com/help/deeplearning/ref/nnet.cnn.layer.sequenceinputlayer.html
Once the input layer is created, you can define a network architecture and training options as needed based on your specific requirements and data characteristics. For more references, you can look at the following documentation: https://www.mathworks.com/help/deeplearning/ug/time-series-forecasting-using-deep-learning.html
Hope this helps!
