What is the format of 'XTrain ' array in the example https://in​.mathworks​.com/help/​deeplearni​ng/ug/trai​n-a-variat​ional-auto​encoder-va​e-to-gener​ate-images​.html

I am not able to run the code given the following link:

https://in.mathworks.com/help/deeplearning/ug/train-a-variational-autoencoder-vae-to-generate-images.html

It is showing error while loading the 'MNIST' dataset. Someone please tell me what is the format of 'XTrain' data in this code. I am trying to run this code using my own dataset. format of 'XTrain' variable using my dataset is 102x1 cell. But I am getting error in mbq = minibatchqueue(dsTrain,numOutputs, ... line. it says Conversion to single from cell is not possible. So, I need to know the format of 'XTrain' variable to remove this error. Someone please help me with this.

Answer 1

[Prasannavenkatesh]

Hi Debojit,

The format of the XTrain variable present in the example "train a variational autoencoder vae to generate images" is of the type 4-D double which is a way to represent a four-dimensional (4-D) numerical array that stores data using double-precision floating-point numbers. Hence, due to mismatch in types you may be getting the conversion error. Hope this helps in solving your error.

[debojit sharma]

Thank you for your response.

Can you provide a sample code for this conversion of XTrain (with 102x1 cell format) to four-dimensional (4-D) numerical array that stores data using double-precision floating-point numbers. I am trying to do it using reshape function but I am getting errors. If possible, please help me with this

[Prasannavenkatesh]

Can you send a bit more information on your XTrain variable, regarding it's data, etc?