MATLAB Answers

0

Use my own .csv with MATLAB sample code; specifically with Wine Classification Example

Asked by Rob Stewart on 25 Sep 2019
Latest activity Answered by Deepak Kumar on 11 Oct 2019
How do I swap out MATLAB sample code for a .csv with my own data?
Specifically, I am looking at the 'Wine Classification' example walks users through how to create a classification nueral network example in MATLAB. It uses the wine_dataset which is pulled directly from MATLAB. However, I imagine the process is similar for other examples
I want to use the same sample code from a CSV instead of the wine_dataset data and run all of the same code. My data is 1,217 records organized as a target column followed by 13 attribute columns. To my knowledge this is exactly how the wine_dataset is organized just with a different number of records.
I have tried csvread, fopen, and a number of other functions.
The original code reads:
[x,t] = wine_dataset;
when I try:
ds = csvread('CellDNA2.csv')
[x,t] = ds;
I get the error message, "Insufficient number of outputs from right hand side of equal sign to satisfy
assignment."
It seems like there should be a simple way to just swap my data in for the sample data but I have spent hours trying to do so and not gotten anywhere. Any help would be greatly appreciated!

  0 Comments

Sign in to comment.

1 Answer

Answer by Deepak Kumar on 11 Oct 2019

I'm reffering the documentation link https://www.mathworks.com/help/deeplearning/examples/wine-classification.html to answer your query. Please note that after running the command [x,t] = wine_dataset; the input matrix x is of size and the target matrix t is of size . Here the Input matrix X has thirteen rows, for the thirteen attributes. Note that both X and t have 178 columns. These represent 178 wine sample attributes (inputs) and associated winery class vectors (targets). For your data the size of x will be and that of t will be , where N= no of samples. Arrange your data in the sheet in such a way that the size of x & t satifies the dimension criterion as mentioned. I will be able to help more if you can share your data.

  0 Comments

Sign in to comment.