How can I perform simple addition/subtraction operations on only certain elements of a row/column based on the first row/column values?

Basically, I have created a table with a range of voltage values across the top row and first column and a fixed 12 volts in every other element. Based on those varying values, I would like to modify the fixed elements to create the proper voltage in each element. All while ignoring the 0 value in (1,1).

I have this code so far, basically creating the table:

TopRowVolt = [linspace(0,5,9); linspace(12,12,9);linspace(12,12,9); ...
    linspace(12,12,9);linspace(12,12,9);linspace(12,12,9);...
    linspace(12,12,9);linspace(12,12,9);linspace(12,12,9);...
    linspace(12,12,9)];
FirstColVolt = [0,linspace(0,3,9)]';
ValueTable = [FirstColVolt , TopRowVolt]

For instance, I want to subtract 0.5V across the table for any value in the first column less than 1V, add 0.2V for values in the first column that fall in the range 1V to 2V, and add 0.5V for values above 2V.

Likewise, based on the top row, I'd like to subtract 0.5V based on values less than 2V and add 0.5V on values greater than 3V. I don't understand how to get the code to determine which values require which action and then perform the specific action only on the proper rows/columns, as the first column and row need to be left alone.

Answer 1

[Image Analyst]

Just use create a mask and use it as an index:

mask = FirstColVolt < 1;
FirstColVolt(mask) = FirstColVolt(mask) - 0.5;
mask = FirstColVolt >= 1 & FirstColVolt < 2;
FirstColVolt(mask) = FirstColVolt(mask) + 0.2;
mask = FirstColVolt >= 2;
FirstColVolt(mask) = FirstColVolt(mask) + 0.5;

mask2 = TopRowVolt < 2;
TopRowVolt(mask2) = TopRowVolt(mask2) - 0.5;
mask2 = TopRowVolt > 3;
TopRowVolt(mask2) = TopRowVolt(mask2) + 0.5;

[Chris Butler]

Here is what I entered (I entered the mask and mask2 after each line just as a test to see if I was entering it right.)

if true
  % code
mask = FirstColVolt(2:end,1) < 1; 
FirstColVolt(mask) = FirstColVolt(mask)-0.5;
mask
mask = FirstColVolt(2:end,1) >= 1 & FirstColVolt(2:end,1) <= 2;
FirstColVolt(mask) = FirstColVolt(mask) + 0.2;
mask
mask = FirstColVolt(2:end,1) > 2;
FirstColVolt(mask) = FirstColVolt(mask) + 0.5;
mask
mask2 = TopRowVolt(1,2:end) < 2;
TopRowVolt(mask2) = TopRowVolt(mask2) - 0.5;
mask2
mask2 = TopRowVolt(1,2:end) > 3;
TopRowVolt(mask2) = TopRowVolt(mask2) + 0.5;
mask2
end

When I entered ValueTable again, unfortunately it returned the original unmodified table. The mask and mask2 tests did return the proper logical answers, but I'm unsure of where I went wrong.

[Steven Lord]

Nowhere in your code did you change ValueTable. I suspect you expected that ValueTable had some sort of "reference", "pointer", or "handle" to the FirstColVolt and TopRowVolt arrays and that changing those arrays would also change ValueTable. That's not the case.

When you make a copy of those arrays (as you did when you created ValueTable) the copies are independent of the original FirstColVolt and TopRowVolt arrays. See the "Behavior of MATLAB Built-In Classes" section on this documentation page for a simpler example with an int32 scalar. Just as changing a in that example didn't change b, so changing FirstColVolt will not change ValueTable.

[Image Analyst]

You don't need to assign ValueTable in advance if you don't want to. I don't see any point in it. As long as TopRowVolt and FirstColVolt still have the same number of elements, you create ValueTable AFTER you execute the code I gave. Like Steve said, if you make it up before, it won't automatically update when you change variables that initially went into creating it.

As an analogy, if I took a photo of you and a photo of your wife and put them into a photo album, and then I took another photo of you and another photo of your wife, would the album automatically change? No, of course not. I'd have to put the new photos in there if I wanted the album to change.

[Chris Butler]

I guess I just don't understand how the mask was supposed to work. (I'm super new to MATLAB.) I thought it would add/subtract certain amounts to the existing voltages below the first row and the the right of the first column instead of creating a new array/table. Sorry. I'm nearly clueless as to how this all operates.

[Chris Butler]

Yeah. I definitely don't understand it. I put the mask code in front of the code where I created ValueTable and it still shows an even 12V across the board. Would hand calculations and hand keying the table be more efficient? I've been able to do these types of operations in programs like excel, which I know is not even remotely similar to MATLAB, but I figured I could possibly make it work similarly anyway.

[Image Analyst]

mask creates a logical vector (1-D array) where the value is "true" where the criteria is met and "false" where the criteria is not met. If you pass mask into an array as a logical index, then the operations you do apply ONLY to the elements where the mask is "true" and NOT to the elements where mask is false.

I'm not really sure what you want to do with those matrices. TopRowVolt is a 10-by-9 matrix, and FirstColVolt is a 10-by-1 vector.

Anyway, I discovered a problem with the code. Some of the earlier operations pushed the values into ranges where they were included in the mask in later operations. So I changed the code to check against only the original array values, not the "current" values which might have been changed.

FirstColVoltOriginal = FirstColVolt; % Make a copy of the original.
mask = FirstColVoltOriginal < 1;
FirstColVolt(mask) = FirstColVoltOriginal(mask) - 0.5;
mask = FirstColVoltOriginal >= 1 & FirstColVoltOriginal < 2;
FirstColVolt(mask) = FirstColVoltOriginal(mask) + 0.2;
mask = FirstColVoltOriginal >= 2;
FirstColVolt(mask) = FirstColVoltOriginal(mask) + 0.5;

TopRowVoltCopy = TopRowVolt; % Make a copy of the original.
mask2 = TopRowVoltCopy < 2;
TopRowVolt(mask2) = TopRowVoltCopy(mask2) - 0.5;
mask2 = TopRowVolt > 3;
TopRowVolt(mask2) = TopRowVoltCopy(mask2) + 0.5;

[Chris Butler]

Does the mask only check the first row of TopRowVolt somehow? Because it seems that it would check the entirety of the TopRowVolt matrix. I used TopRowVolt an FirstColVolt to create a 10x10 matrix (because it seemed faster than typing in every value into every element) where the first row and first column were actually variable voltages from a potentiometer that would affect the output to the standard 12V line. Hence (2:end,2:end) originally being nothing but 12's and the values of (1,1:end) and (1:end,1) being a range of voltages. I probably didn't communicate well enough what my intentions were, as I didn't want anything to modify the firs row or column. Only the 12's from (2:end,2:end) should be modified based on the variable voltages from the potentiometers.

[Chris Butler]

This is what I'm curious about. It keeps changing my values of the top row and first column and they should stay the same. Also, its a uniform change across the board for the initial 12V values, and they should vary based on the initial inputs from the first row or column because those variations should stack on top of one another.

[Chris Butler]

This is what I'm actually working on here. And I've done it already with if/else statements; which was brutal. I gave an example problem that was similar to what I'm doing but different in size, values, variation, etc. I'm trying to learn the code, as opposed to have someone else write it for me, if that makes sense. It's an air/fuel ratio table based on engine temperature and throttle position. As you can see, my variable voltages for throttle position stay their original values across the top row, as do the variable voltages for engine temperature down the first column. Also, the original value for every engine temp./throttle pos. combination was a stoichiometric 14.7 and each value has been varied by both the engine temp. voltage and throttle position voltage so they have a gradual variation from one column/row to the next. (The "0" in (1,1) has been ignored bt the rest of the table.)

[Chris Butler]

Like I previously mentioned, and displayed I guess, I've already figured out a way to accomplish this particular task (with a brutal 142 lines of code for such a task). But I would like to understand how to manipulate arrays more efficiently going forward. The screenshot I've shared here was a school project and that's why I didn't want to give my original problem, because as I stated earlier, I wanted to learn it instead of having someone else write it for me. (My course hasn't even reached masks yet to be honest. That's why I'm clueless as to how to use them.) Thanks so much, by the way for taking the time to try and help me understand this. And sorry if I don't soak this material up very well. I'm a hardcore novice at this.