5 ms·

Fortunately there is a website now capable of doing matrix calculus! http://www.matrixcalculus.org http://www.matrixcalculus.org Mathematica doesn't seem to be

• parrt 9y ago
  Wow! Great little calculator. Thanks for pointing us at it.
  • calebh 9y ago
    You should add a link in the resources section
    • parrt 9y ago
      Done. Added a link.
• BucketSort 9y ago
  Indeed, most tools surprisingly lack this ability. I was shocked when I needed to break my calculations down to a piece-wise form when doing matrix calc with SymPy.
• adyavanapalli 9y ago
  Note: This website presumes denominator layout, which is different from what is used in the guide.
  • calebh 9y ago
    Does the layout matter as long as you're consistent? Do the deep learning libraries presume that you're using a certain layout?
    • jph00 9y ago
      The deep learning libraries largely hide all the calculus - it's all automated.
• improbable22 9y ago
  Mathematica can absolutely do all of this! D[ matrix ,x] ... or things like this: f[x_,y_] := x^2 + Sin[y] vars={x,y}; Table[ D[f[x,y],var1, var2], {var1,vars}, {var2,vars}] // MatrixForm
  • SoerenL 9y ago
    But how do you compute the derivative of x'Ax in Mathematica (x being a vector and A being a matrix)? What you have pointed out is only scalar derivatives, if I am not mistaken here.
    • improbable22 9y ago
      Like this, perhaps? A = {{1,2},{3,4}} vec = {x^2, x^3} D[vec.A.vec, x] Or perhaps like this, again the table of derivatives: xvec = {x1,x2} Table[ D[xvec.A.xvec,x] ,{x,xvec}] (all untested... one typo caught...)
      • claar 9y ago
        http://www.wolframalpha.com/input/?i=D%5B%7Bx%5E2,+x%5E3%7D.%7B%7B1,2%7D,%7B3,4%7D%7D.%7Bx%5E2,+x%5E3%7D,+x%5D http://www.wolframalpha.com/input/?i=D%5B%7Bx%5E2,+x%5E3%7D....
        • parrt 9y ago
          Added Link to Wolfram Alpha...
      • calebh 9y ago
        Mathematica can definitely compute the derivatives if you fix the size of the matrix. This isn't very useful if you're trying to compute the derivative of an expression with arbitrary sized matrices.
        • improbable22 9y ago
          You can do general matrices too, what do you have in mind? aa[x_] = {{1, 2}, {3, 4}} x bb[x_] = {x^2, x^3} D[ a[x].b[x] , x, x] (* for any suitable tensors *) % /. {a -> aa, b -> bb}