It is generally known that the first derivative of $f:\Bbb R^m \to \Bbb R^n$ at $x$ is a linear map $Df(\mathbf x)\in L(\Bbb R^m,\Bbb R^n)$ satisfying $$ \lim_{|\mathbf h|\to 0} \frac{||f(\mathbf x+\mathbf h)-f(\mathbf x)-Df(\mathbf x)h||}{\mathbf |h|} = 0, $$ so technically $Df:\Bbb R^m\to L(\Bbb R^m,\Bbb R^n)$. With the use of partial derivatives, we can express $Df$ w.r.t. the natural coordinates by $$Df= \begin{bmatrix} D_1f_1 & D_2f_1 & \dots & D_mf_1\\ D_1f_2 & D_2f_2 & \dots & D_mf_2\\ \vdots & \vdots & \ddots &\vdots\\ D_1f_n & D_2f_n &\dots & D_mf_n \end{bmatrix} $$ where $f(\mathbf x)=(f_1(\mathbf x),\dots,f_n(\mathbf x))^T$, $\mathbf x=(x_1,\dots,x_m)$.
Now my question is
What's $D^2f$?
What is a nice way to represent it?
Well, I know that we'd need $D^2f:\Bbb R^m \to L(\Bbb R^m,L(\Bbb R^m,\Bbb R^n))$ but what are some nice ways to think about $L(\Bbb R^m,L(\Bbb R^m,\Bbb R^n))$?
Intuitively I kind of see what $D^2f$ do, but I since my background in Differential Geometry and Multi-variable Calculus is quite limited, I've never come across any book that offer a nice symbolic way to work with $D^2f$. I suspect that it has something to do with tensor notation that I'm not familiar with.
Any help is very appreciated, especially if you can point me some nice books/articles treating the topic. My background is mostly in functional analysis so I'm not scared off by terms like Frechet Derivative and such, please feel free to use them.
This is a related thing I found. The answers were good but did not address my question, which is the representation of $D^2f$.