The derivative of bounded linear maps #
For detailed documentation of the FrΓ©chet derivative,
see the module docstring of Mathlib/Analysis/Calculus/FDeriv/Basic.lean.
This file contains the usual formulas (and existence assertions) for the derivative of continuous linear maps, both bundled and unbundled.
We also prove versions of the chain rule when one of the maps is a continuous linear map.
Bundled continuous linear maps #
There are currently two variants of these in mathlib, the bundled version
(named ContinuousLinearMap, and denoted E βL[π] F, works for topological vector spaces),
and the unbundled version (with a predicate IsBoundedLinearMap, requires normed spaces).
This section deals with the first form, see below for the unbundled version
Chain rule for a composition with a continuous linear map #
The chain rule says that given two differentiable functions g and f,
the function g β f is differentiable too, with derivative g' βL f'.
The theorems below deal with the special cases when one of the two functions is a continuous linear map.
These are explicitly not tagged @[fun_prop] because
fun_prop can already prove them with more general lemmas,
and this would only slow it down in failing situations.