feat: Hölder framework for lp spaces#35197
Conversation
PR summary 6cc078933bImport changes for modified filesNo significant changes to the import graph Import changes for all files
|
|
This pull request has conflicts, please merge |
|
This pull request has conflicts, please merge |
…eanprover-community#35198) Currently the API exists only for `Real.HolderConjugate`. This generalizes it in anticipation of leanprover-community#35197, which implements the Hölder framework for `lp` spaces.
|
This PR/issue depends on: |
mike1729
left a comment
There was a problem hiding this comment.
I found a couple of small nits.
Co-authored-by: Michal Swietek <4404982+mike1729@users.noreply.github.com> Co-authored-by: Jireh Loreaux <loreaujy@gmail.com>
| open scoped lp ENNReal NNReal | ||
|
|
||
| namespace lp | ||
| -- the material in this section could be moved to `lpSpace`, but would require some extra imports |
There was a problem hiding this comment.
how heavy are the imports?
|
|
||
| end NontriviallyNormedField | ||
|
|
||
| lemma norm_tsumCLM {α 𝕜 E : Type*} [NontriviallyNormedField 𝕜] |
There was a problem hiding this comment.
I think norm_tsumCLM_le would be better (to match the above naming)
| lemma norm_tsumCLM {α 𝕜 E : Type*} [NontriviallyNormedField 𝕜] | |
| lemma norm_tsumCLM_le {α 𝕜 E : Type*} [NontriviallyNormedField 𝕜] |
| @[simps!] | ||
| noncomputable def holderₗ (B : (i : ι) → E i →L[𝕜] F i →L[𝕜] G i) {K : ℝ} (hBK : ∀ i, ‖B i‖ ≤ K) : | ||
| lp E p →ₗ[𝕜] lp F q →ₗ[𝕜] lp G r := | ||
| .mk₂ 𝕜 (holder r B hBK) ?_ ?_ ?_ ?_ where finally |
There was a problem hiding this comment.
There is a PR about the convention for where and finally: leanprover-community/leanprover-community.github.io#844
This was my take from the zulip discussion, but if you prefer your current version, then feel free to revive the discussion
lp E p#35152Real.HolderTriple#35198