Zulip Chat Archive

Stream: general

Topic: Unclear error "typeclass instance problem is stuck"


ToMi (Apr 27 2025 at 16:18):

I want to state a theorem saying that a function is a bijection. But for "(sigma_g g)" in "lemma sigma_g_bijective" I get "typeclass instance problem is stuck, it is often due to metavariables". Here's the code "```
import MIL.Common

open Function
open Equiv
open MulAction

variable {G A : Type*} [Group G] [MulAction G A]
variable {g : G}
variable {a : A}
-- Permutation repr. of an action is a homemorphism (Dummit & Foote p. 42)

def sigma_g (g : G) : A → A :=
fun a => g • a

lemma sigma_g_bijective : Bijective (sigma_g g) := by
sorry
```" What can be the reason for that? thanks

Ruben Van de Velde (Apr 27 2025 at 17:11):

There is no way for lean to know what A is

Ruben Van de Velde (Apr 27 2025 at 17:12):

And please note that the backticks should be on a line of their own

ToMi (Apr 27 2025 at 17:13):

Ok thanks. Any suggestions how handle testing the function for bijectivity?

Ruben Van de Velde (Apr 27 2025 at 17:16):

I think constructor will split into injective and surjective subgoals


Last updated: May 02 2025 at 03:31 UTC