[isabelle] Sledgehammer proof reconstruction failed
I was trying to infer the following:
fix a n :: nat
assume "a dvd n" "a < n powr (1/3)" "1 < n"
then have "a < n" by (smt divide_le_eq_1_pos of_nat_1_eq_iff of_nat_less_iff powr_less_cancel powr_one_gt_zero_iff)
This smt call has been suggested by Sledgehammer, but with the words
"proof reconstruction failed". Here is the exact error it generates:
exception THM 1 raised (line 2237 of "thm.ML"):
RSN: no unifiers
1 < n
¬ real n + - 1 * real 1 ≤ 0 ⟹ 1 < n
Is this a bug? It certainly seems like one...
This archive was generated by a fusion of
Pipermail (Mailman edition) and