search

d2l-ai / d2l-tvm

Dive into Deep Learning Compiler
https://tvm.d2l.ai
640 stars 98 forks source link

Fix typo #3

Closed cloudhan closed 4 years ago

cloudhan commented 4 years ago

AVX2 computes 8 float32 instructions per second is clearly a typo, otherwise, ENIAC would be faster

mli commented 4 years ago

good catch.