Skip to content

Commit

Permalink
Use float32 scale in AutoScale MNIST training example. (#81)
Browse files Browse the repository at this point in the history 
In some configurations, `float16` does not have enough dynamic range to represent scaling,
meaning the scale factor can quickly overflow. Using `float32` (or equivalently `bfloat16`)
should allow a bit more margin.
  • Loading branch information
@balancap
balancap authored Jan 11, 2024
1 parent 69d428e commit 4e01b4b
Showing 0 changed files with 0 additions and 0 deletions.

0 comments on commit 4e01b4b

Please sign in to comment.