Skip to content

v0.4.6

Choose a tag to compare

View all tags
@ChrisNick92 ChrisNick92 released this 08 May 14:16
· 6 commits to main since this release
v0.4.6
e501038
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's Changed

Early Stopping is Now Opt-In

  • patience defaults to None — early stopping is disabled unless explicitly set. Previously it defaulted to 15, silently terminating training for users who didn't opt in.
  • EarlyStopper is only added to the training loop when patience is provided.

current_patience on Trainer State

  • trainer.state.current_patience now tracks how many consecutive epochs have passed without a validation loss improvement, allowing real-time monitoring of early stopping progress.

Default Device Changed to "cpu"

  • Trainer, Evaluator, and get_device now default to device="cpu" instead of "cuda", preventing an immediate ValueError on CPU-only machines. Users with GPUs should explicitly pass device="cuda" or device="mps".

Full Changelog: v0.4.5...v0.4.6

Assets 2
Loading