tensorless 0.7.0__tar.gz → 0.9.0__tar.gz

{tensorless-0.7.0/tensorless.egg-info → tensorless-0.9.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tensorless
-Version: 0.7.0
+Version: 0.9.0
 Summary: ML with maximum automation and minimum setup.
 Author: Tensorless Contributors
 License: MIT

{tensorless-0.7.0 → tensorless-0.9.0}/docs/training.md

@@ -95,9 +95,10 @@ By default, Tensorless holds out `val_split` of the data (10% for
 datasets with 50+ examples, 0% for smaller ones where a held-out split
 wouldn't be meaningful) and tracks validation loss after each epoch. If
 validation loss doesn't improve by at least `min_delta` for `patience`
-consecutive epochs, training stops early. The automatic default is 3
-consecutive epochs, and the completed model is still written to the `.tl`
-output file.
+consecutive epochs, training stops early. The automatic default is 5
+consecutive epochs. Before the completed model is written to the `.tl` output
+file, Tensorless restores the weights from the epoch with the best validation
+loss. This applies equally to regular training and built-in pretraining.
 
 ## Checkpointing during training
 

{tensorless-0.7.0 → tensorless-0.9.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "tensorless"
-version = "0.7.0"
+version = "0.9.0"
 description = "ML with maximum automation and minimum setup."
 readme = "README.md"
 requires-python = ">=3.9"

{tensorless-0.7.0 → tensorless-0.9.0}/tensorless/auto/config.py

@@ -125,7 +125,7 @@ def resolve_config(ds: Dataset, user: TrainConfig) -> ResolvedConfig:
         grad_clip=user.grad_clip if user.grad_clip is not None else 1.0,
         warmup_steps=user.warmup_steps if user.warmup_steps is not None else min(100, max(1, n // 10)),
         val_split=user.val_split if user.val_split is not None else (0.1 if n >= 50 else 0.0),
-        patience=user.patience if user.patience is not None else 3,
+        patience=user.patience if user.patience is not None else 5,
         min_delta=user.min_delta if user.min_delta is not None else 1e-4,
         device=device,
         precision=precision,