Prepare for the next release of TensorFlow Recommenders.

Includes several important bug fixes as well as the first experimental release of the pre-built ranking model.

PiperOrigin-RevId: 372436083

Author: maciejkula

CHANGELOG.md

@@ -1,19 +1,23 @@
 # Changelog
 
-## [Unreleased][unreleased]
+## [0.5.0][2021-05-06]
 
 ### Changed
 
 -   Fixed the bug in `tfrs.layers.loss.SamplingProbablityCorrection` that logits
-    should substract the log of item probability.
--   `tfrs.experimental.models.RankingModel` can be used as DLRM like model with
-    Dot Product feature interaction or DCN like model with Cross layer.
+    should subtract the log of item probability.
 -   `tfrs.experimental.optimizers.CompositeOptimizer`: an optimizer that
     composes multiple individual optimizers which can be applied to different
     subsets of the model's variables.
 -   `tfrs.layers.dcn.Cross` and `DotInteraction` layers have been moved to
     `tfrs.layers.feature_interaction` package.
 
+## Added
+
+-   `tfrs.experimental.models.Ranking`, an experimental pre-built model for
+    ranking tasks. Can be used as DLRM like model with Dot Product feature
+    interaction or DCN like model with Cross layer.
+
 ## [0.4.0][2021-01-20]
 
 ### Added

setup.py

@@ -17,7 +17,7 @@
 import pathlib
 import setuptools
 
-VERSION = "0.4.0"
+VERSION = "0.5.0"
 
 REQUIRED_PACKAGES = [
     "absl-py >= 0.1.6",

tensorflow_recommenders/__init__.py

@@ -21,7 +21,7 @@
 you the flexibility to build complex models.
 """
 
-__version__ = "v0.4.0"
+__version__ = "v0.5.0"
 
 from tensorflow_recommenders import examples
 from tensorflow_recommenders import experimental