diff --git a/CHANGELOG.md b/CHANGELOG.md
index 4a9c13d..e81d9c5 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,16 @@
 # Changelog
 
+## 1.8.0 - 2023-11-15
+
+### Features
+
+-   Compatibility with TensorFlow 2.15.0.
+-   Inspector support for uplifting.
+
+### Fix
+
+-   Less warnings emitted when loading a model.
+
 ## 1.7.0 - 2023-10-20
 
 ### Features
diff --git a/WORKSPACE b/WORKSPACE
index c141ff4..9980289 100644
--- a/WORKSPACE
+++ b/WORKSPACE
@@ -11,9 +11,9 @@ load("@bazel_tools//tools/build_defs/repo:http.bzl", "http_archive")
 # absl used by tensorflow.
 http_archive(
     name = "org_tensorflow",
-    strip_prefix = "tensorflow-2.14.0",
-    sha256 = "447cdb65c80c86d6c6cf1388684f157612392723eaea832e6392d219098b49de",
-    urls = ["https://github.com/tensorflow/tensorflow/archive/v2.14.0.zip"],
+    strip_prefix = "tensorflow-2.15.0",
+    sha256 = "9cec5acb0ecf2d47b16891f8bc5bc6fbfdffe1700bdadc0d9ebe27ea34f0c220",
+    urls = ["https://github.com/tensorflow/tensorflow/archive/v2.15.0.zip"],
     # Starting with TF 2.14, disable hermetic Python builds.
     patch_args = ["-p1"],
     patches = ["//third_party/tensorflow:tf.patch"],
diff --git a/configure/setup.py b/configure/setup.py
index db9d6c1..57cf46c 100644
--- a/configure/setup.py
+++ b/configure/setup.py
@@ -21,7 +21,7 @@
 from setuptools.command.install import install
 from setuptools.dist import Distribution
 
-_VERSION = "1.7.0"
+_VERSION = "1.8.0"
 
 with open("README.md", "r", encoding="utf-8") as fh:
   long_description = fh.read()
@@ -29,7 +29,7 @@
 REQUIRED_PACKAGES = [
     "numpy",
     "pandas",
-    "tensorflow~=2.14.0",
+    "tensorflow~=2.15.0",
     "six",
     "absl_py",
     "wheel",
diff --git a/documentation/known_issues.md b/documentation/known_issues.md
index b66e198..d08b7a3 100644
--- a/documentation/known_issues.md
+++ b/documentation/known_issues.md
@@ -37,6 +37,8 @@ tensorflow_decision_forests/tensorflow/ops/training/training.so: undefined symbo
 
 -   Use the version of TF-DF that is compatible with your version of TensorFlow.
 
+Note that TF-DF is not compatible with Keras 3 at this time.
+
 ### Compatibility table
 
 The following table shows the compatibility between
diff --git a/tensorflow_decision_forests/__init__.py b/tensorflow_decision_forests/__init__.py
index fe51748..0c33c66 100644
--- a/tensorflow_decision_forests/__init__.py
+++ b/tensorflow_decision_forests/__init__.py
@@ -51,10 +51,10 @@
 
 """
 
-__version__ = "1.7.0"
+__version__ = "1.8.0"
 __author__ = "Mathieu Guillame-Bert"
 
-compatible_tf_versions = ["2.14.0"]
+compatible_tf_versions = ["2.15.0"]
 __git_version__ = "HEAD"  # Modify for release build.
 
 from tensorflow_decision_forests.tensorflow import check_version
diff --git a/tensorflow_decision_forests/keras/wrappers_pre_generated.py b/tensorflow_decision_forests/keras/wrappers_pre_generated.py
index e1d55f2..2052718 100644
--- a/tensorflow_decision_forests/keras/wrappers_pre_generated.py
+++ b/tensorflow_decision_forests/keras/wrappers_pre_generated.py
@@ -785,12 +785,12 @@ def capabilities() -> abstract_learner_pb2.LearnerCapabilities:
 class GradientBoostedTreesModel(core.CoreModel):
   r"""Gradient Boosted Trees learning algorithm.
 
-  A GBT (Gradient Boosted [Decision] Tree;
-  https://statweb.stanford.edu/~jhf/ftp/trebst.pdf) is a set of shallow
-  decision trees trained sequentially. Each tree is trained to predict and then
-  "correct" for the errors of the previously trained trees (more precisely each
-  tree predict the gradient of the loss relative to the model output).
-  GBTs use [early stopping](early_stopping.md) to avoid overfitting.
+  A [Gradient Boosted Trees](https://statweb.stanford.edu/~jhf/ftp/trebst.pdf)
+  (GBT), also known as Gradient Boosted Decision Trees (GBDT) or Gradient
+  Boosted Machines (GBM),  is a set of shallow decision trees trained
+  sequentially. Each tree is trained to predict and then "correct" for the
+  errors of the previously trained trees (more precisely each tree predict the
+  gradient of the loss relative to the model output).
 
   Usage example:
 
@@ -1169,7 +1169,7 @@ class GradientBoostedTreesModel(core.CoreModel):
     subsample: Ratio of the dataset (sampling without replacement) used to train
       individual trees for the random sampling method. If \\"subsample\\" is set
       and if \\"sampling_method\\" is NOT set or set to \\"NONE\\", then
-      \\"sampling_method\\" is implicitely set to \\"RANDOM\\". In other words,
+      \\"sampling_method\\" is implicitly set to \\"RANDOM\\". In other words,
       to enable random subsampling, you only need to set "\\"subsample\\".
       Default: 1.0.
     uplift_min_examples_in_treatment: For uplift models only. Minimum number of
diff --git a/tools/start_compile_docker.sh b/tools/start_compile_docker.sh
index b0a7179..2801a69 100755
--- a/tools/start_compile_docker.sh
+++ b/tools/start_compile_docker.sh
@@ -25,7 +25,7 @@
 #  ./tools/start_compile_docker.sh
 #
 #  # Compile and test TF-DF.
-#  RUN_TESTS=1 PY_VERSION=3.9 TF_VERSION=2.12.0 ./tools/test_bazel.sh
+#  RUN_TESTS=1 PY_VERSION=3.9 TF_VERSION=2.15.0 ./tools/test_bazel.sh
 #
 #  # Create a Pip package for a specific version of python.
 #  ./tools/build_pip_package.sh python3.9
@@ -58,7 +58,7 @@
 #  ./tools/build_pip_package.sh ALL_VERSIONS_ALREADY_ASSEMBLED
 #
 # https://hub.docker.com/r/tensorflow/build/tags?page=1
-DOCKER=tensorflow/build:2.14-python3.9
+DOCKER=tensorflow/build:2.15-python3.9
 
 # Current directory
 # Useful if Yggdrasil Decision Forests is available locally in a neighbor