diff --git a/docs/pipelines/aggregators-custom.md b/docs/pipelines/aggregators-custom.md
index 3ed733e68c..ba5e029258 100644
--- a/docs/pipelines/aggregators-custom.md
+++ b/docs/pipelines/aggregators-custom.md
@@ -50,4 +50,4 @@ requirements-parser==0.2.0
 packaging==19.0.0
 ```
 
-You can install additional PyPI packages and import your own Python packages. See [Python Packages](../advanced/python-packages.md) for more details.
+You can install additional PyPI packages and import your own Python packages. See [Python Packages](python-packages.md) for more details.
diff --git a/docs/pipelines/aggregators.md b/docs/pipelines/aggregators.md
index 62c16d7bfa..7155177246 100644
--- a/docs/pipelines/aggregators.md
+++ b/docs/pipelines/aggregators.md
@@ -2,7 +2,7 @@
 
 An aggregator converts a set of columns and arbitrary values into a single value. Each aggregator has an input type and an output type. Aggregators run before transformers.
 
-Custom aggregators can be implemented in Python or PySpark. See the [implementation docs](../implementations/aggregators.md) for a detailed guide.
+Custom aggregators can be implemented in Python or PySpark. See the [implementation docs](aggregators.md) for a detailed guide.
 
 ## Config
 
diff --git a/docs/pipelines/estimators-custom.md b/docs/pipelines/estimators-custom.md
index a29f5ccf30..f55f7f958a 100644
--- a/docs/pipelines/estimators-custom.md
+++ b/docs/pipelines/estimators-custom.md
@@ -50,7 +50,7 @@ def create_estimator(run_config, model_config):
 
 ## Pre-installed Packages
 
-You can import PyPI packages or your own Python packages to help create more complex models. See [Python Packages](../advanced/python-packages.md) for more details.
+You can import PyPI packages or your own Python packages to help create more complex models. See [Python Packages](python-packages.md) for more details.
 
 The following packages have been pre-installed and can be used in your implementations:
 
@@ -63,7 +63,7 @@ requirements-parser==0.2.0
 packaging==19.0.0
 ```
 
-You can install additional PyPI packages and import your own Python packages. See [Python Packages](../advanced/python-packages.md) for more details.
+You can install additional PyPI packages and import your own Python packages. See [Python Packages](python-packages.md) for more details.
 
 
 # Tensorflow Transformations
diff --git a/docs/pipelines/estimators.md b/docs/pipelines/estimators.md
index c9874d1103..ff48e0ef83 100644
--- a/docs/pipelines/estimators.md
+++ b/docs/pipelines/estimators.md
@@ -2,7 +2,7 @@
 
 An estimator defines how to train a model.
 
-Custom estimators can be implemented in Python or PySpark. See the [implementation docs](../implementations/estimators.md) for a detailed guide.
+Custom estimators can be implemented in Python or PySpark. See the [implementation docs](estimators.md) for a detailed guide.
 
 ## Config
 
diff --git a/docs/pipelines/transformers-custom.md b/docs/pipelines/transformers-custom.md
index a84b7dd626..72f2e62c88 100644
--- a/docs/pipelines/transformers-custom.md
+++ b/docs/pipelines/transformers-custom.md
@@ -95,4 +95,4 @@ requirements-parser==0.2.0
 packaging==19.0.0
 ```
 
-You can install additional PyPI packages and import your own Python packages. See [Python Packages](../advanced/python-packages.md) for more details.
+You can install additional PyPI packages and import your own Python packages. See [Python Packages](python-packages.md) for more details.
diff --git a/docs/pipelines/transformers.md b/docs/pipelines/transformers.md
index 7f3687e071..34f90baafe 100644
--- a/docs/pipelines/transformers.md
+++ b/docs/pipelines/transformers.md
@@ -2,7 +2,7 @@
 
 A transformer converts a set of columns and arbitrary values into a single transformed column. Each transformer has an input type and an output column type.
 
-Custom transformers can be implemented in Python or PySpark. See the [implementation docs](../implementations/transformers.md) for a detailed guide.
+Custom transformers can be implemented in Python or PySpark. See the [implementation docs](transformers.md) for a detailed guide.
 
 ## Config