diff --git a/README.md b/README.md
index 6560a9c7a9..734bd4ddac 100644
--- a/README.md
+++ b/README.md
@@ -19,7 +19,7 @@ of component runs, input and output artifacts, and runtime configuration. This
metadata backend enables advanced functionality like experiment tracking or
warmstarting/resuming ML models from previous runs.
-
+
## Documentation
diff --git a/docs/guide/analysis.svg b/docs/guide/analysis.svg
deleted file mode 100644
index 37e94325a0..0000000000
--- a/docs/guide/analysis.svg
+++ /dev/null
@@ -1 +0,0 @@
-
diff --git a/docs/guide/diag_all.svg b/docs/guide/diag_all.svg
deleted file mode 100644
index 80cca8f806..0000000000
--- a/docs/guide/diag_all.svg
+++ /dev/null
@@ -1 +0,0 @@
-
diff --git a/docs/guide/feature_eng.svg b/docs/guide/feature_eng.svg
deleted file mode 100644
index d457effcd7..0000000000
--- a/docs/guide/feature_eng.svg
+++ /dev/null
@@ -1 +0,0 @@
-
diff --git a/docs/guide/images/analysis.png b/docs/guide/images/analysis.png
new file mode 100644
index 0000000000..a1bf71c8a3
Binary files /dev/null and b/docs/guide/images/analysis.png differ
diff --git a/docs/guide/images/component.png b/docs/guide/images/component.png
new file mode 100644
index 0000000000..6f37d9cd5c
Binary files /dev/null and b/docs/guide/images/component.png differ
diff --git a/docs/guide/images/component.svg b/docs/guide/images/component.svg
deleted file mode 100644
index 5085ea48bf..0000000000
--- a/docs/guide/images/component.svg
+++ /dev/null
@@ -1 +0,0 @@
-
diff --git a/docs/guide/images/diag_all.png b/docs/guide/images/diag_all.png
new file mode 100644
index 0000000000..16f27a81fe
Binary files /dev/null and b/docs/guide/images/diag_all.png differ
diff --git a/docs/guide/images/feature_eng.png b/docs/guide/images/feature_eng.png
new file mode 100644
index 0000000000..f207da8cfd
Binary files /dev/null and b/docs/guide/images/feature_eng.png differ
diff --git a/docs/guide/images/libraries_components.png b/docs/guide/images/libraries_components.png
new file mode 100644
index 0000000000..474301b50f
Binary files /dev/null and b/docs/guide/images/libraries_components.png differ
diff --git a/docs/guide/images/train.png b/docs/guide/images/train.png
new file mode 100644
index 0000000000..ac22ac0c50
Binary files /dev/null and b/docs/guide/images/train.png differ
diff --git a/docs/guide/images/wrangling.png b/docs/guide/images/wrangling.png
new file mode 100644
index 0000000000..712f0166f7
Binary files /dev/null and b/docs/guide/images/wrangling.png differ
diff --git a/docs/guide/index.md b/docs/guide/index.md
index a0521bbdd4..278f5e064a 100644
--- a/docs/guide/index.md
+++ b/docs/guide/index.md
@@ -128,7 +128,7 @@ and helps you validate your exported models, ensuring that they are
This diagram illustrates the flow of data between these components:
-
+
### Anatomy of a Component
@@ -138,7 +138,7 @@ TFX components consist of three main pieces:
* Executor
* Publisher
-
+
#### Driver and Publisher
@@ -161,7 +161,7 @@ will need to develop a `preprocessing_fn`.
TFX includes both libraries and pipeline components. This diagram illustrates
the relationships between TFX libraries and pipeline components:
-
+
TFX provides several Python packages that are the libraries which are used to
create pipeline components. You'll use these libraries to create the components
@@ -386,7 +386,7 @@ code once.
### Data Exploration, Visualization, and Cleaning
-
+
TFX pipelines typically begin with an [ExampleGen](examplegen.md) component, which
accepts input data and formats it as tf.Examples. Often this is done after the
@@ -437,7 +437,7 @@ dataset, and if necessary modify as required.
### Developing and Training Models
-
+
A typical TFX pipeline will include a [Transform](transform.md) component, which
will perform feature engineering by leveraging the capabilities of the
@@ -450,7 +450,7 @@ the Transform component if there is ever a possibility that these will also be
present in data sent for inference requests. [There are some important
considerations](train.md) when designing TensorFlow code for training in TFX.
-
+
The result of a Transform component is a SavedModel which will be imported and
used in your modeling code in TensorFlow, during a [Trainer](trainer.md)
@@ -479,7 +479,7 @@ tfma.export.export_eval_savedmodel(
### Analyzing and Understanding Model Performance
-
+
Following initial model development and training it's important to analyze and
really understand you model's performance. A typical TFX pipeline will include
@@ -527,7 +527,7 @@ inference requests. TFX supports deployment to three classes of deployment
targets. Trained models which have been exported as SavedModels can be deployed
to any or all of these deployment targets.
-
+
### Inference: TensorFlow Serving
diff --git a/docs/guide/inference.svg b/docs/guide/inference.svg
deleted file mode 100644
index 598cd2cd6c..0000000000
--- a/docs/guide/inference.svg
+++ /dev/null
@@ -1 +0,0 @@
-
\ No newline at end of file
diff --git a/docs/guide/libraries_components.svg b/docs/guide/libraries_components.svg
deleted file mode 100644
index edbea9e3e8..0000000000
--- a/docs/guide/libraries_components.svg
+++ /dev/null
@@ -1 +0,0 @@
-
diff --git a/docs/guide/train.svg b/docs/guide/train.svg
deleted file mode 100644
index 9c7e4a776b..0000000000
--- a/docs/guide/train.svg
+++ /dev/null
@@ -1 +0,0 @@
-
diff --git a/docs/guide/wrangling.svg b/docs/guide/wrangling.svg
deleted file mode 100644
index a9ed3aa902..0000000000
--- a/docs/guide/wrangling.svg
+++ /dev/null
@@ -1 +0,0 @@
-