Update 20200705-tfx-ir.md

Fixes format

Author: ruoyu90

rfcs/20200705-tfx-ir.md

@@ -9,7 +9,7 @@
 ## 1. Objective
 
 This RFC documents the data model that supports the TFX DSL
-[semantics](https://github.com/tensorflow/community/blob/master/rfcs/20200601-tfx-uDSL-semantics.md).
+[semantics](https://github.com/tensorflow/community/blob/master/rfcs/20200601-tfx-udsl-semantics.md).
 It also introduces the TFX DSL intermediate representation (IR) and the workflow
 based on that. The IR is the bridge between the DSL and its orchestration /
 execution on all supported platforms and the workflow is the procedure that all
@@ -43,7 +43,7 @@ model became more visible since it would be infeasible to define different
 model when reading and writing ML metadata.
 
 The importance of the data model was further amplified with the proposal of the
-[advanced DSL semantics](https://github.com/tensorflow/community/blob/master/rfcs/20200601-tfx-TFX DSL-semantics.md),
+[advanced DSL semantics](https://github.com/tensorflow/community/blob/master/rfcs/20200601-tfx-udsl-semantics.md),
 in which the concept of synchronous versus asynchronous execution pipelines and
 synchronous versus asynchronous data pipelines was formalized. It is also in
 this proposal that asynchronous execution pipelines are considered to be the
@@ -56,7 +56,7 @@ fetching is still needed and is likely to dominate for use cases where one-shot
 pipelines are good enough. It is also useful even in the context of asynchronous
 execution pipelines, where data synchronization is required. A good example will
 be the
-[sub-pipeline](https://github.com/tensorflow/community/blob/master/rfcs/20200601-tfx-TFX DSL-semantics.md?#sub-pipeline)
+[sub-pipeline](https://github.com/tensorflow/community/blob/master/rfcs/20200601-tfx-udsl-semantics.md?#sub-pipeline)
 idea.
 
 Thus, it is highly desired to have a uniform data model that is consistent
@@ -145,12 +145,12 @@ IR, together with the workflow based on it should be able to:
 
 The design will be introduced in three sections:
 
--   Section [3.1](#data-model) describes how a TFX pipeline execution will be
+-   Section [3.1](#31-data-model) describes how a TFX pipeline execution will be
     represented in MLMD database (a.k.a., the end result of a pipeline
     execution).
--   Section [3.2](#intermediate-representation-ir) describes a specification of
+-   Section [3.2](#32-intermediate-representation-ir) describes a specification of
     a TFX pipeline.
--   Section [3.3](#workflows-scenarios-and-examples) describes how the detailed
+-   Section [3.3](#33-workflows-scenarios-and-examples) describes how the detailed
     workflow given a specification of a pipeline and how that works for
     different semantics.