Update rfcs/20210513-pluggable-profiler-for-tensorflow.md

Co-authored-by: Situ Yi <[email protected]>

Author: jzhoulon

rfcs/20210513-pluggable-profiler-for-tensorflow.md

@@ -76,7 +76,7 @@ Core TensorFlow will register a new ProfilerInterface with [ProfilerFactory](htt
 
 #### Protobuf class
 Profiler uses `XSpace` to store the performance data collected by backends. With this type of data structure, TensorFlow's profiler tools can generate various views of performance, such as timeline, memory consumption, performance of every TensorFlow op and set of summaries. `XSpace` is C++ object generated by protobuf toolchain with a predefined structure in [xplane.proto](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/profiler/protobuf/xplane.proto#L9).
-On the plugin side, plugin will collect performance data with its own profiler infrastructures. When TensorFlow invokes `CollectData()`, the plugin transforms the performance data to `XSpace` and then serializes `XSpace` to the buffer provided by TensorFlow. To successfully serialize the object, plugin writers should keep a copy of `xspace.proto`, and make it exactly the same as that in the TensorFlow side.
+On the plugin side, plugin will collect performance data with its own profiler infrastructures. When TensorFlow invokes `CollectData()`, the plugin transforms the performance data to `XSpace` and then serializes `XSpace` to the buffer provided by TensorFlow. To successfully serialize the object, plugin writers should keep a copy of `xspace.proto`, and make it exactly the same as that on the TensorFlow side.
 
 ### Detailed API
 The C API will be placed in `tensorflow/c/experimental/profiler/profiler.h`.