From 5bc4b276f537aa11eecae3807c5a6f908e37a6b2 Mon Sep 17 00:00:00 2001
From: BenjaminTrapani <ben.trapani1995@gmail.com>
Date: Thu, 7 Mar 2019 18:47:30 -0800
Subject: [PATCH 1/2] Cache previous parsed text sequence rows in memory to
 improve perf for large text files

---
 Source/Readers/CNTKTextFormatReader/TextParser.cpp | 8 ++++++++
 Source/Readers/CNTKTextFormatReader/TextParser.h   | 2 ++
 2 files changed, 10 insertions(+)

diff --git a/Source/Readers/CNTKTextFormatReader/TextParser.cpp b/Source/Readers/CNTKTextFormatReader/TextParser.cpp
index 009e94be023f..eceaa93cb22d 100644
--- a/Source/Readers/CNTKTextFormatReader/TextParser.cpp
+++ b/Source/Readers/CNTKTextFormatReader/TextParser.cpp
@@ -307,6 +307,12 @@ typename TextParser<ElemType>::SequenceBuffer TextParser<ElemType>::LoadSequence
 {
     size_t fileOffset = sequenceDsc.OffsetInChunk() + chunkOffsetInFile;
 
+	auto cachedSequencePos = m_fileOffsetToSequenceBuffer.find(fileOffset);
+	if (cachedSequencePos != m_fileOffsetToSequenceBuffer.end())
+	{
+        return cachedSequencePos->second;
+	}
+
     m_fileReader->SetFileOffset(fileOffset);
 
     size_t bytesToRead = sequenceDsc.SizeInBytes();
@@ -430,6 +436,8 @@ typename TextParser<ElemType>::SequenceBuffer TextParser<ElemType>::LoadSequence
     }
 
     FillSequenceMetadata(sequence, { sequenceDsc.m_key, 0 });
+
+    m_fileOffsetToSequenceBuffer[fileOffset] = sequence;
     return sequence;
 }
 
diff --git a/Source/Readers/CNTKTextFormatReader/TextParser.h b/Source/Readers/CNTKTextFormatReader/TextParser.h
index abf476a92922..3c4459f4ae60 100644
--- a/Source/Readers/CNTKTextFormatReader/TextParser.h
+++ b/Source/Readers/CNTKTextFormatReader/TextParser.h
@@ -10,6 +10,7 @@
 #include "TextConfigHelper.h"
 #include "Index.h"
 #include "CorpusDescriptor.h"
+#include <unordered_map>
 
 namespace CNTK {
 
@@ -135,6 +136,7 @@ class TextParser : public DataDeserializerBase {
     bool m_cacheIndex;
     unsigned int m_numRetries; // specifies the number of times an unsuccessful
                                // file operation should be repeated (default value is 5).
+    std::unordered_map<size_t, SequenceBuffer> m_fileOffsetToSequenceBuffer;
 
     // Corpus descriptor.
     CorpusDescriptorPtr m_corpus;

From 1b7fa257d32d72e8e9cdd3155d518e1c32f48fe3 Mon Sep 17 00:00:00 2001
From: Benjamin Trapani <ben.trapani1995@gmail.com>
Date: Sun, 10 Mar 2019 22:42:50 -0700
Subject: [PATCH 2/2] Convert tabs to spaces

Make indentation consistent with that used in the rest of the file.
---
 Source/Readers/CNTKTextFormatReader/TextParser.cpp | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/Source/Readers/CNTKTextFormatReader/TextParser.cpp b/Source/Readers/CNTKTextFormatReader/TextParser.cpp
index eceaa93cb22d..964c172c3b60 100644
--- a/Source/Readers/CNTKTextFormatReader/TextParser.cpp
+++ b/Source/Readers/CNTKTextFormatReader/TextParser.cpp
@@ -307,11 +307,11 @@ typename TextParser<ElemType>::SequenceBuffer TextParser<ElemType>::LoadSequence
 {
     size_t fileOffset = sequenceDsc.OffsetInChunk() + chunkOffsetInFile;
 
-	auto cachedSequencePos = m_fileOffsetToSequenceBuffer.find(fileOffset);
-	if (cachedSequencePos != m_fileOffsetToSequenceBuffer.end())
-	{
+    auto cachedSequencePos = m_fileOffsetToSequenceBuffer.find(fileOffset);
+    if (cachedSequencePos != m_fileOffsetToSequenceBuffer.end())
+    {
         return cachedSequencePos->second;
-	}
+    }
 
     m_fileReader->SetFileOffset(fileOffset);