Skip to content

Commit

Permalink
Remove Extra Code Comments and unused InternalStreams project (#514)
Browse files Browse the repository at this point in the history 
* First attempt at removing extra code comments

* Round #2

* Removing Microsoft.ML.InternalStreams per comment on #513

* Address notes from @Ivanidzo4ka

* Remove TreeOrderedCandidatesSearch

* Remove whitespace and reinstate commented out tests
  • Loading branch information
@dan-drews @Ivanidzo4ka
dan-drews authored and Ivanidzo4ka committed Jul 13, 2018
1 parent ceac01f commit 669f4fa
Show file tree
Hide file tree
Showing 27 changed files with 10 additions and 19,678 deletions.
7 changes: 0 additions & 7 deletions Microsoft.ML.sln
View file
Original file line number Diff line number Diff line change
Expand Up @@ -36,8 +36,6 @@ Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Microsoft.ML.Tests", "test\
EndProject
Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Microsoft.ML.TestFramework", "test\Microsoft.ML.TestFramework\Microsoft.ML.TestFramework.csproj", "{B5989C06-4FFA-46C1-9D85-9366B34AB0A2}"
EndProject
Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Microsoft.ML.InternalStreams", "src\Microsoft.ML.InternalStreams\Microsoft.ML.InternalStreams.csproj", "{C4F7938F-7109-43C8-92A5-9BE47C7FF7D9}"
EndProject
Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Microsoft.ML.Predictor.Tests", "test\Microsoft.ML.Predictor.Tests\Microsoft.ML.Predictor.Tests.csproj", "{6B047E09-39C9-4583-96F3-685D84CA4117}"
EndProject
Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Microsoft.ML.ResultProcessor", "src\Microsoft.ML.ResultProcessor\Microsoft.ML.ResultProcessor.csproj", "{3769FCC3-9AFF-4C37-97E9-6854324681DF}"
Expand Down Expand Up @@ -150,10 +148,6 @@ Global
{B5989C06-4FFA-46C1-9D85-9366B34AB0A2}.Debug|Any CPU.Build.0 = Debug|Any CPU
{B5989C06-4FFA-46C1-9D85-9366B34AB0A2}.Release|Any CPU.ActiveCfg = Release|Any CPU
{B5989C06-4FFA-46C1-9D85-9366B34AB0A2}.Release|Any CPU.Build.0 = Release|Any CPU
{C4F7938F-7109-43C8-92A5-9BE47C7FF7D9}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
{C4F7938F-7109-43C8-92A5-9BE47C7FF7D9}.Debug|Any CPU.Build.0 = Debug|Any CPU
{C4F7938F-7109-43C8-92A5-9BE47C7FF7D9}.Release|Any CPU.ActiveCfg = Release|Any CPU
{C4F7938F-7109-43C8-92A5-9BE47C7FF7D9}.Release|Any CPU.Build.0 = Release|Any CPU
{6B047E09-39C9-4583-96F3-685D84CA4117}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
{6B047E09-39C9-4583-96F3-685D84CA4117}.Debug|Any CPU.Build.0 = Debug|Any CPU
{6B047E09-39C9-4583-96F3-685D84CA4117}.Release|Any CPU.ActiveCfg = Release|Any CPU
Expand Down Expand Up @@ -218,7 +212,6 @@ Global
{2F636A2C-062C-49F4-85F3-60DCADAB6A43} = {09EADF06-BE25-4228-AB53-95AE3E15B530}
{64BC22D3-1E76-41EF-94D8-C79E471FF2DD} = {AED9C836-31E3-4F3F-8ABC-929555D3F3C4}
{B5989C06-4FFA-46C1-9D85-9366B34AB0A2} = {AED9C836-31E3-4F3F-8ABC-929555D3F3C4}
{C4F7938F-7109-43C8-92A5-9BE47C7FF7D9} = {09EADF06-BE25-4228-AB53-95AE3E15B530}
{6B047E09-39C9-4583-96F3-685D84CA4117} = {AED9C836-31E3-4F3F-8ABC-929555D3F3C4}
{3769FCC3-9AFF-4C37-97E9-6854324681DF} = {09EADF06-BE25-4228-AB53-95AE3E15B530}
{B7B593C5-FB8C-4ADA-A638-5B53B47D087E} = {09EADF06-BE25-4228-AB53-95AE3E15B530}
Expand Down
1 change: 0 additions & 1 deletion src/Microsoft.ML.Console/Microsoft.ML.Console.csproj
View file
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,6 @@
<ProjectReference Include="..\Microsoft.ML.Data\Microsoft.ML.Data.csproj" />
<ProjectReference Include="..\Microsoft.ML.Ensemble\Microsoft.ML.Ensemble.csproj" />
<ProjectReference Include="..\Microsoft.ML.FastTree\Microsoft.ML.FastTree.csproj" />
<ProjectReference Include="..\Microsoft.ML.InternalStreams\Microsoft.ML.InternalStreams.csproj" />
<ProjectReference Include="..\Microsoft.ML.KMeansClustering\Microsoft.ML.KMeansClustering.csproj" />
<ProjectReference Include="..\Microsoft.ML.LightGBM\Microsoft.ML.LightGBM.csproj" />
<ProjectReference Include="..\Microsoft.ML.Maml\Microsoft.ML.Maml.csproj" />
Expand Down
6 changes: 0 additions & 6 deletions src/Microsoft.ML.Core/Utilities/DoubleParser.cs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -433,12 +433,6 @@ public static bool TryParse(out Double value, string s, int ichMin, int ichLim,
if (FloatUtils.GetBits(x) != 0 || FloatUtils.GetBits(value) != TopBit || !neg)
{
System.Diagnostics.Debug.WriteLine("*** FloatParser disagrees with Double.TryParse on: {0} ({1} vs {2})", str, FloatUtils.GetBits(x), FloatUtils.GetBits(value));
//if (!_failed)
//{
// // REVIEW: Double.Parse gets several things wrong, like mapping 148e-325 to 0x2 instead of 0x3.
// _failed = true;
// Contracts.Assert(false, string.Format("FloatParser disagrees with Double.TryParse on: {0} ({1} vs {2})", str, FloatUtils.GetBits(x), FloatUtils.GetBits(value)));
//}
}
}
#endif
Expand Down
4 changes: 0 additions & 4 deletions src/Microsoft.ML.Core/Utilities/MathUtils.cs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -258,10 +258,6 @@ public static Float SoftMax(Float[] inputs, int count)
if (count == 1)
return max;

//else if (leng == 2) {
// return SoftMax(inputs[0], inputs[1]);
//}

double intermediate = 0.0;
Float cutoff = max - LogTolerance;

Expand Down
6 changes: 3 additions & 3 deletions src/Microsoft.ML.Data/DataLoadSave/Text/TextLoaderParser.cs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -137,9 +137,9 @@ private sealed class ParseStats
private volatile int _cref;

// Total number of rows, number of unparsable values, number of format errors.
private /*volatile*/ long _rowCount;
private /*volatile*/ long _badCount;
private /*volatile*/ long _fmtCount;
private long _rowCount;
private long _badCount;
private long _fmtCount;

public ParseStats(IChannelProvider provider, int cref, long maxShow = MaxShow)
{
Expand Down
9 changes: 0 additions & 9 deletions src/Microsoft.ML.FastTree/Application/LogLossApplication.cs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -107,15 +107,6 @@ protected override void GetGradientInOneQuery(int query, int threadIndex)

_gradient[d1] += delta;
_gradient[d2] -= delta;
/*
double labelDiff = (labels[d1] - labels[d2]);
double margin = labelDiff * (_scores[d1] - _scores[d2]);
if (_coef - margin > 0)
{
_gradient[d1] += 0.01 * labelDiff;
_gradient[d2] -= 0.01 * labelDiff;
}
*/
}
}
}
Expand Down
1 change: 0 additions & 1 deletion src/Microsoft.ML.FastTree/Application/WinLossSurplusApplication.cs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -120,7 +120,6 @@ protected override void GetGradientInOneQuery(int query, int threadIndex)

double inverseMaxDCG = _inverseMaxDCGT[query];

//int[] permutation = (threadIndex < 0 ? new int[numDocuments] : _permutationBuffers[threadIndex]);
int[] permutation = _permutationBuffers[threadIndex];

short[] labels = Labels;
Expand Down
30 changes: 1 addition & 29 deletions src/Microsoft.ML.FastTree/Dataset/FeatureFlock.cs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -698,20 +698,6 @@ public void FillSplitCandidatesCategoricalLowPopulation(LeastSquaresRegressionTr
var binStats = virtualBins[i];
catFeatureCount += 1 + binStats.SubFeatures.Length;

/*int feature = features[i];
int subfeature = feature - featureMin;
Contracts.Assert(0 <= subfeature && subfeature < Flock.Count);
Contracts.Assert(subfeature <= feature);
Contracts.Assert(learner.TrainData.FlockToFirstFeature(flock) == feature - subfeature);
Contracts.Assert(featureUseCount[feature] >= 0);
Contracts.Assert(Flock.BinCount(subfeature) == 2);
Contracts.Assert(GetMaxBorder(subfeature) == GetMinBorder(subfeature));
var binStats = GetBinStats(GetMinBorder(subfeature));
sumGTTargets += binStats.SumTargets;
if (hasWeights)
sumGTWeights += binStats.SumWeights;*/

sumGTTargets += binStats.SumTargets;
gtCount += binStats.Count;
docsInCurrentGroup += binStats.Count;
Expand Down Expand Up @@ -933,21 +919,7 @@ public void FillSplitCandidatesCategoricalNeighborBundling(LeastSquaresRegressio
{
var binStats = virtualBins[i];
catFeatureCount += 1 + binStats.SubFeatures.Length;

/*int feature = features[i];
int subfeature = feature - featureMin;
Contracts.Assert(0 <= subfeature && subfeature < Flock.Count);
Contracts.Assert(subfeature <= feature);
Contracts.Assert(learner.TrainData.FlockToFirstFeature(flock) == feature - subfeature);
Contracts.Assert(featureUseCount[feature] >= 0);
Contracts.Assert(Flock.BinCount(subfeature) == 2);
Contracts.Assert(GetMaxBorder(subfeature) == GetMinBorder(subfeature));
var binStats = GetBinStats(GetMinBorder(subfeature));
sumGTTargets += binStats.SumTargets;
if (hasWeights)
sumGTWeights += binStats.SumWeights;*/


sumGTTargets += binStats.SumTargets;
gtCount += binStats.Count;
docsInCurrentGroup += binStats.Count;
Expand Down
3 changes: 0 additions & 3 deletions src/Microsoft.ML.FastTree/Dataset/SegmentIntArray.cs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -168,11 +168,8 @@ public static void StatsOfBestEncoding(uint[] ivalues, int bitsForMaxItem, bool

max = 0;
bits = TransitionCost;
//IEnumerator<int> ienum = ivalues.GetEnumerator();
//while (ienum.MoveNext())
for (int i = 0; i < ivalues.Length; ++i)
{
//uint val = (uint)ienum.Current;
uint val = (uint)ivalues[i];
if (val > max)
max = val;
Expand Down
5 changes: 0 additions & 5 deletions src/Microsoft.ML.FastTree/Dataset/SparseIntArray.cs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -546,13 +546,9 @@ public unsafe int this[int virtualIndex]
{
get
{
//if (virtualIndex < _index) throw new Exception("Index must move forward");

if (virtualIndex < _nextIndex)
return 0;

//if (virtualIndex >= _array._length) throw new IndexOutOfRangeException();

if (virtualIndex == _nextIndex)
return _array._values[_pos];

Expand All @@ -561,7 +557,6 @@ public unsafe int this[int virtualIndex]
{
while (_pos < _array._values.Length)
{
//_index = _nextIndex;
_nextIndex += pDeltas[_pos];
if (virtualIndex < _nextIndex)
return 0;
Expand Down
2 changes: 0 additions & 2 deletions src/Microsoft.ML.FastTree/FastTreeRanking.cs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -580,7 +580,6 @@ private void SetupSecondaryGains(Arguments args)
_secondaryMetricShare = 0.0;
return;
}
//for (int i = 0; i < _secondaryGains.Length; ++i) _secondaryGains[i] *= cmd.secondaryMetricShare;
_secondaryInverseMaxDCGT = DCGCalculator.MaxDCG(_secondaryGains, Dataset.Boundaries,
new int[] { args.lambdaMartMaxTruncation })[0].Select(d => 1.0 / d).ToArray();
}
Expand Down Expand Up @@ -727,7 +726,6 @@ protected override void GetGradientInOneQuery(int query, int threadIndex)
double inverseMaxDcg = _inverseMaxDcgt[query];
double secondaryInverseMaxDcg = _secondaryMetricShare == 0 ? 0.0 : _secondaryInverseMaxDcgt[query];

//int[] permutation = (threadIndex < 0 ? new int[numDocuments] : _permutationBuffers[threadIndex]);
int[] permutation = _permutationBuffers[threadIndex];

short[] labels = _labels;
Expand Down
1 change: 0 additions & 1 deletion src/Microsoft.ML.FastTree/Microsoft.ML.FastTree.csproj
View file
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,6 @@
<ProjectReference Include="..\Microsoft.ML.Core\Microsoft.ML.Core.csproj" />
<ProjectReference Include="..\Microsoft.ML.CpuMath\Microsoft.ML.CpuMath.csproj" />
<ProjectReference Include="..\Microsoft.ML.Data\Microsoft.ML.Data.csproj" />
<ProjectReference Include="..\Microsoft.ML.InternalStreams\Microsoft.ML.InternalStreams.csproj" />
</ItemGroup>

<ItemGroup>
Expand Down
2 changes: 1 addition & 1 deletion src/Microsoft.ML.FastTree/Training/ScoreTracker.cs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -53,7 +53,7 @@ public void Initialize(ScoreTracker scores1, RegressionTree tree, DocumentPartit
}

//InitScores -initScores can be null in such case the scores are reinitialized to Zero
private void InitializeScores(double[] initScores /* = null */)
private void InitializeScores(double[] initScores)
{
if (initScores == null)
{
Expand Down
Loading

0 comments on commit 669f4fa

Please sign in to comment.