Skip to content

FastTree: Instantiate feature map for disk transpose and make Generalized Additive Models predictor resilient when feature map is not available. #122

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
shauheen merged 10 commits into from
May 12, 2018
Merged

FastTree: Instantiate feature map for disk transpose and make Generalized Additive Models predictor resilient when feature map is not available. #122

Show file tree
Hide file tree
Changes from 8 commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
0eec9de
Instantiate feature map for disk transpose and make Generalized Addit…
codemzs May 11, 2018
d81d42c
misc.
codemzs May 11, 2018
0da57bd
misc.
codemzs May 11, 2018
119f50f
misc.
codemzs May 11, 2018
afee8ef
misc.
codemzs May 11, 2018
f7a6553
misc.
codemzs May 11, 2018
f8e9aa3
misc.
codemzs May 11, 2018
0763122
misc.
codemzs May 11, 2018
d768f3b
PR comments.
codemzs May 11, 2018
3b248cc
Merge branch 'master' of https://github.com/dotnet/machinelearning in…
codemzs May 11, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 10 additions & 7 deletions src/Microsoft.ML.FastTree/FastTree.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1545,22 +1545,19 @@ private Dataset Construct(RoleMappedData examples, ref int numExamples, int maxB
{
flocks.Add(CreateOneHotFlockCategorical(ch, pending, binnedValues,
lastOn, true));

if (FeatureMap == null)
FeatureMap = Enumerable.Range(0, NumFeatures)
.Where(f => BinUpperBounds[f].Length > 1).ToArray();
}
iFeature = CategoricalFeatureIndices[catRangeIndex + 1] + 1;
catRangeIndex += 2;
}
else
{
GetFeatureValues(cursor, iFeature, getter, ref temp, ref doubleTemp, copier);

double[] upperBounds = BinUpperBounds[iFeature];
double[] upperBounds = BinUpperBounds[iFeature++];
Host.AssertValue(upperBounds);
if (upperBounds.Length == 1)
continue; //trivial feature, skip it.

flocks.Add(CreateSingletonFlock(ch, ref doubleTemp, binnedValues, upperBounds));
iFeature++;
}
}
}
Expand All @@ -1571,10 +1568,16 @@ private Dataset Construct(RoleMappedData examples, ref int numExamples, int maxB
GetFeatureValues(cursor, i, getter, ref temp, ref doubleTemp, copier);
double[] upperBounds = BinUpperBounds[i];
Host.AssertValue(upperBounds);
if (upperBounds.Length == 1)
continue; //trivial feature, skip it.

flocks.Add(CreateSingletonFlock(ch, ref doubleTemp, binnedValues, upperBounds));
}
}

Contracts.Assert(FeatureMap == null);

FeatureMap = Enumerable.Range(0, NumFeatures).Where(f => BinUpperBounds[f].Length > 1).ToArray();
features = flocks.ToArray();
}
}
Expand Down
18 changes: 11 additions & 7 deletions src/Microsoft.ML.FastTree/GamTrainer.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -574,8 +574,7 @@ protected internal GamPredictorBase(IHostEnvironment env, string name, int input
{
Host.CheckValue(trainSet, nameof(trainSet));
Host.CheckParam(trainSet.NumFeatures <= inputLength, nameof(inputLength), "Must be at least as large as dataset number of features");
Host.CheckValue(featureMap, nameof(featureMap));
Host.CheckParam(featureMap.Length == trainSet.NumFeatures, nameof(featureMap), "Not of right size");
Host.CheckParam(featureMap == null || featureMap.Length == trainSet.NumFeatures, nameof(featureMap), "Not of right size");
Host.CheckValue(binEffects, nameof(binEffects));
Host.CheckParam(binEffects.Length == trainSet.NumFeatures, nameof(binEffects), "Not of right size");

Expand All @@ -584,12 +583,17 @@ protected internal GamPredictorBase(IHostEnvironment env, string name, int input
_numFeatures = binEffects.Length;
_inputType = new VectorType(NumberType.Float, _inputLength);
_featureMap = featureMap;
_inputFeatureToDatasetFeatureMap = new Dictionary<int, int>(featureMap.Length);
for (int i = 0; i < featureMap.Length; i++)

//No features were filtered.
if (_featureMap == null)
_featureMap = Enumerable.Range(0, trainSet.NumFeatures).ToArray();

@TomFinley TomFinley May 11, 2018 •
edited by codemzs
Loading

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Utils.GetIdentityPermutation? #Resolved


_inputFeatureToDatasetFeatureMap = new Dictionary<int, int>(_featureMap.Length);
for (int i = 0; i < _featureMap.Length; i++)
{
Host.CheckParam(0 <= featureMap[i] && featureMap[i] < inputLength, nameof(featureMap), "Contains out of range feature vaule");
Host.CheckParam(!_inputFeatureToDatasetFeatureMap.ContainsValue(featureMap[i]), nameof(featureMap), "Contains duplicate mappings");
_inputFeatureToDatasetFeatureMap[featureMap[i]] = i;
Host.CheckParam(0 <= _featureMap[i] && _featureMap[i] < inputLength, nameof(_featureMap), "Contains out of range feature vaule");
Host.CheckParam(!_inputFeatureToDatasetFeatureMap.ContainsValue(_featureMap[i]), nameof(_featureMap), "Contains duplicate mappings");
_inputFeatureToDatasetFeatureMap[_featureMap[i]] = i;
}

//keep only bin effect and upperbounds where the effect changes.
Expand Down