Skip to content
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

Jump to bottom

Android change #1042

Merged
merged 11 commits into from
Mar 28, 2024
Merged

Android change #1042

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
31f9c47
fixed hierarchical clusterization with ward (#998)
daniyalaliev Nov 29, 2023
04e1e88
master commit
daniyalaliev Jan 4, 2024
b8829d8
Merge branch 'neoml-lib:master' into master
daniyalaliev Jan 4, 2024
b3295b7
Merge branch 'neoml-lib:master' into master
daniyalaliev Jan 9, 2024
b122919
Merge branch 'neoml-lib:master' into master
daniyalaliev Jan 15, 2024
16dee71
Merge branch 'neoml-lib:master' into master
daniyalaliev Jan 20, 2024
ff510c0
Merge branch 'neoml-lib:master' into master
daniyalaliev Jan 29, 2024
f754b54
Merge branch 'neoml-lib:master' into master
daniyalaliev Feb 29, 2024
855afbe
Merge branch 'neoml-lib:master' into master
daniyalaliev Mar 28, 2024
a6b44f1
vulkan minor changes
daniyalaliev Mar 28, 2024
15c47c4
metal same change
daniyalaliev Mar 28, 2024
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
Next Next commit
fixed hierarchical clusterization with ward (#998) 
Signed-off-by: Daniyal Aliev <daniial.aliev@abbyy.com>
Co-authored-by: Valeriy Fedyunin <valery.fedyunin@abbyy.com>
  • Loading branch information
@daniyalaliev
daniyalaliev and Valeriy Fedyunin committed Jan 4, 2024
commit 31f9c47df9760ac71aec4e84c1d838d0d857deb9
2 changes: 1 addition & 1 deletion Build/Inc/ProductBuildNumber.h
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
#define VERINFO_PRODUCT_BRANCH "NeoML-master"
#define VERINFO_MAJOR_VERSION 2
#define VERINFO_MINOR_VERSION 0
#define VERINFO_MODIFICATION_NUMBER 206
#define VERINFO_MODIFICATION_NUMBER 210
#define VERINFO_BUILD_NUMBER 0
11 changes: 6 additions & 5 deletions NeoML/src/TraditionalML/NaiveHierarchicalClustering.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -254,7 +254,8 @@ void CNaiveHierarchicalClustering::mergeClusters( int first, int newClusterIndex
if( i == first || clusters[i] == nullptr ) {
continue;
}
const float distance = recalcDistance( *clusters[i], *clusters[first], firstSize, secondSize,
const float distance = recalcDistance( *clusters[i], *clusters[first], firstSize,
secondSize, clusters[i]->GetElementsCount(),
i < first ? distances[i][first] : prevDistances[i],
i < second ? distances[i][second] : distances[second][i],
mergeDistance );
Expand All @@ -275,7 +276,7 @@ void CNaiveHierarchicalClustering::mergeClusters( int first, int newClusterIndex

// Calculates distance between current cluster and merged cluster based on
float CNaiveHierarchicalClustering::recalcDistance( const CCommonCluster& currCluster, const CCommonCluster& mergedCluster,
int firstSize, int secondSize, float currToFirst, float currToSecond, float firstToSecond ) const
int firstSize, int secondSize, int currSize, float currToFirst, float currToSecond, float firstToSecond ) const
{
static_assert( CHierarchicalClustering::L_Count == 5, "L_Count != 5" );

Expand All @@ -298,9 +299,9 @@ float CNaiveHierarchicalClustering::recalcDistance( const CCommonCluster& currCl
return ::fmaxf( currToFirst, currToSecond );
case CHierarchicalClustering::L_Ward:
{
const int mergedSize = firstSize + secondSize;
return ( firstSize * currToFirst + secondSize * currToSecond
- ( firstSize * secondSize * firstToSecond ) / mergedSize ) / mergedSize;
const int sumSize = firstSize + secondSize + currSize;
return ( firstSize + currSize ) * currToFirst / sumSize + ( secondSize + currSize ) * currToSecond / sumSize -
currSize * firstToSecond / sumSize;
}
default:
NeoAssert( false );
Expand Down
2 changes: 1 addition & 1 deletion NeoML/src/TraditionalML/NaiveHierarchicalClustering.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -96,7 +96,7 @@ class CNaiveHierarchicalClustering {
void findNearestClusters( int& first ) const;
void mergeClusters( int first, int newClusterIndex, CArray<CMergeInfo>* dendrogram );
float recalcDistance( const CCommonCluster& currCluster, const CCommonCluster& mergedCluster,
int firstSize, int secondSize, float currToFirst, float currToSecond, float firstToSecond ) const;
int firstSize, int secondSize, int currSize, float currToFirst, float currToSecond, float firstToSecond ) const;
void fillResult( const CFloatMatrixDesc& matrix, CClusteringResult& result, CArray<int>* dendrogramIndices ) const;
};

Expand Down
11 changes: 6 additions & 5 deletions NeoML/src/TraditionalML/NnChainHierarchicalClustering.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,7 +25,7 @@ namespace NeoML {
// Recalculates distance between current and the result of the merge of the first and second clusters
// Doesn't support centroid linkage!
static float recalcDistance( CHierarchicalClustering::TLinkage linkage, TDistanceFunc distance, int firstSize, int secondSize,
float currToFirst, float currToSecond, float firstToSecond )
int currSize, float currToFirst, float currToSecond, float firstToSecond )
{
switch( linkage ) {
case CHierarchicalClustering::L_Single:
Expand All @@ -44,9 +44,9 @@ static float recalcDistance( CHierarchicalClustering::TLinkage linkage, TDistanc
return ::fmaxf( currToFirst, currToSecond );
case CHierarchicalClustering::L_Ward:
{
const int mergedSize = firstSize + secondSize;
return ( firstSize * currToFirst + secondSize * currToSecond
- ( firstSize * secondSize * firstToSecond ) / mergedSize ) / mergedSize;
const int sumSize = firstSize + secondSize + currSize;
return ( firstSize + currSize ) * currToFirst / sumSize + ( secondSize + currSize ) * currToSecond / sumSize -
currSize * firstToSecond / sumSize;
}
case CHierarchicalClustering::L_Centroid:
default:
Expand Down Expand Up @@ -211,7 +211,8 @@ void CNnChainHierarchicalClustering::mergeClusters( int first, int second )
continue;
}
// We can pass ref to any cluster here because linkage isn't centroid
const float distance = recalcDistance( params.Linkage, params.DistanceType, firstSize, secondSize,
const float distance = recalcDistance( params.Linkage, params.DistanceType, firstSize,
secondSize, clusterSizes[i],
i < first ? distances[i][first] : distances[first][i],
i < second ? distances[i][second] : distances[second][i],
mergeDistance );
Expand Down