| Interface | Description |
|---|---|
| Converter<K1 extends org.apache.hadoop.io.Writable,V1 extends org.apache.hadoop.io.Writable,K2 extends org.apache.hadoop.io.Writable,V2 extends org.apache.hadoop.io.Writable> | |
| CudaSupportFunctions |
DESIGN DOCUMENTATION FOR SUPPORTING LOWER PRECISION:
1.
|
| LibMatrixDNNRotate180.Rotate180Worker | |
| MatrixBlockDataInput |
Any data input that is intended to support fast deserialization / read
of entire blocks should implement this interface.
|
| MatrixBlockDataOutput |
Any data output that is intended to support fast serialization / write
of entire blocks should implement this interface.
|
| Class | Description |
|---|---|
| AdaptivePartialBlock |
Writable used in ReblockMR for intermediate results.
|
| AddDummyWeightConverter | |
| BinaryBlockToBinaryCellConverter | |
| BinaryBlockToRowBlockConverter | |
| BinaryBlockToTextCellConverter | |
| BinaryCellToRowBlockConverter | |
| BinaryCellToTextConverter | |
| CM_N_COVCell | |
| CTableMap |
Ctable map is an abstraction for the hashmap used for ctable's hash group-by
because this structure is passed through various interfaces.
|
| DenseBlock |
This DenseBlock is an abstraction for different dense, row-major
matrix formats.
|
| DenseBlockDRB | |
| DenseBlockFactory | |
| DenseBlockLDRB | |
| DnnParameters |
This class is container that stores parameters required for executing following operations:
conv2d, conv2d_backward_data, conv2d_backward_filter, maxpooling, maxpooling_backward
|
| DoublePrecisionCudaSupportFunctions | |
| FrameBlock | |
| FrameBlock.ColumnMetadata | |
| IdenticalConverter | |
| IJV |
Helper class for external key/value exchange.
|
| InputInfo | |
| LibCommonsMath |
Library for matrix operations that need invocation of
Apache Commons Math library.
|
| LibMatrixAgg |
MB:
Library for matrix aggregations including ak+, uak+ for all
combinations of dense and sparse representations, and corrections.
|
| LibMatrixBincell |
Library for binary cellwise operations (incl arithmetic, relational, etc).
|
| LibMatrixCUDA |
All CUDA kernels and library calls are redirected through this class
|
| LibMatrixCuDNN |
This class contains method that invoke CuDNN operations.
|
| LibMatrixCuDNNConvolutionAlgorithm |
This class is a wrapper that contain necessary data structures to invoke
a cudnn convolution* functions (such as cudnnConvolutionForward, etc)
It implements autocloseable to simplify the LibMatrixCuDNN code and also avoids potential memory leaks.
|
| LibMatrixCuDNNInputRowFetcher |
Performs a slice operation: out = in[(n+1):(n+1), 1:numColumns]
|
| LibMatrixCuDNNPoolingDescriptors |
This class is a wrapper that contain necessary data structures to invoke
a cudnn convolution* functions (such as cudnnConvolutionForward, etc)
It implements autocloseable to simplify the LibMatrixCuDNN code and also avoids potential memory leaks.
|
| LibMatrixCuDNNRnnAlgorithm | |
| LibMatrixCuMatMult | |
| LibMatrixDatagen | |
| LibMatrixDNN | |
| LibMatrixDNNConv2d |
This class contains the set of operators used for performing conv2d
|
| LibMatrixDNNHelper | |
| LibMatrixDNNHelper.CellIndex3 | |
| LibMatrixDNNIm2Col |
This class contains the different implementation of im2col operation
|
| LibMatrixDNNPooling |
This class contains the set of operators used for performing pooling
|
| LibMatrixDNNRelu |
This class contains the different implementation of rotate180 operation
|
| LibMatrixDNNRelu.ReluBackward |
Performs the operation: (X gt 0) * dout
|
| LibMatrixDNNRotate180 |
This class contains the different implementation of rotate180 operation
|
| LibMatrixMult |
MB: Library for matrix multiplications including MM, MV, VV for all
combinations of dense, sparse, ultrasparse representations and special
operations such as transpose-self matrix multiplication.
|
| LibMatrixNative | |
| LibMatrixOuterAgg |
ACS:
Purpose of this library is to make some of the unary outer aggregate operator more efficient.
|
| LibMatrixReorg |
MB:
Library for selected matrix reorg operations including special cases
and all combinations of dense and sparse representations.
|
| MatrixBlock | |
| MatrixBlock.SparsityEstimate | |
| MatrixCell | |
| MatrixIndexes |
This represent the indexes to the blocks of the matrix.
|
| MatrixPackedCell | |
| MatrixValue | |
| MatrixValue.CellIndex | |
| MultipleOutputCommitter | |
| OperationsOnMatrixValues | |
| OutputInfo | |
| Pair<K,V> | |
| PartialBlock | |
| RandomMatrixGenerator | |
| SinglePrecisionCudaSupportFunctions | |
| SparseBlock |
This SparseBlock is an abstraction for different sparse matrix formats.
|
| SparseBlockCOO |
SparseBlock implementation that realizes a traditional 'coordinate matrix'
representation, where the entire sparse block is stored as triples in three arrays:
row indexes, column indexes, and values, where row indexes and colunm indexes are
sorted in order to allow binary search.
|
| SparseBlockCSR |
SparseBlock implementation that realizes a traditional 'compressed sparse row'
representation, where the entire sparse block is stored as three arrays: ptr
of length rlen+1 to store offsets per row, and indexes/values of length nnz
to store column indexes and values of non-zero entries.
|
| SparseBlockFactory | |
| SparseBlockMCSR |
SparseBlock implementation that realizes a 'modified compressed sparse row'
representation, where each compressed row is stored as a separate SparseRow
object which provides flexibility for unsorted row appends without the need
for global reshifting of values/indexes but it incurs additional memory
overhead per row for object/array headers per row which also slows down
memory-bound operations due to higher memory bandwidth requirements.
|
| SparseRow |
Base class for sparse row implementations such as sparse
row vectors and sparse scalars (single value per row).
|
| SparseRowScalar | |
| SparseRowVector | |
| Tagged<T extends org.apache.hadoop.io.WritableComparable> | |
| TaggedAdaptivePartialBlock | |
| TaggedFirstSecondIndexes | |
| TaggedFirstSecondIndexes.Comparator |
A Comparator optimized for TaggedFirstSecondIndexes.
|
| TaggedFirstSecondIndexes.FirstIndexPartitioner |
Partition based on the first index.
|
| TaggedFirstSecondIndexes.FirstIndexRangePartitioner |
Partition based on the first index.
|
| TaggedFirstSecondIndexes.TagPartitioner |
Partition based on the first index.
|
| TaggedMatrixBlock | |
| TaggedMatrixCell | |
| TaggedMatrixIndexes | |
| TaggedMatrixPackedCell | |
| TaggedMatrixValue | |
| TaggedTripleIndexes | |
| TextCellToRowBlockConverter | |
| TextToBinaryCellConverter | |
| TripleIndexes | |
| TripleIndexes.Comparator | |
| TripleIndexes.FirstTwoIndexesPartitioner |
Partition based on the first and second index.
|
| UnPaddedOutputFormat<K extends org.apache.hadoop.io.Writable,V extends org.apache.hadoop.io.Writable> | |
| UnPaddedOutputFormat.UnpaddedRecordWriter<K extends org.apache.hadoop.io.Writable,V extends org.apache.hadoop.io.Writable> | |
| WeightedCell | |
| WeightedCellToSortInputConverter | |
| WeightedPair |
| Enum | Description |
|---|---|
| DenseBlock.Type | |
| LibMatrixBincell.BinaryAccessType | |
| LibMatrixDNN.PoolingType | |
| MatrixBlock.BlockType | |
| SparseBlock.Type |
Copyright © 2018 The Apache Software Foundation. All rights reserved.