|
Cyclops Tensor Framework
parallel arithmetic on multidimensional arrays
|
|
Cyclops Tensor Framework
parallel arithmetic on multidimensional arrays
|
internal distributed tensor class More...
#include <untyped_tensor.h>
Public Member Functions | |
| CTF::Idx_Tensor | operator[] (char const *idx_map) |
| associated an index map with the tensor for future operation More... | |
| tensor () | |
| ~tensor () | |
| class free self More... | |
| void | free_self () |
| destructor More... | |
| tensor (algstrct const *sr, int order, int const *edge_len, int const *sym, CTF::World *wrld, bool alloc_data=true, char const *name=NULL, bool profile=1, bool is_sparse=0) | |
| defines a tensor object with some mapping (if alloc_data) More... | |
| tensor (algstrct const *sr, int order, bool is_sparse, int const *edge_len, int const *sym, CTF::World *wrld, char const *idx, CTF::Idx_Partition const &prl, CTF::Idx_Partition const &blk, char const *name=NULL, bool profile=1) | |
| defines a tensor object with some mapping (if alloc_data) More... | |
| tensor (tensor const *other, bool copy=1, bool alloc_data=1) | |
| creates tensor copy, unfolds other if other is folded More... | |
| tensor (tensor *other, int const *new_sym) | |
| repacks the tensor other to a different symmetry (assumes existing data contains the symmetry and keeps only values with indices in increasing order) WARN: LIMITATION: new_sym must cause unidirectional structural changes, i.e. {NS,NS}->{SY,NS} OK, {SY,NS}->{NS,NS} OK, {NS,NS,SY,NS}->{SY,NS,NS,NS} NOT OK! More... | |
| int * | calc_phase () const |
| compute the cyclic phase of each tensor dimension More... | |
| int | calc_tot_phase () const |
| calculate the total number of blocks of the tensor More... | |
| int64_t | calc_nvirt () const |
| calculate virtualization factor of tensor return virtualization factor More... | |
| int64_t | calc_npe () const |
| calculate the number of processes this tensor is distributed over return number of processes owning a block of the tensor More... | |
| void | set_padding () |
| sets padding and local size of a tensor given a mapping More... | |
| int | set_zero () |
| elects a mapping and sets tensor data to zero More... | |
| int | set (char const *val) |
| sets tensor data to val More... | |
| int | zero_out_padding () |
| sets padded portion of tensor to zero (this should be maintained internally) More... | |
| void | scale_diagonals (int const *sym_mask) |
| scales each element by 1/(number of entries equivalent to it after permutation of indices for which sym_mask is 1) More... | |
| void | addinv () |
| void | print_map (FILE *stream=stdout, bool allcall=1) const |
| displays mapping information More... | |
| void | set_name (char const *name) |
| set the tensor name More... | |
| char const * | get_name () const |
| get the tensor name More... | |
| void | profile_on () |
| turn on profiling More... | |
| void | profile_off () |
| turn off profiling More... | |
| void | get_raw_data (char **data, int64_t *size) const |
| get raw data pointer without copy WARNING: includes padding More... | |
| int | write (int64_t num_pair, char const *alpha, char const *beta, char *mapped_data, char const rw='w') |
| Add tensor data new=alpha*new+beta*old with <key, value> pairs where key is the global index for the value. More... | |
| void | write (int64_t num_pair, char const *alpha, char const *beta, int64_t const *inds, char const *data) |
| Add tensor data new=alpha*new+beta*old with <key, value> pairs where key is the global index for the value. More... | |
| void | read (int64_t num_pair, char const *alpha, char const *beta, int64_t const *inds, char *data) |
| read tensor data with <key, value> pairs where key is the global index for the value, which gets filled in with beta times the old values plus alpha times the values read from the tensor. More... | |
| int | read (int64_t num_pair, char const *alpha, char const *beta, char *mapped_data) |
| read tensor data with <key, value> pairs where key is the global index for the value, which gets filled in with beta times the old values plus alpha times the values read from the tensor. More... | |
| char * | read (char const *idx, CTF::Idx_Partition const &prl, CTF::Idx_Partition const &blk, bool unpack) |
| returns local data of tensor with parallel distribution prl and local blocking blk More... | |
| int | read (int64_t num_pair, char *mapped_data) |
| read tensor data with <key, value> pairs where key is the global index for the value, which gets filled in. More... | |
| int64_t | get_tot_size (bool packed) |
| get number of elements in whole tensor More... | |
| int | allread (int64_t *num_pair, char **all_data, bool unpack) |
| read entire tensor with each processor (in packed layout). WARNING: will use an 'unscalable' amount of memory. More... | |
| int | allread (int64_t *num_pair, char *all_data, bool unpack=true) |
| read entire tensor with each processor (in packed layout). WARNING: will use an 'unscalable' amount of memory. More... | |
| void | slice (int const *offsets_B, int const *ends_B, char const *beta, tensor *A, int const *offsets_A, int const *ends_A, char const *alpha) |
| accumulates out a slice (block) of this tensor = B B[offsets,ends)=beta*B[offsets,ends) + alpha*A[offsets_A,ends_A) More... | |
| int | permute (tensor *A, int *const *permutation_A, char const *alpha, int *const *permutation_B, char const *beta) |
| int | sparsify (char const *threshold=NULL, bool take_abs=true) |
| reduce tensor to sparse format, storing only nonzero data, or data above a specified threshold. makes dense tensors sparse. cleans sparse tensors of any 'computed' zeros. More... | |
| int | sparsify (std::function< bool(char const *)> f) |
| sparsifies tensor keeping only values v such that filter(v) = true More... | |
| int | read_local (int64_t *num_pair, char **mapped_data, bool unpack_sym=false) const |
| read tensor data pairs local to processor including those with zero values WARNING: for sparse tensors this includes the zeros to maintain consistency with the behavior for dense tensors, use read_local_nnz to get only nonzeros More... | |
| int | read_local_nnz (int64_t *num_pair, char **mapped_data, bool unpack_sym=false) const |
| read tensor data pairs local to processor that have nonzero values More... | |
| void | read_local (int64_t *num_pair, int64_t **inds, char **data, bool unpack_sym=false) const |
| read tensor data pairs local to processor including those with zero values WARNING: for sparse tensors this includes the zeros to maintain consistency with the behavior for dense tensors, use read_local_nnz to get only nonzeros More... | |
| void | read_local_nnz (int64_t *num_pair, int64_t **inds, char **data, bool unpack_sym=false) const |
| read tensor data pairs local to processor that have nonzero values More... | |
| int | align (tensor const *B) |
| align mapping of thisa tensor to that of B More... | |
| int | reduce_sum (char *result) |
| Performs an elementwise summation reduction on a tensor. More... | |
| int | reduce_sum (char *result, algstrct const *sr_other) |
| Performs an elementwise summation reduction on a tensor with summation defined by sr_other. More... | |
| int | reduce_sumabs (char *result) |
| Performs an elementwise absolute value summation reduction on a tensor. More... | |
| int | reduce_sumabs (char *result, algstrct const *sr_other) |
| Performs an elementwise absolute value summation reduction on a tensor. More... | |
| int | reduce_sumsq (char *result) |
| computes the sum of squares of the elements More... | |
| int | get_max_abs (int n, char *data) const |
| obtains the largest n elements (in absolute value) of the tensor More... | |
| void | print (FILE *fp=stdout, char const *cutoff=NULL) const |
| prints tensor data to file using process 0 More... | |
| void | prnt () const |
| void | compare (const tensor *A, FILE *fp, char const *cutoff) |
| prints two sets of tensor data side-by-side to file using process 0 More... | |
| void | orient_subworld (CTF::World *greater_world, int &bw_mirror_rank, int &fw_mirror_rank, distribution *&odst, char **sub_buffer_) |
| maps data from this world (subcomm) to the correct order of processors with respect to a parent (greater_world) comm More... | |
| void | add_to_subworld (tensor *tsr_sub, char const *alpha, char const *beta) |
| accumulates this tensor to a tensor object defined on a different world More... | |
| void | add_from_subworld (tensor *tsr_sub, char const *alpha, char const *beta) |
| accumulates this tensor from a tensor object defined on a different world More... | |
| void | unfold (bool was_mod=0) |
| undo the folding of a local tensor block unsets is_folded and deletes rec_tsr More... | |
| void | remove_fold () |
| removes folding without doing transpose unsets is_folded and deletes rec_tsr More... | |
| double | est_time_unfold () |
| estimate cost of potential transpose involved in undoing the folding of a local tensor block More... | |
| void | fold (int nfold, int const *fold_idx, int const *idx_map, int *all_fdim, int **all_flen) |
| fold a tensor by putting the symmetry-preserved portion in the leading dimensions of the tensor sets is_folded and creates rec_tsr with aliased data More... | |
| void | pull_alias (tensor const *other) |
| pulls data from an tensor with an aliased buffer More... | |
| void | clear_mapping () |
| zeros out mapping More... | |
| int | redistribute (distribution const &old_dist, int const *old_offsets=NULL, int *const *old_permutation=NULL, int const *new_offsets=NULL, int *const *new_permutation=NULL) |
| permutes the data of a tensor to its new layout More... | |
| double | est_redist_time (distribution const &old_dist, double nnz_frac) |
| int64_t | get_redist_mem (distribution const &old_dist, double nnz_frac) |
| int | map_tensor_rem (int num_phys_dims, CommData *phys_comm, int fill=0) |
| map the remainder of a tensor More... | |
| int | extract_diag (int const *idx_map, int rw, tensor *&new_tsr, int **idx_map_new) |
| extracts the diagonal of a tensor if the index map specifies to do so More... | |
| void | set_sym (int const *sym) |
| sets symmetry, WARNING: for internal use only !!!! More... | |
| void | set_new_nnz_glb (int64_t const *nnz_blk) |
| sets the number of nonzeros both locally (nnz_loc) and overall globally (nnz_tot) More... | |
| void | spmatricize (int m, int n, int nrow_idx, bool csr) |
| transposes local data in preparation for summation or contraction, transforms to COO or CSR format for sparse More... | |
| void | despmatricize (int nrow_idx, bool csr) |
| transposes back local data from sparse matrix format to key-value pair format More... | |
| void | leave_home_with_buffer () |
| degister home buffer More... | |
| void | register_size (int64_t size) |
| register buffer allocation for this tensor More... | |
| void | deregister_size () |
| deregister buffer allocation for this tensor More... | |
| void | write_dense_to_file (MPI_File &file, int64_t offset=0) |
| write all tensor data to binary file in element order, unpacking from sparse or symmetric formats More... | |
| void | write_dense_to_file (char const *filename) |
| write all tensor data to binary file in element order, unpacking from sparse or symmetric formats More... | |
| void | read_dense_from_file (MPI_File &file, int64_t offset=0) |
| read all tensor data from binary file in element order, which should be stored as nonsymmetric and dense as done in write_dense_to_file() More... | |
| void | read_dense_from_file (char const *filename) |
| read all tensor data from binary file in element order, which should be stored as nonsymmetric and dense as done in write_dense_to_file() More... | |
| template<typename dtype_A , typename dtype_B > | |
| void | conv_type (tensor *B) |
| convert this tensor from dtype_A to dtype_B and store the result in B (primarily needed for python interface) More... | |
| template<typename dtype_A , typename dtype_B > | |
| void | exp_helper (tensor *A) |
| template<typename dtype > | |
| void | compare_elementwise (tensor *A, tensor *B) |
| do an elementwise comparison(==) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor) More... | |
| template<typename dtype > | |
| void | not_equals (tensor *A, tensor *B) |
| do an elementwise comparison(!=) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor) More... | |
| template<typename dtype > | |
| void | smaller_than (tensor *A, tensor *B) |
| do an elementwise comparison(<) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor) More... | |
| template<typename dtype > | |
| void | smaller_equal_than (tensor *A, tensor *B) |
| do an elementwise comparison(<=) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor) More... | |
| template<typename dtype > | |
| void | larger_than (tensor *A, tensor *B) |
| do an elementwise comparison(>) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor) More... | |
| template<typename dtype > | |
| void | larger_equal_than (tensor *A, tensor *B) |
| do an elementwise comparison(>=) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor) More... | |
| template<typename dtype > | |
| void | true_divide (tensor *A) |
| tensor * | self_reduce (int const *idx_A, int **new_idx_A, int order_B, int const *idx_B, int **new_idx_B, int order_C=0, int const *idx_C=NULL, int **new_idx_C=NULL) |
| performs a partial reduction on the tensor (used in summation and contraction) More... | |
Data Fields | |
| CTF::World * | wrld |
| distributed processor context on which tensor is defined More... | |
| algstrct * | sr |
| algstrct on which tensor elements and operations are defined More... | |
| int * | sym |
| symmetries among tensor dimensions More... | |
| int | order |
| number of tensor dimensions More... | |
| int * | lens |
| unpadded tensor edge lengths More... | |
| int * | pad_edge_len |
| padded tensor edge lengths More... | |
| int * | padding |
| padding along each edge length (less than distribution phase) More... | |
| char * | name |
| name given to tensor More... | |
| int | is_scp_padded |
| whether tensor data has additional padding More... | |
| int * | scp_padding |
| additional padding, may be greater than ScaLAPACK phase More... | |
| int * | sym_table |
| order-by-order table of dimensional symmetry relations More... | |
| bool | is_mapped |
| whether a mapping has been selected More... | |
| topology * | topo |
| topology to which the tensor is mapped More... | |
| mapping * | edge_map |
| mappings of each tensor dimension onto topology dimensions More... | |
| int64_t | size |
| current size of local tensor data chunk (mapping-dependent) More... | |
| int64_t | registered_alloc_size |
| size CTF keeps track of for memory usage More... | |
| bool | is_folded |
| whether the data is folded/transposed into a (lower-order) tensor More... | |
| int * | inner_ordering |
| ordering of the dimensions according to which the tensori s folded More... | |
| tensor * | rec_tsr |
| representation of folded tensor (shares data pointer) More... | |
| bool | is_cyclic |
| whether the tensor data is cyclically distributed (blocked if false) More... | |
| bool | is_data_aliased |
| whether the tensor data is an alias of another tensor object's data More... | |
| tensor * | slay |
| tensor object associated with tensor object whose data pointer needs to be preserved, needed for ScaLAPACK wrapper FIXME: home buffer should presumably take care of this... More... | |
| bool | has_zero_edge_len |
| if true tensor has a zero edge length, so is zero, which short-cuts stuff More... | |
| char * | data |
| tensor data, either the data or the key-value pairs should exist at any given time More... | |
| bool | has_home |
| whether the tensor has a home mapping/buffer More... | |
| char * | home_buffer |
| buffer associated with home mapping of tensor, to which it is returned More... | |
| int64_t | home_size |
| size of home buffer More... | |
| bool | is_home |
| whether the latest tensor data is in the home buffer More... | |
| bool | left_home_transp |
| whether the tensor left home to transpose More... | |
| bool | profile |
| whether profiling should be done for contractions/sums involving this tensor More... | |
| bool | is_sparse |
| whether only the non-zero elements of the tensor are stored More... | |
| bool | is_csr |
| whether CSR or COO if folded More... | |
| int | nrow_idx |
| how many modes are folded into matricized row More... | |
| int64_t | nnz_loc |
| number of local nonzero elements More... | |
| int64_t | nnz_tot |
| maximum number of local nonzero elements over all procs More... | |
| int64_t * | nnz_blk |
| nonzero elements in each block owned locally More... | |
Protected Member Functions | |
| void | init (algstrct const *sr, int order, int const *edge_len, int const *sym, CTF::World *wrld, bool alloc_data, char const *name, bool profile, bool is_sparse) |
| initializes tensor data More... | |
| PairIterator | read_all_pairs (int64_t *num_pair, bool unpack) |
| read all pairs with each processor (packed) More... | |
| void | copy_tensor_data (tensor const *other) |
| copies all tensor data from other More... | |
| void | set_distribution (char const *idx, CTF::Idx_Partition const &prl, CTF::Idx_Partition const &blk) |
| set edge mappings as specified More... | |
internal distributed tensor class
Definition at line 20 of file untyped_tensor.h.
| CTF_int::tensor::tensor | ( | ) |
Definition at line 36 of file untyped_tensor.cxx.
Referenced by CTF::Tensor< dtype >::add_from_subworld(), and CTF::Tensor< dtype >::add_to_subworld().
| CTF_int::tensor::~tensor | ( | ) |
class free self
Definition at line 78 of file untyped_tensor.cxx.
| CTF_int::tensor::tensor | ( | algstrct const * | sr, |
| int | order, | ||
| int const * | edge_len, | ||
| int const * | sym, | ||
| CTF::World * | wrld, | ||
| bool | alloc_data = true, |
||
| char const * | name = NULL, |
||
| bool | profile = 1, |
||
| bool | is_sparse = 0 |
||
| ) |
defines a tensor object with some mapping (if alloc_data)
| [in] | sr | defines the tensor arithmetic for this tensor |
| [in] | order | number of dimensions of tensor |
| [in] | edge_len | edge lengths of tensor |
| [in] | sym | symmetries of tensor (e.g. symmetric matrix -> sym={SY, NS}) |
| [in] | wrld | a distributed context for the tensor to live in |
| [in] | alloc_data | whether to allocate and set data to zero immediately |
| [in] | name | an optionary name for the tensor |
| [in] | profile | set to 1 to profile contractions involving this tensor |
| [in] | is_sparse | set to 1 to store only nontrivial tensor elements |
Definition at line 82 of file untyped_tensor.cxx.
| CTF_int::tensor::tensor | ( | algstrct const * | sr, |
| int | order, | ||
| bool | is_sparse, | ||
| int const * | edge_len, | ||
| int const * | sym, | ||
| CTF::World * | wrld, | ||
| char const * | idx, | ||
| CTF::Idx_Partition const & | prl, | ||
| CTF::Idx_Partition const & | blk, | ||
| char const * | name = NULL, |
||
| bool | profile = 1 |
||
| ) |
defines a tensor object with some mapping (if alloc_data)
| [in] | sr | defines the tensor arithmetic for this tensor |
| [in] | order | number of dimensions of tensor |
| [in] | is_sparse | whether to make tensor sparse |
| [in] | edge_len | edge lengths of tensor |
| [in] | sym | symmetries of tensor (e.g. symmetric matrix -> sym={SY, NS}) |
| [in] | wrld | a distributed context for the tensor to live in |
| [in] | idx | assignment of characters to each dim |
| [in] | prl | mesh processor topology with character labels |
| [in] | blk | local blocking with processor labels |
| [in] | name | an optionary name for the tensor |
| [in] | profile | set to 1 to profile contractions involving this tensor |
Definition at line 94 of file untyped_tensor.cxx.
References CTF_int::algstrct::addid(), CTF_int::alloc(), CTF_int::algstrct::alloc(), CTF_int::accumulatable::el_size, and CTF_int::algstrct::set().
| CTF_int::tensor::tensor | ( | tensor const * | other, |
| bool | copy = 1, |
||
| bool | alloc_data = 1 |
||
| ) |
creates tensor copy, unfolds other if other is folded
| [in] | other | tensor to copy |
| [in] | copy | whether to copy mapping and data |
| [in] | alloc_data | whether th allocate data |
Definition at line 136 of file untyped_tensor.cxx.
References CTF_int::alloc(), CTF_int::cdealloc(), DPRINTF, has_zero_edge_len, is_sparse, lens, name, order, profile, CTF::World::rank, sr, sym, and wrld.
| CTF_int::tensor::tensor | ( | tensor * | other, |
| int const * | new_sym | ||
| ) |
repacks the tensor other to a different symmetry (assumes existing data contains the symmetry and keeps only values with indices in increasing order) WARN: LIMITATION: new_sym must cause unidirectional structural changes, i.e. {NS,NS}->{SY,NS} OK, {SY,NS}->{NS,NS} OK, {NS,NS,SY,NS}->{SY,NS,NS,NS} NOT OK!
| [in] | other | tensor to copy |
| [in] | new_sym | new symmetry array (replaces this->sym) |
Definition at line 157 of file untyped_tensor.cxx.
References CTF_int::alloc(), CTF_int::cdealloc(), DPRINTF, has_zero_edge_len, is_sparse, lens, name, NS, order, profile, CTF::World::rank, sr, CTF_int::summation::sum_tensors(), sym, and wrld.
| void CTF_int::tensor::add_from_subworld | ( | tensor * | tsr_sub, |
| char const * | alpha, | ||
| char const * | beta | ||
| ) |
accumulates this tensor from a tensor object defined on a different world
| [in] | tsr_sub | id of tensor on a subcomm of this CTF inst |
| [in] | alpha | scaling factor for this tensor |
| [in] | beta | scaling factor for tensor tsr |
Definition at line 1037 of file untyped_tensor.cxx.
References CTF_int::cyclic_reshuffle(), order, and orient_subworld().
Referenced by CTF::Tensor< dtype >::add_from_subworld().
| void CTF_int::tensor::add_to_subworld | ( | tensor * | tsr_sub, |
| char const * | alpha, | ||
| char const * | beta | ||
| ) |
accumulates this tensor to a tensor object defined on a different world
| [in] | tsr_sub | tensor on a subcomm of this world |
| [in] | alpha | scaling factor for this tensor |
| [in] | beta | scaling factor for tensor tsr |
Definition at line 991 of file untyped_tensor.cxx.
References ASSERT, CTF_int::cyclic_reshuffle(), data, CTF_int::algstrct::mdtype(), order, orient_subworld(), CTF_int::distribution::size, slice(), and sr.
Referenced by CTF::Tensor< dtype >::add_to_subworld(), and CTF::Schedule::partition_and_execute().
| void CTF_int::tensor::addinv | ( | ) |
Definition at line 2682 of file untyped_tensor.cxx.
Referenced by CTF_int::summation::estimate_time(), force_integration(), and force_integration_sparse().
| int CTF_int::tensor::align | ( | tensor const * | B | ) |
align mapping of thisa tensor to that of B
brief copy A into this (B). Realloc if necessary param[in] A tensor to copy
| [in] | B | tensor handle of B |
Definition at line 1749 of file untyped_tensor.cxx.
References ASSERT, CTF_int::comp_dim_map(), CTF_int::copy_mapping(), edge_map, is_folded, order, CTF_int::SUCCESS, topo, and wrld.
Referenced by CTF::Tensor< dtype >::align(), and compare().
| int CTF_int::tensor::allread | ( | int64_t * | num_pair, |
| char ** | all_data, | ||
| bool | unpack | ||
| ) |
read entire tensor with each processor (in packed layout). WARNING: will use an 'unscalable' amount of memory.
| [out] | num_pair | number of values read |
| [in,out] | all_data | values read (allocated by library) |
| [in] | unpack | if true any symmetric tensor is unpacked, otherwise only unique elements are read |
Definition at line 1724 of file untyped_tensor.cxx.
References CTF_int::PairIterator::ptr, CTF_int::PairIterator::read_val(), and CTF_int::SUCCESS.
Referenced by CTF::Tensor< dtype >::read_all().
| int CTF_int::tensor::allread | ( | int64_t * | num_pair, |
| char * | all_data, | ||
| bool | unpack = true |
||
| ) |
read entire tensor with each processor (in packed layout). WARNING: will use an 'unscalable' amount of memory.
| [out] | num_pair | number of values read |
| [in,out] | all_data | preallocated mapped_data values read |
| [in] | unpack | if true any symmetric tensor is unpacked, otherwise only unique elements are read |
Definition at line 1738 of file untyped_tensor.cxx.
References CTF_int::PairIterator::read_val(), and CTF_int::SUCCESS.
| int64_t CTF_int::tensor::calc_npe | ( | ) | const |
calculate the number of processes this tensor is distributed over return number of processes owning a block of the tensor
Definition at line 435 of file untyped_tensor.cxx.
References CTF_int::mapping::child, CTF_int::mapping::has_child, CTF_int::mapping::np, CTF_int::PHYSICAL_MAP, and CTF_int::mapping::type.
Referenced by CTF_int::get_len_ordering().
| int64_t CTF_int::tensor::calc_nvirt | ( | ) | const |
calculate virtualization factor of tensor return virtualization factor
Definition at line 417 of file untyped_tensor.cxx.
References CTF_int::mapping::child, CTF_int::mapping::has_child, CTF_int::mapping::np, CTF_int::mapping::type, and CTF_int::VIRTUAL_MAP.
Referenced by copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), CTF_int::nosym_transpose(), CTF_int::symmetrize(), and CTF_int::tspsum::tspsum().
| int * CTF_int::tensor::calc_phase | ( | ) | const |
compute the cyclic phase of each tensor dimension
Definition at line 394 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), and CTF_int::mapping::calc_phase().
| int CTF_int::tensor::calc_tot_phase | ( | ) | const |
calculate the total number of blocks of the tensor
Definition at line 406 of file untyped_tensor.cxx.
References CTF_int::cdealloc().
| void CTF_int::tensor::clear_mapping | ( | ) |
zeros out mapping
Definition at line 2189 of file untyped_tensor.cxx.
References CTF_int::mapping::clear().
Referenced by CTF_int::desymmetrize(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), and read().
| void CTF_int::tensor::compare | ( | const tensor * | A, |
| FILE * | fp, | ||
| char const * | cutoff | ||
| ) |
prints two sets of tensor data side-by-side to file using process 0
| [in] | A | tensor to compare against |
| [in] | fp | file to print to e.g. stdout |
| [in] | cutoff | do not print values of absolute value smaller than this |
Definition at line 1930 of file untyped_tensor.cxx.
References CTF_int::algstrct::abs, align(), CTF_int::alloc_ptr(), CTF_int::cdealloc(), CTF::World::cdt, CTF_int::CommData::cm, CTF_int::accumulatable::el_size, CTF_int::algstrct::isequal(), lens, CTF_int::algstrct::min(), CTF_int::CommData::np, order, CTF_int::algstrct::pair_alloc(), CTF_int::algstrct::pair_size(), CTF_int::algstrct::print(), print_map(), CTF_int::CommData::rank, read_local(), CTF_int::PairIterator::sort(), sr, and wrld.
Referenced by CTF::Tensor< dtype >::compare().
| template void CTF_int::tensor::compare_elementwise< bool > | ( | tensor * | A, |
| tensor * | B | ||
| ) |
do an elementwise comparison(==) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor)
| [in] | A | first operand |
| [in] | B | second operand |
Definition at line 27 of file untyped_tensor_tmpl.h.
References ctf.core::a, ctf.core::b, ctf.core::dtype, CTF_int::accumulatable::el_size, operator[](), order, and sr.
| void CTF_int::tensor::conv_type | ( | tensor * | B | ) |
convert this tensor from dtype_A to dtype_B and store the result in B (primarily needed for python interface)
| [in] | B | output tensor |
Definition at line 7 of file untyped_tensor_tmpl.h.
References ctf.core::a, and order.
|
protected |
copies all tensor data from other
| [in] | other | tensor to copy from |
Definition at line 210 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), ASSERT, calc_nvirt(), CTF_int::cdealloc(), CTF_int::copy_mapping(), data, edge_map, has_home, home_buffer, home_size, inner_ordering, is_cyclic, is_folded, is_home, is_mapped, is_sparse, lens, name, nnz_blk, nnz_loc, nnz_tot, order, pad_edge_len, padding, rec_tsr, size, sr, sym, topo, and wrld.
Referenced by CTF::Tensor< dtype >::operator=().
| void CTF_int::tensor::deregister_size | ( | ) |
deregister buffer allocation for this tensor
Definition at line 2889 of file untyped_tensor.cxx.
References CTF_int::inc_tot_mem_used().
| void CTF_int::tensor::despmatricize | ( | int | nrow_idx, |
| bool | csr | ||
| ) |
transposes back local data from sparse matrix format to key-value pair format
| [in] | nrow_idx | number of indices to fold into column |
| [in] | csr | whether to go from csr (1) or coo (0) layout |
Definition at line 2790 of file untyped_tensor.cxx.
References CTF_int::COO_Matrix::all_data, ASSERT, CTF_int::cdealloc(), CTF_int::COO_Matrix::get_data(), CTF_int::COO_Matrix::nnz(), CTF_int::CSR_Matrix::nnz(), TAU_FSTART, and TAU_FSTOP.
| double CTF_int::tensor::est_redist_time | ( | distribution const & | old_dist, |
| double | nnz_frac | ||
| ) |
Definition at line 2354 of file untyped_tensor.cxx.
References CTF_int::blres_est_time(), CTF_int::can_block_reshuffle(), CTF_int::dgtog_est_time(), CTF_int::distribution::phase, CTF_int::distribution::size, CTF_int::spredist_est_time(), and CTF_int::distribution::virt_phase.
Referenced by CTF_int::get_len_ordering().
| double CTF_int::tensor::est_time_unfold | ( | ) |
estimate cost of potential transpose involved in undoing the folding of a local tensor block
Definition at line 2075 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), CTF_int::calc_dim(), CTF_int::cdealloc(), CTF_int::est_time_transp(), NS, and CTF_int::sy_packed_size().
| template void CTF_int::tensor::exp_helper< int32_t, float > | ( | tensor * | A | ) |
function store the e**value in tensor A into this (primarily needed for python interface)
Definition at line 18 of file untyped_tensor_tmpl.h.
References ctf.core::a, ctf.core::exp(), operator[](), and order.
| int CTF_int::tensor::extract_diag | ( | int const * | idx_map, |
| int | rw, | ||
| tensor *& | new_tsr, | ||
| int ** | idx_map_new | ||
| ) |
extracts the diagonal of a tensor if the index map specifies to do so
| [in] | idx_map | index map of tensor for this operation |
| [in] | rw | if 1 this writes to the diagonal, if 0 it reads the diagonal |
| [in,out] | new_tsr | if rw=1 this will be output as new tensor if rw=0 this should be input as the tensor of the extracted diagonal |
| [out] | idx_map_new | if rw=1 this will be the new index map |
Definition at line 2456 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), CTF_int::cdealloc(), CTF_int::summation::execute(), CTF_int::PairIterator::k(), CTF_int::NEGATIVE, NS, read_local_nnz(), CTF_int::SUCCESS, ctf.core::sum(), write(), and CTF_int::PairIterator::write_val().
Referenced by CTF_int::summation::estimate_time(), CTF_int::get_len_ordering(), and CTF_int::summation::sum_tensors().
| void CTF_int::tensor::fold | ( | int | nfold, |
| int const * | fold_idx, | ||
| int const * | idx_map, | ||
| int * | all_fdim, | ||
| int ** | all_flen | ||
| ) |
fold a tensor by putting the symmetry-preserved portion in the leading dimensions of the tensor sets is_folded and creates rec_tsr with aliased data
| [in] | nfold | number of global indices we are folding |
| [in] | fold_idx | which global indices we are folding |
| [in] | idx_map | how this tensor indices map to the global indices |
| [out] | all_fdim | number of dimensions including unfolded dimensions |
| [out] | all_flen | edge lengths including unfolded dimensions |
Definition at line 2101 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), CTF_int::calc_dim(), CTF_int::cdealloc(), NS, and CTF_int::sy_packed_size().
| void CTF_int::tensor::free_self | ( | ) |
destructor
Definition at line 40 of file untyped_tensor.cxx.
References CTF_int::cdealloc(), and DPRINTF.
Referenced by ao_mo_transf_naive(), ao_mo_transf_slice(), CTF::Scalar< dtype >::operator=(), and CTF::Tensor< dtype >::operator=().
| int CTF_int::tensor::get_max_abs | ( | int | n, |
| char * | data | ||
| ) | const |
obtains the largest n elements (in absolute value) of the tensor
| [in] | n | number of elements to fill |
| [in,out] | data | preallocated array of size at least n, in which to put the elements |
Referenced by CTF::Tensor< dtype >::get_max_abs().
| char const * CTF_int::tensor::get_name | ( | ) | const |
| void CTF_int::tensor::get_raw_data | ( | char ** | data, |
| int64_t * | size | ||
| ) | const |
get raw data pointer without copy WARNING: includes padding
| [out] | data | raw local data in char * format |
| [out] | size | number of elements in data |
Definition at line 715 of file untyped_tensor.cxx.
Referenced by CTF::Idx_Tensor::execute().
| int64_t CTF_int::tensor::get_redist_mem | ( | distribution const & | old_dist, |
| double | nnz_frac | ||
| ) |
Definition at line 2385 of file untyped_tensor.cxx.
References CTF_int::can_block_reshuffle(), CTF_int::distribution::phase, and CTF_int::distribution::size.
Referenced by CTF_int::get_len_ordering().
| int64_t CTF_int::tensor::get_tot_size | ( | bool | packed = false | ) |
get number of elements in whole tensor
| [in] | packed | if false (default) ignore symmetry |
Definition at line 1712 of file untyped_tensor.cxx.
References CTF_int::packed_size().
|
protected |
initializes tensor data
| [in] | sr | defines the tensor arithmetic for this tensor |
| [in] | order | number of dimensions of tensor |
| [in] | edge_len | edge lengths of tensor |
| [in] | sym | symmetries of tensor (e.g. symmetric matrix -> sym={SY, NS}) |
| [in] | wrld | a distributed context for the tensor to live in |
| [in] | alloc_data | set to 1 if tensor should be mapped and data buffer allocated |
| [in] | name | an optionary name for the tensor |
| [in] | profile | set to 1 to profile contractions involving this tensor |
| [in] | is_sparse | set to 1 to store only nontrivial tensor elements |
Definition at line 306 of file untyped_tensor.cxx.
References CTF_int::alloc(), CTF_int::alloc_ptr(), ASSERT, CTF_int::algstrct::clone(), DPRINTF, CTF_int::NOT_MAPPED, CTF_int::SUCCESS, TAU_FSTART, TAU_FSTOP, and CTF_int::mapping::type.
Referenced by CTF::Scalar< dtype >::operator=(), and CTF::Tensor< dtype >::operator=().
| template void CTF_int::tensor::larger_equal_than< bool > | ( | tensor * | A, |
| tensor * | B | ||
| ) |
do an elementwise comparison(>=) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor)
| [in] | A | first operand |
| [in] | B | second operand |
Definition at line 87 of file untyped_tensor_tmpl.h.
References ctf.core::a, ctf.core::b, ctf.core::dtype, CTF_int::accumulatable::el_size, operator[](), order, and sr.
| template void CTF_int::tensor::larger_than< bool > | ( | tensor * | A, |
| tensor * | B | ||
| ) |
do an elementwise comparison(>) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor)
| [in] | A | first operand |
| [in] | B | second operand |
Definition at line 75 of file untyped_tensor_tmpl.h.
References ctf.core::a, ctf.core::b, ctf.core::dtype, CTF_int::accumulatable::el_size, operator[](), order, and sr.
| void CTF_int::tensor::leave_home_with_buffer | ( | ) |
| int CTF_int::tensor::map_tensor_rem | ( | int | num_phys_dims, |
| CommData * | phys_comm, | ||
| int | fill = 0 |
||
| ) |
map the remainder of a tensor
| [in] | num_phys_dims | number of physical processor grid dimensions |
| [in] | phys_comm | dimensional communicators |
| [in] | fill | whether to map everything |
Definition at line 2405 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), CTF_int::cdealloc(), CTF_int::mapping::cdt, CTF_int::mapping::child, CTF_int::mapping::has_child, CTF_int::map_tensor(), CTF_int::NOT_MAPPED, CTF_int::PHYSICAL_MAP, and CTF_int::mapping::type.
Referenced by CTF_int::scaling::execute(), and CTF_int::get_len_ordering().
| template void CTF_int::tensor::not_equals< bool > | ( | tensor * | A, |
| tensor * | B | ||
| ) |
do an elementwise comparison(!=) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor)
| [in] | A | first operand |
| [in] | B | second operand |
Definition at line 39 of file untyped_tensor_tmpl.h.
References ctf.core::a, ctf.core::b, ctf.core::dtype, CTF_int::accumulatable::el_size, operator[](), order, and sr.
| Idx_Tensor CTF_int::tensor::operator[] | ( | char const * | idx_map | ) |
associated an index map with the tensor for future operation
| [in] | idx_map | index assignment for this tensor |
Definition at line 31 of file untyped_tensor.cxx.
Referenced by compare_elementwise(), exp_helper(), larger_equal_than(), larger_than(), not_equals(), smaller_equal_than(), smaller_than(), and true_divide().
| void CTF_int::tensor::orient_subworld | ( | CTF::World * | greater_world, |
| int & | bw_mirror_rank, | ||
| int & | fw_mirror_rank, | ||
| distribution *& | odst, | ||
| char ** | sub_buffer_ | ||
| ) |
maps data from this world (subcomm) to the correct order of processors with respect to a parent (greater_world) comm
| [in] | greater_world | comm with respect to which the data needs to be ordered |
| [out] | bw_mirror_rank | processor rank in greater_world from which data is received |
| [out] | fw_mirror_rank | processor rank in greater_world to which data is sent |
| [out] | odst | distribution mapping of data on output defined on oriented subworld |
| [out] | sub_buffer_ | allocated buffer of received data on oriented subworld |
Definition at line 777 of file untyped_tensor.cxx.
References CTF_int::alloc(), CTF_int::CommData::allred(), ASSERT, CTF_int::CommData::bcast(), CTF_int::cdealloc(), CTF::World::cdt, CTF::World::comm, CTF_int::get_distribution_size(), CTF::World::rank, CTF_int::distribution::serialize(), and CTF_int::distribution::size.
Referenced by add_from_subworld(), and add_to_subworld().
| int CTF_int::tensor::permute | ( | tensor * | A, |
| int *const * | permutation_A, | ||
| char const * | alpha, | ||
| int *const * | permutation_B, | ||
| char const * | beta | ||
| ) |
Permutes a tensor along each dimension skips if perm set to -1, generalizes slice. one of permutation_A or permutation_B has to be set to NULL, if multiworld read, then the parent world tensor should not be being permuted
| [in] | A | pure-operand tensor |
| [in] | permutation_A | mappings for each dimension of A indices |
| [in] | alpha | scaling factor for A |
| [in] | permutation_B | mappings for each dimension of B (this) indices |
| [in] | beta | scaling factor for current values of B |
Definition at line 720 of file untyped_tensor.cxx.
References ASSERT, CTF_int::depermute_keys(), has_zero_edge_len, is_sparse, lens, CTF::World::np, order, CTF_int::algstrct::pair_dealloc(), CTF_int::permute_keys(), read_local(), sr, write(), and wrld.
Referenced by ctf.core.tensor::__setitem__(), and CTF::Tensor< dtype >::permute().
| void CTF_int::tensor::print | ( | FILE * | fp = stdout, |
| char const * | cutoff = NULL |
||
| ) | const |
prints tensor data to file using process 0
| [in] | fp | file to print to e.g. stdout |
| [in] | cutoff | do not print values of absolute value smaller than this |
Definition at line 1825 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), CTF_int::cdealloc(), read_local_nnz(), and sparsify().
Referenced by CTF::Tensor< dtype >::print(), and CTF::Tensor< dtype >::prnt().
| void CTF_int::tensor::print_map | ( | FILE * | stream = stdout, |
| bool | allcall = 1 |
||
| ) | const |
displays mapping information
| [in] | stream | output log (e.g. stdout) |
| [in] | allcall | (if 1 print only with proc 0) |
Definition at line 652 of file untyped_tensor.cxx.
References ctf.core::dim, ctf.core::np(), and CTF_int::PHYSICAL_MAP.
Referenced by compare(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), read(), and CTF_int::summation::sum_tensors().
| void CTF_int::tensor::prnt | ( | ) | const |
Definition at line 1822 of file untyped_tensor.cxx.
| void CTF_int::tensor::profile_off | ( | ) |
turn off profiling
Definition at line 711 of file untyped_tensor.cxx.
Referenced by CTF::Tensor< dtype >::profile_off().
| void CTF_int::tensor::profile_on | ( | ) |
turn on profiling
Definition at line 707 of file untyped_tensor.cxx.
Referenced by CTF::Tensor< dtype >::profile_on().
| void CTF_int::tensor::pull_alias | ( | tensor const * | other | ) |
pulls data from an tensor with an aliased buffer
| [in] | other | tensor with aliased data to pull from |
Definition at line 2176 of file untyped_tensor.cxx.
References ASSERT, CTF_int::copy_mapping(), data, edge_map, has_home, home_buffer, is_data_aliased, is_home, order, and topo.
Referenced by CTF_int::summation::estimate_time(), and CTF_int::get_len_ordering().
| void CTF_int::tensor::read | ( | int64_t | num_pair, |
| char const * | alpha, | ||
| char const * | beta, | ||
| int64_t const * | inds, | ||
| char * | data | ||
| ) |
read tensor data with <key, value> pairs where key is the global index for the value, which gets filled in with beta times the old values plus alpha times the values read from the tensor.
| [in] | num_pair | number of pairs to read |
| [in] | alpha | scaling factor of read value |
| [in] | beta | scaling factor of old value |
| [in] | indices | 64-bit global indices |
| [in] | data | values (num_pair of them to read) |
Definition at line 1203 of file untyped_tensor.cxx.
References CTF_int::PairIterator::read_val(), CTF_int::PairIterator::write_key(), and CTF_int::PairIterator::write_val().
Referenced by ctf.core.tensor::__getitem__(), CTF::Tensor< dtype >::get_mapped_data(), CTF::Tensor< dtype >::read(), and read().
| int CTF_int::tensor::read | ( | int64_t | num_pair, |
| char const * | alpha, | ||
| char const * | beta, | ||
| char * | mapped_data | ||
| ) |
read tensor data with <key, value> pairs where key is the global index for the value, which gets filled in with beta times the old values plus alpha times the values read from the tensor.
| [in] | num_pair | number of pairs to read |
| [in] | alpha | scaling factor of read value |
| [in] | beta | scaling factor of old value |
| [in] | mapped_data | pairs to write |
Definition at line 1221 of file untyped_tensor.cxx.
Referenced by ctf.core.tensor::__getitem__().
| char * CTF_int::tensor::read | ( | char const * | idx, |
| CTF::Idx_Partition const & | prl, | ||
| CTF::Idx_Partition const & | blk, | ||
| bool | unpack | ||
| ) |
returns local data of tensor with parallel distribution prl and local blocking blk
| [in] | idx | assignment of characters to each dim |
| [in] | prl | mesh processor topology with character labels |
| [in] | blk | local blocking with processor labels |
| [in] | unpack | whether to unpack from symmetric layout |
Definition at line 1299 of file untyped_tensor.cxx.
References clear_mapping(), data, has_home, is_data_aliased, is_home, NS, print_map(), read(), redistribute(), set_distribution(), and ctf.core::tsr.
Referenced by ctf.core.tensor::__getitem__().
| int CTF_int::tensor::read | ( | int64_t | num_pair, |
| char * | mapped_data | ||
| ) |
read tensor data with <key, value> pairs where key is the global index for the value, which gets filled in.
| [in] | num_pair | number of pairs to read |
| [in,out] | mapped_data | pairs to read |
Definition at line 1228 of file untyped_tensor.cxx.
Referenced by ctf.core.tensor::__getitem__().
|
protected |
read all pairs with each processor (packed)
| [out] | num_pair | number of values read |
| [in] | unpack | whether to read all or unique pairs return pair iterator with allocated all pairs read |
Definition at line 1648 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), CTF_int::cdealloc(), and CTF_int::PairIterator::sort().
| void CTF_int::tensor::read_dense_from_file | ( | MPI_File & | file, |
| int64_t | offset = 0 |
||
| ) |
read all tensor data from binary file in element order, which should be stored as nonsymmetric and dense as done in write_dense_to_file()
| [in] | file | stream to read from, the user should open, (optionally) set view, and close after function |
| [in] | offset | displacement in bytes at which to start in the file (ought ot be the same on all processors) |
Definition at line 2941 of file untyped_tensor.cxx.
References CTF_int::alloc(), CTF_int::cdealloc(), NS, CTF_int::packed_size(), read_dense_from_file(), and CTF_int::summation::sum_tensors().
Referenced by checkpoint(), and read_dense_from_file().
| void CTF_int::tensor::read_dense_from_file | ( | char const * | filename | ) |
read all tensor data from binary file in element order, which should be stored as nonsymmetric and dense as done in write_dense_to_file()
| [in] | filename | stream to read from |
Definition at line 2982 of file untyped_tensor.cxx.
| int CTF_int::tensor::read_local | ( | int64_t * | num_pair, |
| char ** | mapped_data, | ||
| bool | unpack_sym = false |
||
| ) | const |
read tensor data pairs local to processor including those with zero values WARNING: for sparse tensors this includes the zeros to maintain consistency with the behavior for dense tensors, use read_local_nnz to get only nonzeros
| [out] | num_pair | number of values read |
| [out] | mapped_data | values read |
Definition at line 1553 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), ASSERT, CTF_int::mapping::calc_phase(), CTF_int::mapping::calc_phys_phase(), CTF_int::mapping::calc_phys_rank(), CTF_int::cdealloc(), CTF_int::mapping::cdt, data, edge_map, CTF_int::summation::execute(), has_zero_edge_len, is_folded, is_mapped, is_sparse, CTF_int::topology::lda, ctf.core::np(), NS, order, pad_edge_len, padding, CTF_int::PHYSICAL_MAP, CTF_int::read_loc_pairs(), read_local(), size, CTF_int::SUCCESS, sym, TAU_FSTART, TAU_FSTOP, topo, ctf.core::tsr, CTF_int::mapping::type, and write().
Referenced by compare(), CTF::Tensor< dtype >::get_local_data(), CTF::Tensor< dtype >::get_local_pairs(), permute(), CTF::Tensor< dtype >::read_local(), read_local(), and slice().
| void CTF_int::tensor::read_local | ( | int64_t * | num_pair, |
| int64_t ** | inds, | ||
| char ** | data, | ||
| bool | unpack_sym = false |
||
| ) | const |
read tensor data pairs local to processor including those with zero values WARNING: for sparse tensors this includes the zeros to maintain consistency with the behavior for dense tensors, use read_local_nnz to get only nonzeros
| [out] | num_pair | number of values read |
| [out] | indices | 64-bit global indices |
| [out] | data | values (num_pair of them to read) |
Definition at line 1505 of file untyped_tensor.cxx.
References CTF_int::alloc(), CTF_int::ConstPairIterator::k(), and CTF_int::ConstPairIterator::read_val().
| int CTF_int::tensor::read_local_nnz | ( | int64_t * | num_pair, |
| char ** | mapped_data, | ||
| bool | unpack_sym = false |
||
| ) | const |
read tensor data pairs local to processor that have nonzero values
| [out] | num_pair | number of values read |
| [out] | mapped_data | values read |
Definition at line 1538 of file untyped_tensor.cxx.
References data, is_data_aliased, nnz_loc, sparsify(), and CTF_int::SUCCESS.
Referenced by extract_diag(), CTF::Tensor< dtype >::get_local_data(), CTF::Tensor< dtype >::get_local_pairs(), print(), and ctf.core.tensor::reshape().
| void CTF_int::tensor::read_local_nnz | ( | int64_t * | num_pair, |
| int64_t ** | inds, | ||
| char ** | data, | ||
| bool | unpack_sym = false |
||
| ) | const |
read tensor data pairs local to processor that have nonzero values
| [out] | num_pair | number of values read |
| [out] | indices | 64-bit global indices |
| [out] | data | values (num_pair of them to read) |
Definition at line 1522 of file untyped_tensor.cxx.
References CTF_int::alloc(), CTF_int::ConstPairIterator::k(), and CTF_int::ConstPairIterator::read_val().
Referenced by ctf.core.tensor::reshape().
| int CTF_int::tensor::redistribute | ( | distribution const & | old_dist, |
| int const * | old_offsets = NULL, |
||
| int *const * | old_permutation = NULL, |
||
| int const * | new_offsets = NULL, |
||
| int *const * | new_permutation = NULL |
||
| ) |
permutes the data of a tensor to its new layout
| [in] | old_dist | previous distribution to remap data from |
| [in] | old_offsets | offsets from corner of tensor |
| [in] | old_permutation | permutation of rows/cols/... |
| [in] | new_offsets | offsets from corner of tensor |
| [in] | new_permutation | permutation of rows/cols/... |
Definition at line 2201 of file untyped_tensor.cxx.
References ABORT, CTF_int::alloc(), CTF_int::block_reshuffle(), CTF_int::can_block_reshuffle(), CTF_int::cdealloc(), dgtog_reshuffle(), DPRINTF, CTF_int::padded_reshuffle(), CTF_int::distribution::phase, CTF_int::distribution::size, CTF_int::spredist_mdl, CTF::Timer::start(), CTF::Timer::stop(), CTF_int::SUCCESS, and VPRINTF.
Referenced by CTF_int::scaling::execute(), CTF_int::get_len_ordering(), and read().
| int CTF_int::tensor::reduce_sum | ( | char * | result | ) |
Performs an elementwise summation reduction on a tensor.
| [out] | result | result of reduction operation |
Definition at line 1771 of file untyped_tensor.cxx.
Referenced by CTF::Tensor< dtype >::reduce().
| int CTF_int::tensor::reduce_sum | ( | char * | result, |
| algstrct const * | sr_other | ||
| ) |
Performs an elementwise summation reduction on a tensor with summation defined by sr_other.
| [out] | result | result of reduction operation |
| [in] | sr_other | an algebraic structure (at least a monoid) defining the summation operation |
Definition at line 1775 of file untyped_tensor.cxx.
References ASSERT, data, CTF_int::accumulatable::el_size, CTF_int::summation::execute(), CTF_int::algstrct::mulid(), and CTF_int::SUCCESS.
| int CTF_int::tensor::reduce_sumabs | ( | char * | result | ) |
Performs an elementwise absolute value summation reduction on a tensor.
| [out] | result | result of reduction operation |
Definition at line 1789 of file untyped_tensor.cxx.
Referenced by CTF::Tensor< dtype >::reduce().
| int CTF_int::tensor::reduce_sumabs | ( | char * | result, |
| algstrct const * | sr_other | ||
| ) |
Performs an elementwise absolute value summation reduction on a tensor.
| [out] | result | result of reduction operation |
| [in] | sr_other | an algebraic structure (at least a monoid) defining the summation operation |
Definition at line 1793 of file untyped_tensor.cxx.
References CTF_int::algstrct::abs, ASSERT, data, CTF_int::summation::execute(), CTF_int::algstrct::mulid(), CTF_int::SUCCESS, and univar_function().
| int CTF_int::tensor::reduce_sumsq | ( | char * | result | ) |
computes the sum of squares of the elements
| [out] | result | result of reduction operation |
Definition at line 1808 of file untyped_tensor.cxx.
References ASSERT, data, CTF_int::contraction::execute(), and CTF_int::SUCCESS.
Referenced by CTF::Tensor< dtype >::reduce().
| void CTF_int::tensor::register_size | ( | int64_t | size | ) |
register buffer allocation for this tensor
Definition at line 2883 of file untyped_tensor.cxx.
References CTF_int::inc_tot_mem_used().
| void CTF_int::tensor::remove_fold | ( | ) |
removes folding without doing transpose unsets is_folded and deletes rec_tsr
Definition at line 2067 of file untyped_tensor.cxx.
References CTF_int::cdealloc().
Referenced by CTF_int::get_len_ordering().
| void CTF_int::tensor::scale_diagonals | ( | int const * | sym_mask | ) |
scales each element by 1/(number of entries equivalent to it after permutation of indices for which sym_mask is 1)
| [in] | sym_mask | identifies which tensor indices are part of the symmetric group which diagonals we want to scale (i.e. sym_mask [1,1] does A["ii"]= (1./2.)*A["ii"]) |
Definition at line 2629 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), ASSERT, CTF_int::mapping::calc_phase(), CTF_int::mapping::calc_phys_phase(), CTF_int::mapping::calc_phys_rank(), CTF_int::cdealloc(), CTF_int::mapping::cdt, ctf.core::np(), CTF_int::PHYSICAL_MAP, CTF_int::scal_diag(), TAU_FSTART, TAU_FSTOP, and CTF_int::mapping::type.
Referenced by CTF_int::get_len_ordering().
| tensor * CTF_int::tensor::self_reduce | ( | int const * | idx_A, |
| int ** | new_idx_A, | ||
| int | order_B, | ||
| int const * | idx_B, | ||
| int ** | new_idx_B, | ||
| int | order_C = 0, |
||
| int const * | idx_C = NULL, |
||
| int ** | new_idx_C = NULL |
||
| ) |
performs a partial reduction on the tensor (used in summation and contraction)
| [in] | idx_A | index map of this tensor as defined by summation/contraction |
| [out] | new_idx_A | how idx_A needs to be transformed |
| [in] | order_B | number of modes in tensor B |
| [in] | idx_B | index map containing all indices in another tensor involved in the operaiton |
| [out] | new_idx_B | how idx_B needs to be transformed |
| [in] | order_C | number of modes in tensor C |
| [in] | idx_C | index map containing all indices in another tensor involved in the operaiton (should be NULL for summation) |
| [out] | new_idx_C | how idx_C needs to be transformed, untouched if idx_C is NULL |
| [in] | idx_C | index map containing all indices in another tensor involved in the operaiton (should be NULL for summation) |
| [out] | new_idx_C | how idx_C needs to be transformed, untouched if idx_C is NULL |
Definition at line 2990 of file untyped_tensor.cxx.
References CTF_int::alloc(), CTF_int::summation::execute(), and NS.
Referenced by CTF_int::summation::estimate_time(), and CTF_int::get_len_ordering().
| int CTF_int::tensor::set | ( | char const * | val | ) |
sets tensor data to val
Definition at line 493 of file untyped_tensor.cxx.
|
protected |
set edge mappings as specified
| [in] | idx | assignment of characters to each dim |
| [in] | prl | mesh processor topology with character labels |
| [in] | blk | local blocking with processor labels |
Definition at line 1233 of file untyped_tensor.cxx.
References ASSERT, CTF_int::cdealloc(), CTF_int::mapping::cdt, CTF_int::check_self_mapping(), CTF_int::mapping::child, CTF_int::conv_idx(), CTF_int::find_topology(), CTF_int::mapping::has_child, CTF::Idx_Partition::idx, CTF::Partition::lens, CTF_int::NOT_MAPPED, CTF_int::mapping::np, CTF::Partition::order, CTF::Idx_Partition::part, CTF_int::PHYSICAL_MAP, CTF::Idx_Partition::reduce_order(), CTF_int::mapping::type, and CTF_int::VIRTUAL_MAP.
Referenced by read().
| void CTF_int::tensor::set_name | ( | char const * | name | ) |
set the tensor name
| [in] | name | to set |
Definition at line 697 of file untyped_tensor.cxx.
References CTF_int::alloc(), and CTF_int::cdealloc().
Referenced by CTF::Tensor< dtype >::set_name().
| void CTF_int::tensor::set_new_nnz_glb | ( | int64_t const * | nnz_blk | ) |
sets the number of nonzeros both locally (nnz_loc) and overall globally (nnz_tot)
| [in] | nnz_blk | number of nonzeros in each block |
Definition at line 2716 of file untyped_tensor.cxx.
Referenced by CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), CTF_int::summation::sum_tensors(), and CTF_int::symmetrize().
| void CTF_int::tensor::set_padding | ( | ) |
sets padding and local size of a tensor given a mapping
Definition at line 456 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), CTF_int::mapping::calc_phase(), CTF_int::cdealloc(), and CTF_int::sy_packed_size().
Referenced by CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), CTF_int::symmetrize(), and write().
| void CTF_int::tensor::set_sym | ( | int const * | sym | ) |
sets symmetry, WARNING: for internal use only !!!!
| [in] | sym |
Definition at line 2701 of file untyped_tensor.cxx.
References NS.
Referenced by CTF_int::get_len_ordering(), and CTF_int::summation::sum_tensors().
| int CTF_int::tensor::set_zero | ( | ) |
elects a mapping and sets tensor data to zero
btopo = get_best_topo(bnvirt, btopo, wrld->cdt, 0, bmemuse);
Definition at line 499 of file untyped_tensor.cxx.
References CTF_int::alloc(), CTF_int::alloc_ptr(), ASSERT, CTF_int::cdealloc(), DPRINTF, CTF_int::ERROR, CTF_int::get_best_topo(), CTF_int::lcm(), CTF_int::map_tensor(), ctf.core::np(), CTF_int::proc_bytes_available(), CTF_int::SUCCESS, TAU_FSTART, and TAU_FSTOP.
Referenced by ctf.core.tensor::__idiv__(), ctf.core.tensor::__imul__(), ctf.core.tensor::__itruediv__(), ctf.core.tensor::__setitem__(), btwn_cnt_fast(), CTF_int::desymmetrize(), CTF_int::get_len_ordering(), and CTF::Idx_Tensor::operator=().
| void CTF_int::tensor::slice | ( | int const * | offsets_B, |
| int const * | ends_B, | ||
| char const * | beta, | ||
| tensor * | A, | ||
| int const * | offsets_A, | ||
| int const * | ends_A, | ||
| char const * | alpha | ||
| ) |
accumulates out a slice (block) of this tensor = B B[offsets,ends)=beta*B[offsets,ends) + alpha*A[offsets_A,ends_A)
| [in] | offsets_B | bottom left corner of block |
| [in] | ends_B | top right corner of block |
| [in] | beta | scaling factor of this tensor |
| [in] | A | tensor who owns pure-operand slice |
| [in] | offsets_A | bottom left corner of block of A |
| [in] | ends_A | top right corner of block of A |
| [in] | alpha | scaling factor of tensor A |
Definition at line 861 of file untyped_tensor.cxx.
References CTF_int::alloc(), ASSERT, CTF_int::cdealloc(), CTF_int::depad_tsr(), has_zero_edge_len, lens, CTF::World::np, NS, order, CTF_int::pad_key(), CTF_int::algstrct::pair_alloc(), CTF_int::algstrct::pair_dealloc(), read_local(), sr, sym, write(), and wrld.
Referenced by add_to_subworld(), and CTF::Tensor< dtype >::slice().
| template void CTF_int::tensor::smaller_equal_than< bool > | ( | tensor * | A, |
| tensor * | B | ||
| ) |
do an elementwise comparison(<=) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor)
| [in] | A | first operand |
| [in] | B | second operand |
Definition at line 63 of file untyped_tensor_tmpl.h.
References ctf.core::a, ctf.core::b, ctf.core::dtype, CTF_int::accumulatable::el_size, operator[](), order, and sr.
| template void CTF_int::tensor::smaller_than< bool > | ( | tensor * | A, |
| tensor * | B | ||
| ) |
do an elementwise comparison(<) of two tensors with elements of type dtype (primarily needed for python interface), store result in this tensor (has to be boolean tensor)
| [in] | A | first operand |
| [in] | B | second operand |
Definition at line 51 of file untyped_tensor_tmpl.h.
References ctf.core::a, ctf.core::b, ctf.core::dtype, CTF_int::accumulatable::el_size, operator[](), order, and sr.
| int CTF_int::tensor::sparsify | ( | char const * | threshold = NULL, |
| bool | take_abs = true |
||
| ) |
reduce tensor to sparse format, storing only nonzero data, or data above a specified threshold. makes dense tensors sparse. cleans sparse tensors of any 'computed' zeros.
| [in] | threshold | all values smaller or equal to than this one will be removed/not stored (by default is NULL, meaning only zeros are removed, so same as threshold=additive identity) |
| [in] | take_abs | whether to take absolute value when comparing to threshold |
Definition at line 1332 of file untyped_tensor.cxx.
References CTF_int::SUCCESS.
Referenced by CTF_int::get_len_ordering(), print(), read_local_nnz(), CTF::Tensor< dtype >::sparsify(), CTF_int::subsample(), and CTF_int::summation::sum_tensors().
| int CTF_int::tensor::sparsify | ( | std::function< bool(char const *)> | f | ) |
sparsifies tensor keeping only values v such that filter(v) = true
| [in] | f | boolean function to apply to values to determine whether to keep them, must be deterministic |
Definition at line 1355 of file untyped_tensor.cxx.
References CTF_int::alloc(), CTF_int::alloc_ptr(), ASSERT, CTF_int::mapping::calc_phase(), CTF_int::mapping::calc_phys_phase(), CTF_int::mapping::calc_phys_rank(), CTF_int::cdealloc(), CTF_int::mapping::cdt, CTF_int::depad_tsr(), CTF_int::pad_key(), CTF_int::PHYSICAL_MAP, CTF_int::spsfy_tsr(), CTF_int::SUCCESS, TAU_FSTART, TAU_FSTOP, CTF_int::mapping::type, and CTF_int::PairIterator::write().
| void CTF_int::tensor::spmatricize | ( | int | m, |
| int | n, | ||
| int | nrow_idx, | ||
| bool | csr | ||
| ) |
transposes local data in preparation for summation or contraction, transforms to COO or CSR format for sparse
| [in] | m | number of rows in matrix |
| [in] | n | number of columns in matrix |
| [in] | nrow_idx | number of indices to fold into column |
| [in] | csr | whether to do csr (1) or coo (0) layout |
Definition at line 2728 of file untyped_tensor.cxx.
References CTF_int::COO_Matrix::all_data, CTF_int::alloc(), CTF_int::alloc_ptr(), ASSERT, CTF_int::cdealloc(), CTF_int::get_coo_size(), CTF_int::get_csr_size(), CTF_int::COO_Matrix::set_data(), TAU_FSTART, and TAU_FSTOP.
| template void CTF_int::tensor::true_divide< bool > | ( | tensor * | A | ) |
Definition at line 99 of file untyped_tensor_tmpl.h.
References ctf.core::a, ctf.core::dtype, operator[](), and order.
| void CTF_int::tensor::unfold | ( | bool | was_mod = 0 | ) |
undo the folding of a local tensor block unsets is_folded and deletes rec_tsr
| [in] | was_mod | true if data was modified, controls whether to discard sparse data |
Definition at line 2029 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), ASSERT, CTF_int::calc_dim(), CTF_int::cdealloc(), CTF_int::nosym_transpose(), NS, and CTF_int::sy_packed_size().
Referenced by CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), and CTF_int::summation::sum_tensors().
| int CTF_int::tensor::write | ( | int64_t | num_pair, |
| char const * | alpha, | ||
| char const * | beta, | ||
| char * | mapped_data, | ||
| char const | rw = 'w' |
||
| ) |
Add tensor data new=alpha*new+beta*old with <key, value> pairs where key is the global index for the value.
| [in] | num_pair | number of pairs to write |
| [in] | alpha | scaling factor of written (read) value |
| [in] | beta | scaling factor of old (existing) value |
| [in] | mapped_data | pairs to write |
| [in] | rw | weather to read (r) or write (w) |
Definition at line 1082 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), ASSERT, CTF_int::mapping::calc_phase(), CTF_int::mapping::calc_phys_phase(), CTF_int::mapping::calc_phys_rank(), CTF_int::cdealloc(), CTF_int::mapping::cdt, data, DEBUG_PRINTF, edge_map, CTF_int::ERROR, has_zero_edge_len, is_mapped, CTF_int::topology::lda, order, pad_edge_len, padding, CTF_int::PHYSICAL_MAP, CTF_int::proc_bytes_available(), set_padding(), CTF_int::SUCCESS, sym, TAU_FSTART, TAU_FSTOP, topo, ctf.core::tsr, CTF_int::mapping::type, and CTF_int::wr_pairs_layout().
Referenced by ctf.core.tensor::__setitem__(), extract_diag(), ctf.core.tensor::from_nparray(), permute(), read_local(), slice(), and CTF::Tensor< dtype >::write().
| void CTF_int::tensor::write | ( | int64_t | num_pair, |
| char const * | alpha, | ||
| char const * | beta, | ||
| int64_t const * | inds, | ||
| char const * | data | ||
| ) |
Add tensor data new=alpha*new+beta*old with <key, value> pairs where key is the global index for the value.
| [in] | num_pair | number of pairs to write |
| [in] | alpha | scaling factor of written (read) value |
| [in] | beta | scaling factor of old (existing) value |
| [in] | indices | 64-bit global indices |
| [in] | data | values (num_pair of them) |
Definition at line 1067 of file untyped_tensor.cxx.
References CTF_int::PairIterator::write_key(), and CTF_int::PairIterator::write_val().
Referenced by ctf.core.tensor::__setitem__(), and ctf.core.tensor::from_nparray().
| void CTF_int::tensor::write_dense_to_file | ( | MPI_File & | file, |
| int64_t | offset = 0 |
||
| ) |
write all tensor data to binary file in element order, unpacking from sparse or symmetric formats
| [in,out] | file | stream to write to, the user should open, (optionally) set view, and close after function |
| [in] | offset | displacement in bytes at which to start in the file (ought ot be the same on all processors) |
Definition at line 2894 of file untyped_tensor.cxx.
References CTF_int::alloc(), CTF_int::cdealloc(), NS, CTF_int::packed_size(), CTF_int::PairIterator::read_val(), write_dense_to_file(), and CTF_int::PairIterator::write_key().
Referenced by checkpoint(), and write_dense_to_file().
| void CTF_int::tensor::write_dense_to_file | ( | char const * | filename | ) |
write all tensor data to binary file in element order, unpacking from sparse or symmetric formats
| [in] | filename | stream to write to |
Definition at line 2933 of file untyped_tensor.cxx.
| int CTF_int::tensor::zero_out_padding | ( | ) |
sets padded portion of tensor to zero (this should be maintained internally)
Definition at line 2570 of file untyped_tensor.cxx.
References CTF_int::alloc_ptr(), CTF_int::mapping::calc_phase(), CTF_int::mapping::calc_phys_phase(), CTF_int::mapping::calc_phys_rank(), CTF_int::cdealloc(), CTF_int::mapping::cdt, ctf.core::np(), CTF_int::PHYSICAL_MAP, CTF_int::SUCCESS, TAU_FSTART, TAU_FSTOP, CTF_int::mapping::type, and CTF_int::zero_padding().
Referenced by CTF_int::desymmetrize(), CTF::fill_random_base(), CTF_int::get_len_ordering(), CTF_int::summation::sum_tensors(), and CTF_int::symmetrize().
| char* CTF_int::tensor::data |
tensor data, either the data or the key-value pairs should exist at any given time
Definition at line 117 of file untyped_tensor.h.
Referenced by add_to_subworld(), copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF::fill_random_base(), CTF_int::get_len_ordering(), CTF::Tensor< dtype >::get_raw_data(), CTF_int::nosym_transpose(), CTF_int::Term::operator double(), CTF_int::Term::operator float(), CTF_int::Term::operator int(), CTF_int::Term::operator int64_t(), CTF::Matrix< dtype >::print_matrix(), pull_alias(), read(), read_local(), read_local_nnz(), CTF::Matrix< dtype >::read_mat(), reduce_sum(), reduce_sumabs(), reduce_sumsq(), CTF_int::summation::sum_tensors(), CTF_int::symmetrize(), CTF_int::tsum::tsum(), write(), and CTF::Matrix< dtype >::write_mat().
| mapping* CTF_int::tensor::edge_map |
mappings of each tensor dimension onto topology dimensions
Definition at line 96 of file untyped_tensor.h.
Referenced by align(), CTF_int::check_self_mapping(), copy_tensor_data(), CTF_int::ctr_2d_gen_build(), CTF_int::desymmetrize(), CTF_int::distribution::distribution(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF::Matrix< dtype >::get_desc(), CTF_int::get_len_ordering(), CTF_int::map_self_indices(), pull_alias(), read_local(), CTF::Matrix< dtype >::read_mat(), CTF_int::seq_tsr_spsum::seq_tsr_spsum(), CTF_int::spctr_pin_keys::spctr_pin_keys(), CTF::Matrix< dtype >::svd(), CTF_int::symmetrize(), CTF_int::tspsum_map::tspsum_map(), CTF_int::tspsum_pin_keys::tspsum_pin_keys(), write(), and CTF::Matrix< dtype >::write_mat().
| bool CTF_int::tensor::has_home |
whether the tensor has a home mapping/buffer
Definition at line 119 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), CTF_int::nosym_transpose(), pull_alias(), and read().
| bool CTF_int::tensor::has_zero_edge_len |
if true tensor has a zero edge length, so is zero, which short-cuts stuff
Definition at line 115 of file untyped_tensor.h.
Referenced by CTF_int::scaling::execute(), CTF_int::get_len_ordering(), permute(), read_local(), slice(), tensor(), and write().
| char* CTF_int::tensor::home_buffer |
buffer associated with home mapping of tensor, to which it is returned
Definition at line 121 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), CTF_int::nosym_transpose(), and pull_alias().
| int64_t CTF_int::tensor::home_size |
size of home buffer
Definition at line 123 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), and CTF_int::get_len_ordering().
| int* CTF_int::tensor::inner_ordering |
ordering of the dimensions according to which the tensori s folded
Definition at line 104 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::get_len_ordering(), CTF_int::seq_tsr_ctr::seq_tsr_ctr(), and CTF_int::seq_tsr_spctr::seq_tsr_spctr().
| bool CTF_int::tensor::is_csr |
whether CSR or COO if folded
Definition at line 133 of file untyped_tensor.h.
| bool CTF_int::tensor::is_cyclic |
whether the tensor data is cyclically distributed (blocked if false)
Definition at line 108 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::distribution::distribution(), CTF_int::scaling::execute(), and CTF_int::get_len_ordering().
| bool CTF_int::tensor::is_data_aliased |
whether the tensor data is an alias of another tensor object's data
Definition at line 110 of file untyped_tensor.h.
Referenced by CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), pull_alias(), read(), and read_local_nnz().
| bool CTF_int::tensor::is_folded |
whether the data is folded/transposed into a (lower-order) tensor
Definition at line 102 of file untyped_tensor.h.
Referenced by align(), copy_tensor_data(), and read_local().
| bool CTF_int::tensor::is_home |
whether the latest tensor data is in the home buffer
Definition at line 125 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), CTF_int::nosym_transpose(), pull_alias(), read(), and CTF_int::symmetrize().
| bool CTF_int::tensor::is_mapped |
whether a mapping has been selected
Definition at line 92 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), read_local(), CTF_int::symmetrize(), and write().
| int CTF_int::tensor::is_scp_padded |
whether tensor data has additional padding
Definition at line 86 of file untyped_tensor.h.
| bool CTF_int::tensor::is_sparse |
whether only the non-zero elements of the tensor are stored
Definition at line 131 of file untyped_tensor.h.
Referenced by block_sparse(), copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::scaling::execute(), CTF::fill_random_base(), CTF_int::get_full_intm(), CTF_int::get_len_ordering(), CTF::Scalar< dtype >::operator=(), CTF::Tensor< dtype >::operator=(), permute(), read_local(), CTF::Matrix< dtype >::read_mat(), CTF_int::seq_tsr_spsum::seq_tsr_spsum(), setup(), CTF_int::spctr::spctr(), CTF_int::summation::sum_tensors(), CTF_int::symmetrize(), tensor(), CTF_int::tspsum::tspsum(), and CTF_int::tspsum_map::tspsum_map().
| bool CTF_int::tensor::left_home_transp |
whether the tensor left home to transpose
Definition at line 127 of file untyped_tensor.h.
Referenced by CTF_int::nosym_transpose().
| int* CTF_int::tensor::lens |
unpadded tensor edge lengths
Definition at line 78 of file untyped_tensor.h.
Referenced by ao_mo_transf_slice(), block_sparse(), ccsd(), check_asym(), check_sym(), compare(), copy_tensor_data(), divide_EaEi(), CTF::fill_sp_random_base(), CTF::get_core_tensor(), get_core_tensor(), get_core_tensor_hooi(), get_factor_matrices(), CTF::get_factor_matrices(), CTF_int::get_full_intm(), CTF_int::get_len_ordering(), get_rand_as_tsr(), CTF::Matrix< dtype >::Matrix(), mp3(), CTF::Scalar< dtype >::operator=(), CTF::Tensor< dtype >::operator=(), permute(), CTF::read_sparse_from_file_base(), rec_scan(), setup(), slice(), CTF::Tensor< dtype >::slice(), CTF_int::spctr_pin_keys::spctr_pin_keys(), tensor(), CTF_int::tspsum_map::tspsum_map(), CTF_int::tspsum_pin_keys::tspsum_pin_keys(), vcycle(), CTF::Vector< dtype >::Vector(), and CTF::write_sparse_to_file_base().
| char* CTF_int::tensor::name |
name given to tensor
Definition at line 84 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF::Tensor< dtype >::fill_random(), CTF::Tensor< dtype >::fill_sp_random(), CTF_int::get_len_ordering(), CTF::Tensor< dtype >::norm1(), CTF::Tensor< dtype >::norm2(), CTF::Tensor< dtype >::norm_infty(), CTF_int::tensor_name_less::operator()(), CTF::Scalar< dtype >::operator=(), CTF::Tensor< dtype >::operator=(), CTF_int::symmetrize(), and tensor().
| int64_t* CTF_int::tensor::nnz_blk |
nonzero elements in each block owned locally
Definition at line 141 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), CTF_int::summation::sum_tensors(), CTF_int::symmetrize(), and CTF_int::tspsum::tspsum().
| int64_t CTF_int::tensor::nnz_loc |
number of local nonzero elements
Definition at line 137 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::summation::estimate_time(), read_local_nnz(), CTF_int::summation::sum_tensors(), and CTF_int::tspsum::tspsum().
| int64_t CTF_int::tensor::nnz_tot |
maximum number of local nonzero elements over all procs
Definition at line 139 of file untyped_tensor.h.
Referenced by btwn_cnt_fast(), copy_tensor_data(), CTF_int::summation::estimate_time(), CTF_int::get_len_ordering(), mis(), mis2(), setup_3d_Poisson(), setup_unstructured(), test_mis(), and test_mis2().
| int CTF_int::tensor::nrow_idx |
how many modes are folded into matricized row
Definition at line 135 of file untyped_tensor.h.
| int CTF_int::tensor::order |
number of tensor dimensions
Definition at line 76 of file untyped_tensor.h.
Referenced by ctf.core.tensor::__get__(), ctf.core.tensor::__getitem__(), ctf.core.tensor::__setitem__(), CTF_int::abs_helper(), add_from_subworld(), CTF_int::add_sym_perm(), add_to_subworld(), align(), block_sparse(), CTF_int::calc_fold_lnmk(), check_asym(), CTF_int::check_self_mapping(), check_sym(), compare(), compare_elementwise(), CTF_int::conj_helper(), CTF_int::contraction::contraction(), conv_type(), copy_tensor_data(), CTF_int::ctr_2d_gen_build(), CTF_int::ctr_virt::ctr_virt(), CTF_int::desymmetrize(), CTF_int::distribution::distribution(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), exp_helper(), fast_tensor_ctr(), CTF::fill_sp_random_base(), CTF::get_core_tensor(), get_core_tensor(), get_core_tensor_hooi(), get_factor_matrices(), CTF::get_factor_matrices(), CTF_int::get_full_intm(), CTF_int::get_imag(), CTF_int::get_len_ordering(), get_rand_as_tsr(), CTF_int::get_real(), CTF_int::get_sym_perms(), CTF::Idx_Tensor::get_uniq_inds(), hooi(), ctf.core.tensor::i(), CTF::Idx_Tensor::Idx_Tensor(), larger_equal_than(), larger_than(), CTF_int::map_self_indices(), CTF::Matrix< dtype >::Matrix(), not_equals(), CTF::Scalar< dtype >::operator=(), CTF::Tensor< dtype >::operator=(), CTF_int::order_perm(), permute(), ctf.core.tensor::permute(), pull_alias(), read_local(), CTF::read_sparse_from_file_base(), CTF::real_norm1(), rec_scan(), CTF_int::seq_tsr_ctr::seq_tsr_ctr(), CTF_int::seq_tsr_spctr::seq_tsr_spctr(), CTF_int::seq_tsr_spsum::seq_tsr_spsum(), CTF_int::seq_tsr_sum::seq_tsr_sum(), CTF_int::set_imag(), CTF_int::set_real(), slice(), CTF::Tensor< dtype >::slice(), smaller_equal_than(), smaller_than(), CTF_int::spctr_pin_keys::spctr_pin_keys(), CTF_int::spctr_virt::spctr_virt(), CTF_int::sum_bool_tsr(), CTF_int::summation::sum_tensors(), CTF_int::summation::summation(), CTF_int::symmetrize(), tensor(), ctf.core.tensor::transpose(), true_divide(), CTF_int::tspsum_map::tspsum_map(), CTF_int::tspsum_permute::tspsum_permute(), CTF_int::tspsum_pin_keys::tspsum_pin_keys(), CTF_int::tspsum_virt::tspsum_virt(), CTF_int::tsum_virt::tsum_virt(), CTF::Vector< dtype >::Vector(), write(), and CTF::write_sparse_to_file_base().
| int* CTF_int::tensor::pad_edge_len |
padded tensor edge lengths
Definition at line 80 of file untyped_tensor.h.
Referenced by CTF_int::calc_fold_lnmk(), copy_tensor_data(), CTF_int::distribution::distribution(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF::Matrix< dtype >::get_desc(), CTF_int::get_len_ordering(), read_local(), CTF_int::seq_tsr_spsum::seq_tsr_spsum(), CTF_int::tspsum_map::tspsum_map(), and write().
| int* CTF_int::tensor::padding |
padding along each edge length (less than distribution phase)
Definition at line 82 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::distribution::distribution(), read_local(), and write().
| bool CTF_int::tensor::profile |
whether profiling should be done for contractions/sums involving this tensor
Definition at line 129 of file untyped_tensor.h.
Referenced by CTF_int::desymmetrize(), CTF_int::get_len_ordering(), CTF::Scalar< dtype >::operator=(), CTF::Tensor< dtype >::operator=(), CTF_int::symmetrize(), and tensor().
| tensor* CTF_int::tensor::rec_tsr |
representation of folded tensor (shares data pointer)
Definition at line 106 of file untyped_tensor.h.
Referenced by copy_tensor_data(), CTF_int::summation::estimate_time(), CTF_int::get_len_ordering(), CTF_int::seq_tsr_ctr::seq_tsr_ctr(), and CTF_int::seq_tsr_spctr::seq_tsr_spctr().
| int64_t CTF_int::tensor::registered_alloc_size |
size CTF keeps track of for memory usage
Definition at line 100 of file untyped_tensor.h.
| int* CTF_int::tensor::scp_padding |
additional padding, may be greater than ScaLAPACK phase
Definition at line 88 of file untyped_tensor.h.
| int64_t CTF_int::tensor::size |
current size of local tensor data chunk (mapping-dependent)
Definition at line 98 of file untyped_tensor.h.
Referenced by ctf.core.tensor::__float__(), ctf.core.tensor::__get__(), ctf.core.tensor::__int__(), ctf.core.tensor::__nonzero__(), copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::distribution::distribution(), CTF_int::scaling::execute(), CTF::fill_random_base(), flatten_block_sparse_matrix(), CTF_int::get_len_ordering(), CTF_int::nosym_transpose(), CTF::Matrix< dtype >::qr(), read_local(), CTF::Matrix< dtype >::read_mat(), CTF_int::spctr_pin_keys::spctr_pin_keys(), CTF::Matrix< dtype >::svd(), CTF_int::symmetrize(), and CTF::Matrix< dtype >::write_mat().
| tensor* CTF_int::tensor::slay |
tensor object associated with tensor object whose data pointer needs to be preserved, needed for ScaLAPACK wrapper FIXME: home buffer should presumably take care of this...
Definition at line 113 of file untyped_tensor.h.
| algstrct* CTF_int::tensor::sr |
algstrct on which tensor elements and operations are defined
Definition at line 72 of file untyped_tensor.h.
Referenced by CTF::Tensor< dtype >::add_from_subworld(), add_to_subworld(), CTF::Tensor< dtype >::add_to_subworld(), compare(), compare_elementwise(), copy_tensor_data(), CTF_int::ctr::ctr(), CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF::fill_sp_random_base(), CTF_int::get_full_intm(), CTF_int::get_len_ordering(), CTF::Tensor< dtype >::get_local_data(), larger_equal_than(), larger_than(), CTF_int::matrix_qr(), CTF_int::matrix_qr_cmplx(), CTF_int::matrix_svd(), CTF_int::matrix_svd_cmplx(), CTF_int::nosym_transpose(), not_equals(), CTF::Scalar< dtype >::operator=(), CTF::Tensor< dtype >::operator=(), CTF::Schedule::partition_and_execute(), permute(), CTF::Matrix< dtype >::print_matrix(), CTF::Matrix< dtype >::qr(), CTF::Tensor< dtype >::read(), CTF::Tensor< dtype >::read_local(), CTF::Matrix< dtype >::read_mat(), CTF::read_sparse_from_file_base(), rec_scan(), CTF::Tensor< dtype >::reduce(), scan(), slice(), CTF::Tensor< dtype >::slice(), smaller_equal_than(), smaller_than(), CTF::Sparse_Tensor< dtype >::Sparse_Tensor(), CTF_int::summation::sum_tensors(), CTF_int::symmetrize(), tensor(), CTF::Tensor< dtype >::Tensor(), CTF_int::tsum::tsum(), CTF::Tensor< dtype >::write(), and CTF::write_sparse_to_file_base().
| int* CTF_int::tensor::sym |
symmetries among tensor dimensions
Definition at line 74 of file untyped_tensor.h.
Referenced by ctf.core.tensor::__get__(), CTF_int::add_sym_perm(), block_sparse(), ccsd(), check_asym(), check_sym(), copy_tensor_data(), CTF_int::desymmetrize(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_full_intm(), CTF_int::get_len_ordering(), get_rand_as_tsr(), CTF_int::get_sym_perms(), CTF::Matrix< dtype >::Matrix(), CTF::Scalar< dtype >::operator=(), CTF::Tensor< dtype >::operator=(), CTF_int::order_perm(), read_local(), CTF_int::seq_tsr_ctr::seq_tsr_ctr(), CTF_int::seq_tsr_spctr::seq_tsr_spctr(), CTF_int::seq_tsr_spsum::seq_tsr_spsum(), CTF_int::seq_tsr_sum::seq_tsr_sum(), slice(), CTF::Tensor< dtype >::slice(), CTF_int::summation::sum_tensors(), CTF_int::symmetrize(), tensor(), ctf.core.tensor::transpose(), and write().
| int* CTF_int::tensor::sym_table |
order-by-order table of dimensional symmetry relations
Definition at line 90 of file untyped_tensor.h.
Referenced by CTF_int::get_len_ordering(), and CTF_int::map_self_indices().
| topology* CTF_int::tensor::topo |
topology to which the tensor is mapped
Definition at line 94 of file untyped_tensor.h.
Referenced by align(), copy_tensor_data(), CTF_int::ctr_2d_gen_build(), CTF_int::ctr_replicate::ctr_replicate(), CTF_int::desymmetrize(), CTF_int::distribution::distribution(), CTF_int::summation::estimate_time(), CTF_int::scaling::execute(), CTF_int::get_len_ordering(), pull_alias(), read_local(), CTF_int::spctr_pin_keys::spctr_pin_keys(), CTF_int::spctr_replicate::spctr_replicate(), CTF_int::summation::sum_tensors(), CTF::Matrix< dtype >::svd(), CTF_int::symmetrize(), CTF_int::tspsum_pin_keys::tspsum_pin_keys(), CTF_int::tspsum_replicate::tspsum_replicate(), CTF_int::tsum_replicate::tsum_replicate(), and write().
| CTF::World* CTF_int::tensor::wrld |
distributed processor context on which tensor is defined
Definition at line 70 of file untyped_tensor.h.
Referenced by align(), CTF::Tensor< dtype >::align(), ao_mo_transf_naive(), ao_mo_transf_slice(), btwn_cnt_fast(), btwn_cnt_naive(), check_asym(), check_sym(), compare(), copy_tensor_data(), CTF_int::desymmetrize(), divide_EaEi(), CTF_int::Bifun_Term::execute(), CTF_int::scaling::execute(), CTF::Tensor< dtype >::fill_random(), CTF::Tensor< dtype >::fill_sp_random(), CTF::fill_sp_random_base(), CTF::get_core_tensor(), CTF::Matrix< dtype >::get_desc(), CTF::get_factor_matrices(), CTF_int::get_full_intm(), CTF_int::get_len_ordering(), get_rand_as_tsr(), CTF::Scalar< dtype >::get_val(), mis2(), mp3(), CTF::Tensor< dtype >::norm1(), CTF::Tensor< dtype >::norm2(), CTF::Tensor< dtype >::norm_infty(), CTF_int::Term::operator double(), CTF_int::Term::operator float(), CTF_int::Term::operator int(), CTF_int::Term::operator int64_t(), CTF::Scalar< dtype >::operator=(), CTF::Tensor< dtype >::operator=(), permute(), CTF::Matrix< dtype >::print_matrix(), CTF::Matrix< dtype >::qr(), CTF::Matrix< dtype >::read_mat(), CTF::read_sparse_from_file_base(), rec_scan(), recursive_matmul(), scan(), CTF_int::seq_tsr_ctr::seq_tsr_ctr(), CTF_int::seq_tsr_spctr::seq_tsr_spctr(), setup(), slice(), CTF::Tensor< dtype >::slice(), smooth_jacobi(), sparse_mp3(), CTF_int::sum_bool_tsr(), CTF_int::summation::sum_tensors(), CTF::Matrix< dtype >::svd(), CTF_int::symmetrize(), tensor(), test_alg_multigrid(), test_mis(), test_mis2(), vcycle(), CTF::Idx_Tensor::where_am_i(), CTF::Matrix< dtype >::write_mat(), and CTF::write_sparse_to_file_base().
1.8.11 
