Package Biskit :: Module SparseArray :: Class SparseArray
[show private]
[frames] | no frames]

Class SparseArray

source code

Memory - saving representation of a sparse array.

Instance Methods [show private]
  __init__(self, array_or_shape=0, typecode='f', default=0.)
Create a sparse array from a normal Numeric array or create an empty sparse array with a certain shape.
str typecode(self)
Get type code of object.
number default(self)
Get default value, defined in __init__.
[ int ] nondefault(self)
Get a 1D list of indices that have a non-default value in a raveled version of this array.
[ number ] nondefault_values(self)
Get a 1D-list of all values != L.default() in the order that they would have in a raveled array.
array toarray(self)
Reconstruct dense array:
list tolist(self)
Reconstruct list from dense array:
  put(self, i, v)
Replace one or several values, L.put( i, v )
  __setitem__(self, i, v)
Set position specifyed by the index i to value v.
  __getitem__(self, i)
Value for specified position.
int __len__(self)
Get length: supports len( this )
0|1 __eq__(self, o)
Comparison equal: supports this == other
0|1 __ne__(self, o)
Comparison not equal: supports this != other
SparseArray __getslice__(self, a, b)
Slice a sparce array:
0|1 __contains__(self, v)
Sparse array contains value: supports v in this -> 0|1
int count(self, v)
Count the occuravces of value in sparse array:
int index(self, v)
position of first occurrence of value:
1|0 empty(self)
Check if object is empty.
  __delitem__(self, i)
Delete value at index i: supports del this[i]
  insert(self, i, v)
Insert value before index:
  append(self, v, axis=0)
Enlarge this array along an axis by one value:
  extend(self, lst)
Extend list by appending elements from the iterable:

Method Details [show private]

__init__(self, array_or_shape=0, typecode='f', default=0.)
(Constructor)

source code 

Create a sparse array from a normal Numeric array or create an empty sparse array with a certain shape.
Parameters:
  • array_or_shape (( int, ) OR int OR array) - craeate sparse array from:
                            ( int, ),   shape of array
                             OR int,     length of array
                             OR array,   numeric (dense) array
    
  • typecode (str) - single char type of values ['f' OR input type]
  • default (any) - value of majority of array content (default: 0.)

typecode(self)

source code 

Get type code of object.
Returns: str
typecode of lowest dimension

default(self)

source code 

Get default value, defined in __init__.
Returns: number
default value for array elements (of lowest dimension)

nondefault(self)

source code 

Get a 1D list of indices that have a non-default value in a raveled version of this array. If L.default()==0 this would be equivalent to nonzero( ravel( L.toarray() ) ) (except that the Numeric array is never constructed).
Returns: [ int ]
list of indices with none default values

nondefault_values(self)

source code 

Get a 1D-list of all values != L.default() in the order that they would have in a raveled array. If L.default()==0 this would be equivalent to take( L.toarray(), nonzero( ravel( L.toarray() ) ) ) (except that the Numeric array is never constructed).
Returns: [ number ]
list of none default values

toarray(self)

source code 

Reconstruct dense array:
 L.toarray() -> Numeric.array, normal dense array
Returns: array
normal dense array

tolist(self)

source code 

Reconstruct list from dense array:
 L.tolist() -> [ ([) number (]) ], same as L.toarray().tolist()
Returns: list
list from normal dense array

put(self, i, v)

source code 

Replace one or several values, L.put( i, v )
Parameters:
  • i (int OR [ int ]) - indices
  • v (any OR [ any ]) - values

__setitem__(self, i, v)
(Index assignment operator)

source code 

Set position specifyed by the index i to value v.
Parameters:
  • i (int OR (int,int)) - index
  • v (any) - value
Raises:

__getitem__(self, i)
(Indexing operator)

source code 

Value for specified position.
Parameters:
  • i (int OR (int,int)) - array index

__len__(self)
(Length operator)

source code 

Get length: supports len( this )
Returns: int
object length

__eq__(self, o)
(Equality operator)

source code 

Comparison equal: supports this == other
Returns: 0|1
result of comparison

__ne__(self, o)

source code 

Comparison not equal: supports this != other
Returns: 0|1
result of comparison

__getslice__(self, a, b)
(Slicling operator)

source code 

Slice a sparce array:
 this[ a : b ] -> SparseArray
Returns: SparseArray
sliced sparse array

__contains__(self, v)
(In operator)

source code 

Sparse array contains value: supports v in this -> 0|1
Parameters:
  • v (any) - value
Returns: 0|1
result of comparison

count(self, v)

source code 

Count the occuravces of value in sparse array:
 count( value ) -> int, number of occurences of value.
Parameters:
  • v (any) - value
Returns: int
number of occurances

index(self, v)

source code 

position of first occurrence of value:
 index( value ) -> int
Parameters:
  • v (any) - value to look for
Returns: int
index of first occurance
Raises:
  • ValueError - if value is not contained in this list

empty(self)

source code 

Check if object is empty.
Returns: 1|0
true if lenght is 0

__delitem__(self, i)
(Index deletion operator)

source code 

Delete value at index i: supports del this[i]
Parameters:
  • i (int) - index

Note: del this[int:int] is not supported

insert(self, i, v)

source code 

Insert value before index:
 this.insert(index, value) 
Parameters:
  • v (any) - value to insert
  • i (int) - index
Raises:
  • IndexError - if i < 0 or i > len( this )

append(self, v, axis=0)

source code 

Enlarge this array along an axis by one value:
 L.append( value, [ axis=0 ])
Parameters:
  • v (number OR SparseArray) - v MUST be a sparse array if the given axis is not the last axis of this array
  • axis (int) - dimension along wich v should be added
Raises:

extend(self, lst)

source code 

Extend list by appending elements from the iterable:
 L.extend(iterable)
Parameters:
  • lst ([any] OR SparseArray) - list OR SparseArray with extend values