Wolfram Language Paclet Repository
Community-contributed installable additions to the Wolfram Language
Wolfram`TensorNetworks`
TensorNetworkTensors  |
|
| | ||||
|
| | ||||
Details and Options
Examples
(3)
Scope
(1)
The TN form returns the raw tensor list:
In[1]:=
SeedRandom[11];tn=TensorNetwork[{RandomReal[{-1,1},{2,3}],RandomReal[{-1,1},{3,4}]},{{i,j},{j,k}}];Map[Dimensions,TensorNetworkTensors[tn]]
Out[1]=
{{2,3},{3,4}}
The graph form reads the same tensors from the "Tensor" annotation on each vertex:
In[1]:=
SeedRandom[11];tn=TensorNetwork[{RandomReal[{-1,1},{2,3}],RandomReal[{-1,1},{3,4}]},{{i,j},{j,k}}];Map[Dimensions,TensorNetworkTensors[ToTensorNetworkGraph[tn]]]
Out[1]=
{{2,3},{3,4}}
The sparse form replaces every nonzero-rank tensor with a SparseArray:
In[1]:=
SeedRandom[11];tn=TensorNetwork[{RandomReal[{-1,1},{2,3}],RandomReal[{-1,1},{3,4}]},{{i,j},{j,k}}];Map[Head,TensorNetworkTensors[SparseTensorNetwork[tn]]]
Out[1]=
{SparseArray,SparseArray}
The property form is a synonym of the function call:
In[1]:=
SeedRandom[42];tn=RandomTensorNetwork["MPS"[3,2,2]];TensorNetworkTensors[tn]===tn["Tensors"]
Out[1]=
True
Applications
(1)
Properties & Relations
(1)
 |
|
 |
|
 |
|
Get the list of tensors in a tensor network:
In[1]:=
tn=TensorNetwork[{RandomReal[{-1,1},{2,3}],RandomReal[{-1,1},{3,4}]},{{i,j},{j,k}}];
In[2]:=
TensorNetworkTensors[tn]
A hypergraph-only network is auto-filled with symbolic ArraySymbol tensors:
In[1]:=
TensorNetworkTensors[TensorNetwork[{{1,2},{2,3}}]]
Out[1]=
ArraySymbolT.,d.,d.,ArraySymbolT.,d.,d.
A three-site MPS gives a list of three tensors:
In[1]:=
SeedRandom[7];Map[Dimensions,TensorNetworkTensors[RandomTensorNetwork["MPS"[3,2,2]]]]
Out[1]=
{{2,2},{2,2,2},{2,2}}
""