Skip to content

Commit

Permalink
update tensor module
Browse files Browse the repository at this point in the history
  • Loading branch information
@obouchaara
obouchaara committed Jun 11, 2024
1 parent ab9333a commit 1720ff1
Show file tree
Hide file tree
Showing 2 changed files with 470 additions and 88 deletions.
146 changes: 81 additions & 65 deletions src/mechpy/core/symbolic/tensor.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,20 @@


class SymbolicTensor:
def __init__(self, data, name=None, tensor_params=None):
if not isinstance(data, sp.NDimArray):
raise ValueError("Input data must be a SymPy NDimArray.")

self.data = data
self.name = name
self.tensor_params = tensor_params or {}

if not isinstance(self.tensor_params, dict):
raise ValueError("Tensor parameters must be a dict.")

def __repr__(self):
return f"{self.__class__.__name__}(\n{self.data},\n{self.name}\n)"

@staticmethod
def notation_standard_map(n):
"""
Expand Down Expand Up @@ -62,16 +76,6 @@ def notation_standard_inverse_map(n):

return index_map_inverse

def __init__(self, data, name=None):
if isinstance(data, sp.ImmutableDenseNDimArray):
self.data = data
self.name = name
else:
raise ValueError("Input data must be a SymPy ImmutableDenseNDimArray")

def __repr__(self):
return f"{self.__class__.__name__}(\n{self.data},\n{self.name}\n)"

def is_second_rank(self):
return len(self.data.shape) == 2

Expand All @@ -85,9 +89,9 @@ def is_symmetric(self):
return self.is_square() and self.data == self.data.transpose()

def to_matrix(self):
if self.is_square():
return sp.Matrix(self.data.tolist())
raise ValueError("The Tensor should be a second rank square")
if not self.is_square():
raise ValueError("The Tensor should be a second rank square")
return sp.Matrix(self.data.tolist())

def to_3x3(self):
if self.data.shape == (3, 3):
Expand All @@ -106,35 +110,42 @@ def to_6x6(self):

@classmethod
def from_list(cls, components, shape):
if isinstance(components, list):
if isinstance(shape, tuple):
try:
data = sp.ImmutableDenseNDimArray(components, shape)
return cls(data)
except:
raise ValueError("Invalid components or shape value")
raise ValueError("Invalid shape parameter")
raise ValueError("Input must be a list")
if not isinstance(components, list):
raise ValueError("Components must be a list.")
if not isinstance(shape, tuple):
raise ValueError("Shape must be a tuple.")

# need components length validation

data = sp.NDimArray(components, shape=shape)
return cls(data)

@classmethod
def create(cls, shape, name):
if not isinstance(shape, tuple):
raise ValueError("Shape must be a tuple")
raise ValueError("Shape must be a tuple.")
if name is None:
raise ValueError("Name cannot be None.")

indices = itertools.product(*(range(1, dim_len + 1) for dim_len in shape))
components = [sp.symbols(f"{name}_{''.join(map(str, idx))}") for idx in indices]
data = sp.ImmutableDenseNDimArray(components, shape)
data = sp.NDimArray(components, shape)
return cls(data, name=name)

def subs_tensor_params(self, param_values):
pass

def subs(self, subs_dict, keys=False):
try:
if keys:
for k, v in subs_dict.items():
component = self.data[k]
self.data = self.data.subs({component: v})
else:
self.data = self.data.subs(subs_dict)
except Exception as e:
raise RuntimeError(f"An error occurred during substitution: {e}")
raise DeprecationWarning
# try:
# if keys:
# for k, v in subs_dict.items():
# component = self.data[k]
# self.data = self.data.subs({component: v})
# else:
# self.data = self.data.subs(subs_dict)
# except Exception as e:
# raise RuntimeError(f"An error occurred during substitution: {e}")

def __matmul__(self, other):
if not isinstance(other, SymbolicTensor):
Expand All @@ -148,7 +159,7 @@ def __matmul__(self, other):
"Shapes are not aligned for matrix multiplication: "
f"{self.data.shape} and {other.data.shape}"
)
result_data = sp.ImmutableDenseNDimArray(self.to_matrix() @ other.to_matrix())
result_data = sp.NDimArray(self.to_matrix() @ other.to_matrix())
return SymbolicTensor(result_data)

def __getitem__(self, key):
Expand All @@ -167,44 +178,38 @@ def diagonalize(self):
class SymbolicThreeByThreeTensor(SymbolicTensor):
shape = (3, 3)

def __init__(self, data, name=None):
if isinstance(data, sp.MutableDenseNDimArray):
data = sp.ImmutableDenseNDimArray(data)
if isinstance(data, sp.ImmutableDenseNDimArray) and data.shape == self.shape:
super().__init__(data, name)
else:
raise ValueError("Input data must be a 3x3 SymPy Array")
def __init__(self, data, name=None, tensor_params=None):
if not isinstance(data, sp.NDimArray) or data.shape != self.shape:
raise ValueError("Data must be a 3x3 SymPy Array.")
super().__init__(data, name, tensor_params)

@classmethod
def from_list(cls, components):
return super().from_list(components, cls.shape)

@classmethod
def create(cls, name):
return super().create(cls.shape, name=name)
return super().create(cls.shape, name)

def to_symmetric(self, notation=1):
def to_symmetric(self, notation="Standard notation"):
if self.is_symmetric():
NOTATIONS = SymbolicSymmetricThreeByThreeTensor.NOTATIONS
if notation in NOTATIONS.keys():
mapping = NOTATIONS[notation]["inverse_map"]
components = [self.data[mapping[i]] for i in range(6)]
data = sp.ImmutableDenseNDimArray(components)
data = sp.NDimArray(components)
return SymbolicSymmetricThreeByThreeTensor(data, notation=notation)
raise NotImplementedError(f"Notation {notation} not implemented")
raise ValueError("The tensor is not symmetric")
raise ValueError("The tensor is not symmetric.")


class SymbolicSixBySixTensor(SymbolicTensor):
shape = (6, 6)

def __init__(self, data, name=None):
if isinstance(data, sp.MutableDenseNDimArray):
data = sp.ImmutableDenseNDimArray(data)
if isinstance(data, sp.ImmutableDenseNDimArray) and data.shape == self.shape:
super().__init__(data, name)
else:
raise ValueError("Input data must be a 6x6 SymPy Array")
def __init__(self, data, name=None, tensor_params=None):
if not isinstance(data, sp.NDimArray) or data.shape != self.shape:
raise ValueError("Data must be a 6x6 SymPy Array.")
super().__init__(data, name, tensor_params)

@classmethod
def from_list(cls, components):
Expand Down Expand Up @@ -261,6 +266,14 @@ class SymbolicSymmetricThreeByThreeTensor(SymbolicTensor):
}

NOTATIONS = {
"Standard notation": {
"map": STANDARD_MAPPING,
"inverse_map": INVERSE_STANDARD_MAPPING,
},
"Voight notation": {
"map": VOIGT_MAPPING,
"inverse_map": INVERSE_VOIGT_MAPPING,
},
1: {
"name": "Standard notation",
"map": STANDARD_MAPPING,
Expand All @@ -273,17 +286,20 @@ class SymbolicSymmetricThreeByThreeTensor(SymbolicTensor):
},
}

def __init__(self, data, name=None, notation=1):
if isinstance(data, sp.MutableDenseNDimArray):
data = sp.ImmutableDenseNDimArray(data)
if isinstance(data, sp.ImmutableDenseNDimArray) and data.shape == self.shape:
if notation in self.NOTATIONS.keys():
super().__init__(data, name=name)
self.notation = notation
else:
raise NotImplementedError(f"Notation {notation} not implemented")
else:
raise ValueError(f"Input data must be a SymPy Array ans sahpe={self.shape}")
def __init__(
self,
data,
name=None,
notation="Standard notation",
tensor_params=None,
):
if not isinstance(data, sp.NDimArray) or data.shape != self.shape:
raise ValueError("Data must be a 6x1 SymPy Array.")
if not notation in self.NOTATIONS.keys():
raise NotImplementedError(f"Notation {notation} not implemented")

super().__init__(data, name, tensor_params)
self.notation = notation

def __repr__(self):
return f"{self.__class__.__name__}(\n{self.data},\n{self.name},\n{self.notation}\n)"
Expand All @@ -296,7 +312,7 @@ def from_list(cls, components, name=None, notation=1):
if isinstance(components, list):
if len(components) == 6:
if notation in cls.NOTATIONS.keys():
data = sp.ImmutableDenseNDimArray(components)
data = sp.NDimArray(components)
return cls(data, name=name, notation=notation)
raise NotImplementedError(f"Notation {notation} not implemented")
elif len(components) == 9:
Expand All @@ -320,7 +336,7 @@ def create(cls, name, notation=1):
def to_general(self):
mapping = self.NOTATIONS[self.notation]["map"]
components = [[self.data[mapping[(i, j)]] for j in range(3)] for i in range(3)]
data = sp.ImmutableDenseNDimArray(components)
data = sp.NDimArray(components)
return SymbolicThreeByThreeTensor(data)

def __getitem__(self, key):
Expand Down

0 comments on commit 1720ff1

Please sign in to comment.