# sparse – Symbolic Sparse Matrices#

In the tutorial section, you can find a sparse tutorial.

The sparse submodule is not loaded when we import PyTensor. You must import pytensor.sparse to enable it.

The sparse module provides the same functionality as the tensor module. The difference lies under the covers because sparse matrices do not store data in a contiguous array. The sparse module has been used in:

• NLP: Dense linear transformations of sparse vectors.

• Audio: Filterbank in the Fourier domain.

## Compressed Sparse Format#

This section tries to explain how information is stored for the two sparse formats of SciPy supported by PyTensor.

PyTensor supports two compressed sparse formats: csc and csr, respectively based on columns and rows. They have both the same attributes: data, indices, indptr and shape.

• The data attribute is a one-dimensional ndarray which contains all the non-zero elements of the sparse matrix.

• The indices and indptr attributes are used to store the position of the data in the sparse matrix.

• The shape attribute is exactly the same as the shape attribute of a dense (i.e. generic) matrix. It can be explicitly specified at the creation of a sparse matrix if it cannot be inferred from the first three attributes.

### CSC Matrix#

In the Compressed Sparse Column format, indices stands for indexes inside the column vectors of the matrix and indptr tells where the column starts in the data and in the indices attributes. indptr can be thought of as giving the slice which must be applied to the other attribute in order to get each column of the matrix. In other words, slice(indptr[i], indptr[i+1]) corresponds to the slice needed to find the i-th column of the matrix in the data and indices fields.

The following example builds a matrix and returns its columns. It prints the i-th column, i.e. a list of indices in the column and their corresponding value in the second list.

>>> import numpy as np
>>> import scipy.sparse as sp
>>> data = np.asarray([7, 8, 9])
>>> indices = np.asarray([0, 1, 2])
>>> indptr = np.asarray([0, 2, 3, 3])
>>> m = sp.csc_matrix((data, indices, indptr), shape=(3, 3))
>>> m.toarray()
array([[7, 0, 0],
[8, 0, 0],
[0, 9, 0]])
>>> i = 0
>>> m.indices[m.indptr[i]:m.indptr[i+1]], m.data[m.indptr[i]:m.indptr[i+1]]
(array([0, 1], dtype=int32), array([7, 8]))
>>> i = 1
>>> m.indices[m.indptr[i]:m.indptr[i+1]], m.data[m.indptr[i]:m.indptr[i+1]]
(array([2], dtype=int32), array([9]))
>>> i = 2
>>> m.indices[m.indptr[i]:m.indptr[i+1]], m.data[m.indptr[i]:m.indptr[i+1]]
(array([], dtype=int32), array([], dtype=int64))


### CSR Matrix#

In the Compressed Sparse Row format, indices stands for indexes inside the row vectors of the matrix and indptr tells where the row starts in the data and in the indices attributes. indptr can be thought of as giving the slice which must be applied to the other attribute in order to get each row of the matrix. In other words, slice(indptr[i], indptr[i+1]) corresponds to the slice needed to find the i-th row of the matrix in the data and indices fields.

The following example builds a matrix and returns its rows. It prints the i-th row, i.e. a list of indices in the row and their corresponding value in the second list.

>>> import numpy as np
>>> import scipy.sparse as sp
>>> data = np.asarray([7, 8, 9])
>>> indices = np.asarray([0, 1, 2])
>>> indptr = np.asarray([0, 2, 3, 3])
>>> m = sp.csr_matrix((data, indices, indptr), shape=(3, 3))
>>> m.toarray()
array([[7, 8, 0],
[0, 0, 9],
[0, 0, 0]])
>>> i = 0
>>> m.indices[m.indptr[i]:m.indptr[i+1]], m.data[m.indptr[i]:m.indptr[i+1]]
(array([0, 1], dtype=int32), array([7, 8]))
>>> i = 1
>>> m.indices[m.indptr[i]:m.indptr[i+1]], m.data[m.indptr[i]:m.indptr[i+1]]
(array([2], dtype=int32), array([9]))
>>> i = 2
>>> m.indices[m.indptr[i]:m.indptr[i+1]], m.data[m.indptr[i]:m.indptr[i+1]]
(array([], dtype=int32), array([], dtype=int64))


## List of Implemented Operations#

• Moving from and to sparse
• dense_from_sparse. Both grads are implemented. Structured by default.

• csr_from_dense, csc_from_dense. The grad implemented is structured.

• PyTensor SparseVariable objects have a method toarray() that is the same as dense_from_sparse.

• Construction of Sparses and their Properties
• CSM and CSC, CSR to construct a matrix. The grad implemented is regular.

• csm_properties. to get the properties of a sparse matrix. The grad implemented is regular.

• csm_indices(x), csm_indptr(x), csm_data(x) and csm_shape(x) or x.shape.

• sp_ones_like. The grad implemented is regular.

• sp_zeros_like. The grad implemented is regular.

• square_diagonal. The grad implemented is regular.

• construct_sparse_from_list. The grad implemented is regular.

• Cast
• cast with bcast, wcast, icast, lcast, fcast, dcast, ccast, and zcast. The grad implemented is regular.

• Transpose
• transpose. The grad implemented is regular.

• Basic Arithmetic
• neg. The grad implemented is regular.

• eq.

• neq.

• gt.

• ge.

• lt.

• le.

• add. The grad implemented is regular.

• sub. The grad implemented is regular.

• mul. The grad implemented is regular.

• col_scale to multiply by a vector along the columns. The grad implemented is structured.

• row_scale to multiply by a vector along the rows. The grad implemented is structured.

• Monoid (Element-wise operation with only one sparse input).

They all have a structured grad.

• structured_sigmoid

• structured_exp

• structured_log

• structured_pow

• structured_minimum

• structured_maximum

• structured_add

• sin

• arcsin

• tan

• arctan

• sinh

• arcsinh

• tanh

• arctanh

• rad2deg

• deg2rad

• rint

• ceil

• floor

• trunc

• sign

• log1p

• expm1

• sqr

• sqrt

• Dot Product
• One of the inputs must be sparse, the other sparse or dense.

• The grad implemented is regular.

• No C code for perform and no C code for grad.

• Returns a dense for perform and a dense for grad.

• The first input is sparse, the second can be sparse or dense.

• The grad implemented is structured.

• C code for perform and grad.

• It returns a sparse output if both inputs are sparse and dense one if one of the inputs is dense.

• Returns a sparse grad for sparse inputs and dense grad for dense inputs.

• The first input is sparse, the second can be sparse or dense.

• The grad implemented is regular.

• No C code for perform and no C code for grad.

• Returns a Sparse.

• The gradient returns a Sparse for sparse inputs and by default a dense for dense inputs. The parameter grad_preserves_dense can be set to False to return a sparse grad for dense inputs.

• sampling_dot.

• Both inputs must be dense.

• The grad implemented is structured for p.

• Sample of the dot and sample of the gradient.

• C code for perform but not for grad.

• Returns sparse for perform and grad.

• usmm.

• You shouldn’t insert this op yourself!
• There is a rewrite that transforms a dot to Usmm when possible.

• This Op is the equivalent of gemm for sparse dot.

• There is no grad implemented for this Op.

• One of the inputs must be sparse, the other sparse or dense.

• Returns a dense from perform.

• Slice Operations
• sparse_variable[N, N], returns a tensor scalar. There is no grad implemented for this operation.

• sparse_variable[M:N, O:P], returns a sparse matrix There is no grad implemented for this operation.

• Sparse variables don’t support [M, N:O] and [M:N, O] as we don’t support sparse vectors and returning a sparse matrix would break the numpy interface. Use [M:M+1, N:O] and [M:N, O:O+1] instead.

• diag. The grad implemented is regular.

• Concatenation
• Probability

There is no grad implemented for these operations.

• Poisson and poisson

• Binomial and csc_fbinomial, csc_dbinomial csr_fbinomial, csr_dbinomial

• Multinomial and multinomial

• Internal Representation

They all have a regular grad implemented.

• ensure_sorted_indices.

• remove0.

• clean to resort indices and remove zeros

• To help testing
• tests.sparse.test_basic.sparse_random_inputs()

# sparse – Sparse Op#

Classes for handling sparse matrices.

TODO: Automatic methods for determining best sparse format?

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, y)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, y)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

Add two sparse matrices assuming they have the same sparsity pattern.

Notes

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, y)[source]#
Parameters:
• x – Sparse matrix.

• y – Sparse matrix.

Notes

x and y are assumed to have the same sparsity pattern.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.CSM(format, kmap=None)[source]#

Construct a CSM matrix from constituent parts.

Notes

The grad method returns a dense vector, so it provides a regular grad.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(data, indices, indptr, shape)[source]#
Parameters:
• data – One dimensional tensor representing the data of the sparse matrix to construct.

• indices – One dimensional tensor of integers representing the indices of the sparse matrix to construct.

• indptr – One dimensional tensor of integers representing the indice pointer for the sparse matrix to construct.

• shape – One dimensional tensor of integers representing the shape of the sparse matrix to construct.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

Compute the gradient of a CSM.

Note

CSM creates a matrix from data, indices, and indptr vectors; it’s gradient is the gradient of the data vector only. There are two complexities to calculate this gradient:

1. The gradient may be sparser than the input matrix defined by (data, indices, indptr). In this case, the data vector of the gradient will have less elements than the data vector of the input because sparse formats remove 0s. Since we are only returning the gradient of the data vector, the relevant 0s need to be added back. 2. The elements in the sparse dimension are not guaranteed to be sorted. Therefore, the input data vector may have a different order than the gradient data vector.

make_node(x_data, x_indices, x_indptr, x_shape, g_data, g_indices, g_indptr, g_shape)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.CSMProperties(kmap=None)[source]#

Create arrays containing all the properties of a given sparse matrix.

More specifically, this Op extracts the .data, .indices, .indptr and .shape fields.

For specific field, csm_data, csm_indices, csm_indptr and csm_shape are provided.

Notes

The grad implemented is regular, i.e. not structured. infer_shape method is not available for this Op.

We won’t implement infer_shape for this op now. This will ask that we implement an GetNNZ op, and this op will keep the dependence on the input of this op. So this won’t help to remove computations in the graph. To remove computation, we will need to make an infer_sparse_pattern feature to remove computations. Doing this is trickier then the infer_shape feature. For example, how do we handle the case when some op create some 0 values? So there is dependence on the values themselves. We could write an infer_shape for the last output that is the shape, but I dough this will get used.

We don’t return a view of the shape, we create a new ndarray from the shape tuple.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(csm)[source]#

The output vectors correspond to the tuple (data, indices, indptr, shape), i.e. the properties of a csm array.

Parameters:

csm – Sparse matrix in CSR or CSC format.

perform(node, inputs, out)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

view_map: dict[int, list[int]] = {0: [0], 1: [0], 2: [0]}[source]#

A dict that maps output indices to the input indices of which they are a view.

Examples

view_map = {0: [1]} # first output is a view of second input
view_map = {1: [0]} # second output is a view of first input

class pytensor.sparse.basic.Cast(out_type)[source]#

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.ColScaleCSC[source]#

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, s)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.ConstructSparseFromList[source]#

Constructs a sparse matrix out of a list of 2-D matrix rows.

Notes

The grad implemented is regular, i.e. not structured.

R_op(inputs, eval_points)[source]#

Construct a graph for the R-operator.

This method is primarily used by Rop.

Parameters:
• inputs – The Op inputs.

• eval_points – A Variable or list of Variables with the same length as inputs. Each element of eval_points specifies the value of the corresponding input at the point where the R-operator is to be evaluated.

Return type:

rval[i] should be Rop(f=f_i(inputs), wrt=inputs, eval_points=eval_points).

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, values, ilist)[source]#

This creates a sparse matrix with the same shape as x. Its values are the rows of values moved. It operates similar to the following pseudo-code:

output = csc_matrix.zeros_like(x, dtype=values.dtype)
for in_idx, out_idx in enumerate(ilist):
output[out_idx] = values[in_idx]

Parameters:
• x – A dense matrix that specifies the output shape.

• values – A dense matrix with the values to use for output.

• ilist – A dense vector with the same length as the number of rows of values. It specifies where in the output to put the corresponding rows.

perform(node, inp, out_)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.DenseFromSparse(structured=True)[source]#

Convert a sparse matrix to a dense one.

Notes

The grad implementation can be controlled through the constructor via the structured parameter. True will provide a structured grad while False will provide a regular grad. By default, the grad is structured.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x)[source]#
Parameters:

x – A sparse matrix.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.Diag[source]#

Extract the diagonal of a square sparse matrix as a dense vector.

Notes

The grad implemented is regular, i.e. not structured, since the output is a dense vector.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x)[source]#
Parameters:

x – A square sparse matrix in csc format.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.Dot[source]#

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, y)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, out)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.EnsureSortedIndices(inplace)[source]#

Re-sort indices of a sparse matrix.

CSR column indices are not necessarily sorted. Likewise for CSC row indices. Use ensure_sorted_indices when sorted indices are required (e.g. when passing data to other libraries).

Notes

The grad implemented is regular, i.e. not structured.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x)[source]#
Parameters:

x – A sparse matrix.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.EqualSD[source]#
class pytensor.sparse.basic.EqualSS[source]#
class pytensor.sparse.basic.GetItem2Lists[source]#

Select elements of sparse matrix, returning them in a vector.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, ind1, ind2)[source]#
Parameters:
• x – Sparse matrix.

• index – List of two lists, first list indicating the row of each element and second list indicating its column.

perform(node, inp, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

make_node(x, ind1, ind2, gz)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inp, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.GetItem2d[source]#

Implement a subtensor of sparse variable, returning a sparse matrix.

If you want to take only one element of a sparse matrix see GetItemScalar that returns a tensor scalar.

Notes

Subtensor selection always returns a matrix, so indexing with [a:b, c:d] is forced. If one index is a scalar, for instance, x[a:b, c] or x[a, b:c], an error will be raised. Use instead x[a:b, c:c+1] or x[a:a+1, b:c].

The above indexing methods are not supported because the return value would be a sparse matrix rather than a sparse vector, which is a deviation from numpy indexing rule. This decision is made largely to preserve consistency between numpy and pytensor. This may be revised when sparse vectors are supported.

The grad is not implemented for this op.

make_node(x, index)[source]#
Parameters:
• x – Sparse matrix.

• index – Tuple of slice object.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.GetItemList[source]#

Select row of sparse matrix, returning them as a new sparse matrix.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, index)[source]#
Parameters:
• x – Sparse matrix.

• index – List of rows.

perform(node, inp, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

make_node(x, index, gz)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inp, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.GetItemScalar[source]#

Subtensor of a sparse variable that takes two scalars as index and returns a scalar.

If you want to take a slice of a sparse matrix see GetItem2d that returns a sparse matrix.

Notes

The grad is not implemented for this op.

make_node(x, index)[source]#
Parameters:
• x – Sparse matrix.

• index – Tuple of scalars.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.GreaterEqualSD[source]#
class pytensor.sparse.basic.GreaterEqualSS[source]#
class pytensor.sparse.basic.GreaterThanSD[source]#
class pytensor.sparse.basic.GreaterThanSS[source]#
class pytensor.sparse.basic.HStack(format=None, dtype=None)[source]#

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(*mat)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, block, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.LessEqualSD[source]#
class pytensor.sparse.basic.LessEqualSS[source]#
class pytensor.sparse.basic.LessThanSD[source]#
class pytensor.sparse.basic.LessThanSS[source]#
class pytensor.sparse.basic.MulSD[source]#

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, y)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.MulSS[source]#

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, y)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.MulSV[source]#

Element-wise multiplication of sparse matrix by a broadcasted dense vector element wise.

Notes

The grad implemented is regular, i.e. not structured.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, y)[source]#
Parameters:
• x – Sparse matrix to multiply.

• y – Tensor broadcastable vector.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.Neg[source]#

Negative of the sparse matrix (i.e. multiply by -1).

Notes

The grad is regular, i.e. not structured.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x)[source]#
Parameters:

x – Sparse matrix.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.NotEqualSD[source]#
class pytensor.sparse.basic.NotEqualSS[source]#
class pytensor.sparse.basic.Remove0(inplace=False)[source]#

Remove explicit zeros from a sparse matrix.

Notes

The grad implemented is regular, i.e. not structured.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x)[source]#
Parameters:

x – Sparse matrix.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.RowScaleCSC[source]#

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, s)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

view_map: dict[int, list[int]] = {0: [0]}[source]#

A dict that maps output indices to the input indices of which they are a view.

Examples

view_map = {0: [1]} # first output is a view of second input
view_map = {1: [0]} # second output is a view of first input

class pytensor.sparse.basic.SamplingDot[source]#

Compute the dot product dot(x, y.T) = z for only a subset of z.

This is equivalent to p * (x . y.T) where * is the element-wise product, x and y operands of the dot product and p is a matrix that contains 1 when the corresponding element of z should be calculated and 0 when it shouldn’t. Note that SamplingDot has a different interface than dot because it requires x to be a m x k matrix while y is a n x k matrix instead of the usual k x n matrix.

Notes

It will work if the pattern is not binary value, but if the pattern doesn’t have a high sparsity proportion it will be slower then a more optimized dot followed by a normal elemwise multiplication.

The grad implemented is regular, i.e. not structured.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, y, p)[source]#
Parameters:
• x – Tensor matrix.

• y – Tensor matrix.

• p – Sparse matrix in csr format.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

WARNING: judgement call… We are not using the structured in the comparison or hashing because it doesn’t change the perform method therefore, we do want Sums with different structured values to be merged by the merge optimization and this requires them to compare equal.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.SparseBlockDiagonal(n_inputs: int, format: Literal['csc', 'csr'] = 'csc')[source]#
make_node(*matrices)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, output_storage, params=None)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.SparseConstant(type: _TensorTypeType, data, name=None)[source]#
class pytensor.sparse.basic.SparseConstantSignature(iterable=(), /)[source]#
class pytensor.sparse.basic.SparseFromDense(format)[source]#

Convert a dense matrix to a sparse matrix.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x)[source]#
Parameters:

x – A dense matrix.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.SparseVariable(type: _TensorTypeType, owner: OptionalApplyType, index=None, name=None)[source]#
class pytensor.sparse.basic.SquareDiagonal[source]#

Produce a square sparse (csc) matrix with a diagonal given by a dense vector.

Notes

The grad implemented is regular, i.e. not structured.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(diag)[source]#
Parameters:

x – Dense vector for the diagonal.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

Structured addition of a sparse matrix and a dense vector.

The elements of the vector are only added to the corresponding non-zero elements of the sparse matrix. Therefore, this operation outputs another sparse matrix.

Notes

The grad implemented is structured since the op is structured.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, y)[source]#
Parameters:
• x – Sparse matrix.

• y – Tensor type vector.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.StructuredDot[source]#

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(a, b)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

c_code(node, name, inputs, outputs, sub)[source]#

Return the C implementation of an Op.

Returns C code that does the computation associated to this Op, given names for the inputs and outputs.

Parameters:
• node (Apply instance) – The node for which we are compiling the current C code. The same Op may be used in more than one node.

• name (str) – A name that is automatically assigned and guaranteed to be unique.

• inputs (list of strings) – There is a string for each input of the function, and the string is the name of a C variable pointing to that input. The type of the variable depends on the declared type of the input. There is a corresponding python variable that can be accessed by prepending "py_" to the name in the list.

• outputs (list of strings) – Each string is the name of a C variable where the Op should store its output. The type depends on the declared type of the output. There is a corresponding Python variable that can be accessed by prepending "py_" to the name in the list. In some cases the outputs will be preallocated and the value of the variable may be pre-filled. The value for an unallocated output is type-dependent.

• sub (dict of strings) – Extra symbols defined in CLinker sub symbols (such as 'fail').

c_code_cache_version()[source]#

Return a tuple of integers indicating the version of this Op.

An empty tuple indicates an “unversioned” Op that will not be cached between processes.

The cache mechanism may erase cached modules that have been superseded by newer versions. See ModuleCache for details.

c_code_cache_version_apply

make_node(a_indices, a_indptr, b, g_ab)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

c_code(node, name, inputs, outputs, sub)[source]#

Return the C implementation of an Op.

Returns C code that does the computation associated to this Op, given names for the inputs and outputs.

Parameters:
• node (Apply instance) – The node for which we are compiling the current C code. The same Op may be used in more than one node.

• name (str) – A name that is automatically assigned and guaranteed to be unique.

• inputs (list of strings) – There is a string for each input of the function, and the string is the name of a C variable pointing to that input. The type of the variable depends on the declared type of the input. There is a corresponding python variable that can be accessed by prepending "py_" to the name in the list.

• outputs (list of strings) – Each string is the name of a C variable where the Op should store its output. The type depends on the declared type of the output. There is a corresponding Python variable that can be accessed by prepending "py_" to the name in the list. In some cases the outputs will be preallocated and the value of the variable may be pre-filled. The value for an unallocated output is type-dependent.

• sub (dict of strings) – Extra symbols defined in CLinker sub symbols (such as 'fail').

c_code_cache_version()[source]#

Return a tuple of integers indicating the version of this Op.

An empty tuple indicates an “unversioned” Op that will not be cached between processes.

The cache mechanism may erase cached modules that have been superseded by newer versions. See ModuleCache for details.

c_code_cache_version_apply

make_node(a_indices, a_indptr, b, g_ab)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.Transpose[source]#

Transpose of a sparse matrix.

Notes

The returned matrix will not be in the same format. csc matrix will be changed in csr matrix and csr matrix in csc matrix.

The grad is regular, i.e. not structured.

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x)[source]#
Parameters:

x – Sparse matrix.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

view_map: dict[int, list[int]] = {0: [0]}[source]#

A dict that maps output indices to the input indices of which they are a view.

Examples

view_map = {0: [1]} # first output is a view of second input
view_map = {1: [0]} # second output is a view of first input


Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

make_node(x, y)[source]#

Construct an Apply node that represent the application of this operation to the given inputs.

This must be implemented by sub-classes.

Returns:

node – The constructed Apply node.

Return type:

Apply

perform(node, inp, out_)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.Usmm[source]#

Computes the dense matrix resulting from alpha * x @ y + z.

Notes

At least one of x or y must be a sparse matrix.

make_node(alpha, x, y, z)[source]#
Parameters:
• alpha – A scalar.

• x – Matrix variable.

• y – Matrix variable.

• z – Dense matrix.

perform(node, inputs, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

class pytensor.sparse.basic.VStack(format=None, dtype=None)[source]#

Construct a graph for the gradient with respect to each input variable.

Each returned Variable represents the gradient with respect to that input computed based on the symbolic gradients with respect to each output. If the output is not differentiable with respect to an input, then this method should return an instance of type NullType for that input.

Using the reverse-mode AD characterization given in [1]_, for a $$C = f(A, B)$$ representing the function implemented by the Op and its two arguments $$A$$ and $$B$$, given by the Variables in inputs, the values returned by Op.grad represent the quantities $$\bar{A} \equiv \frac{\partial S_O}{A}$$ and $$\bar{B}$$, for some scalar output term $$S_O$$ of $$C$$ in

$\operatorname{Tr}\left(\bar{C}^\top dC\right) = \operatorname{Tr}\left(\bar{A}^\top dA\right) + \operatorname{Tr}\left(\bar{B}^\top dB\right)$
Parameters:
• inputs – The input variables.

Returns:

The gradients with respect to each Variable in inputs.

Return type:

References

perform(node, block, outputs)[source]#

Calculate the function on the inputs and put the variables in the output storage.

Parameters:
• node – The symbolic Apply node that represents this computation.

• inputs – Immutable sequence of non-symbolic/numeric inputs. These are the values of each Variable in node.inputs.

• output_storage – List of mutable single-element lists (do not change the length of these lists). Each sub-list corresponds to value of each Variable in node.outputs. The primary purpose of this method is to set the values of these sub-lists.

Notes

The output_storage list might contain data. If an element of output_storage is not None, it has to be of the right type, for instance, for a TensorVariable, it has to be a NumPy ndarray with the right number of dimensions and the correct dtype. Its shape and stride pattern can be arbitrary. It is not guaranteed that such pre-set values were produced by a previous call to this Op.perform(); they could’ve been allocated by another Op’s perform method. An Op is free to reuse output_storage as it sees fit, or to discard it and allocate new memory.

Add two matrices, at least one of which is sparse.

This method will provide the right op according to the inputs.

Parameters:
• x – A matrix variable.

• y – A matrix variable.

Returns:

x + y

Return type:

A sparse matrix

Notes

At least one of x and y must be a sparse matrix.

The grad will be structured only when one of the variable will be a dense matrix.

pytensor.sparse.basic.as_sparse(x, name=None, ndim=None, **kwargs)[source]#

Wrapper around SparseVariable constructor to construct a Variable with a sparse matrix with the same dtype and format.

Parameters:

x – A sparse matrix.

Returns:

SparseVariable version of x.

Return type:

object

pytensor.sparse.basic.as_sparse_variable(x, name=None, ndim=None, **kwargs)[source]#

Wrapper around SparseVariable constructor to construct a Variable with a sparse matrix with the same dtype and format.

Parameters:

x – A sparse matrix.

Returns:

SparseVariable version of x.

Return type:

object

pytensor.sparse.basic.block_diag(*matrices: TensorVariable, format: Literal['csc', 'csr'] = 'csc')[source]#

Construct a block diagonal matrix from a sequence of input matrices.

Given the inputs A, B and C, the output will have these arrays arranged on the diagonal:

[[A, 0, 0],

[0, B, 0], [0, 0, C]]

Parameters:
• A (tensors) –

Input tensors to form the block diagonal matrix. last two dimensions of the inputs will be used, and all inputs should have at least 2 dimensins.

Note that the input matrices need not be sparse themselves, and will be automatically converted to the requested format if they are not.

• B (tensors) –

Input tensors to form the block diagonal matrix. last two dimensions of the inputs will be used, and all inputs should have at least 2 dimensins.

Note that the input matrices need not be sparse themselves, and will be automatically converted to the requested format if they are not.

• ... (C) –

Input tensors to form the block diagonal matrix. last two dimensions of the inputs will be used, and all inputs should have at least 2 dimensins.

Note that the input matrices need not be sparse themselves, and will be automatically converted to the requested format if they are not.

• format (str, optional) – The format of the output sparse matrix. One of ‘csr’ or ‘csc’. Default is ‘csr’. Ignored if sparse=False.

Returns:

out – Symbolic sparse matrix in the specified format.

Return type:

sparse matrix tensor

Examples

Create a sparse block diagonal matrix from two sparse 2x2 matrices:

..code-block:: python

import numpy as np from pytensor.sparse import block_diag from scipy.sparse import csr_matrix

A = csr_matrix([[1, 2], [3, 4]]) B = csr_matrix([[5, 6], [7, 8]]) result_sparse = block_diag(A, B, format=’csr’, name=’X’)

print(result_sparse) >>> SparseVariable{csr,int32}

print(result_sparse.toarray().eval()) >>> array([[1, 2, 0, 0], >>> [3, 4, 0, 0], >>> [0, 0, 5, 6], >>> [0, 0, 7, 8]])

pytensor.sparse.basic.cast(variable, dtype)[source]#

Cast sparse variable to the desired dtype.

Parameters:
• variable – Sparse matrix.

• dtype – The dtype wanted.

Return type:

Same as x but having dtype as dtype.

Notes

The grad implemented is regular, i.e. not structured.

pytensor.sparse.basic.clean(x)[source]#

Remove explicit zeros from a sparse matrix, and re-sort indices.

CSR column indices are not necessarily sorted. Likewise for CSC row indices. Use clean when sorted indices are required (e.g. when passing data to other libraries) and to ensure there are no zeros in the data.

Parameters:

x – A sparse matrix.

Returns:

The same as x with indices sorted and zeros removed.

Return type:

A sparse matrix

Notes

The grad implemented is regular, i.e. not structured.

pytensor.sparse.basic.col_scale(x, s)[source]#

Scale each columns of a sparse matrix by the corresponding element of a dense vector.

Parameters:
• x – A sparse matrix.

• s – A dense vector with length equal to the number of columns of x.

Returns:

• A sparse matrix in the same format as x which each column had been

• multiply by the corresponding element of s.

Notes

pytensor.sparse.basic.csm_data(csm)[source]#

Return the data field of the sparse variable.

alias of CSMGrad

pytensor.sparse.basic.csm_indices(csm)[source]#

Return the indices field of the sparse variable.

pytensor.sparse.basic.csm_indptr(csm)[source]#

Return the indptr field of the sparse variable.

pytensor.sparse.basic.csm_shape(csm)[source]#

Return the shape field of the sparse variable.

pytensor.sparse.basic.dot(x, y)[source]#

Efficiently compute the dot product when one or all operands are sparse.

Supported formats are CSC and CSR. The output of the operation is dense.

Parameters:
• x – Sparse or dense matrix variable.

• y – Sparse or dense matrix variable.

Return type:

The dot product x @ y in a dense format.

Notes

The grad implemented is regular, i.e. not structured.

At least one of x or y must be a sparse matrix.

When the operation has the form dot(csr_matrix, dense) the gradient of this operation can be performed inplace by UsmmCscDense. This leads to significant speed-ups.

pytensor.sparse.basic.hstack(blocks, format=None, dtype=None)[source]#

Stack sparse matrices horizontally (column wise).

This wrap the method hstack from scipy.

Parameters:
• blocks – List of sparse array of compatible shape.

• format – String representing the output format. Default is csc.

• dtype – Output dtype.

Returns:

The concatenation of the sparse array column wise.

Return type:

array

Notes

The number of line of the sparse matrix must agree.

The grad implemented is regular, i.e. not structured.

pytensor.sparse.basic.mul(x, y)[source]#

Multiply elementwise two matrices, at least one of which is sparse.

This method will provide the right op according to the inputs.

Parameters:
• x – A matrix variable.

• y – A matrix variable.

Returns:

x * y

Return type:

A sparse matrix

Notes

At least one of x and y must be a sparse matrix. The grad is regular, i.e. not structured.

pytensor.sparse.basic.row_scale(x, s)[source]#

Scale each row of a sparse matrix by the corresponding element of a dense vector.

Parameters:
• x – A sparse matrix.

• s – A dense vector with length equal to the number of rows of x.

Returns:

A sparse matrix in the same format as x whose each row has been multiplied by the corresponding element of s.

Return type:

A sparse matrix

Notes

pytensor.sparse.basic.sp_ones_like(x)[source]#

Construct a sparse matrix of ones with the same sparsity pattern.

Parameters:

x – Sparse matrix to take the sparsity pattern.

Returns:

The same as x with data changed for ones.

Return type:

A sparse matrix

Calculate the sum of a sparse matrix along the specified axis.

It operates a reduction along the specified axis. When axis is None, it is applied along all axes.

Parameters:
• x – Sparse matrix.

• axis – Axis along which the sum is applied. Integer or None.

• sparse_grad (bool) – True to have a structured grad.

Returns:

The sum of x in a dense format.

Return type:

object

Notes

The grad implementation is controlled with the sparse_grad parameter. True will provide a structured grad and False will provide a regular grad. For both choices, the grad returns a sparse matrix having the same format as x.

This op does not return a sparse matrix, but a dense tensor matrix.

pytensor.sparse.basic.sp_zeros_like(x)[source]#

Construct a sparse matrix of zeros.

Parameters:

x – Sparse matrix to take the shape.

Returns:

The same as x with zero entries for all element.

Return type:

A sparse matrix

pytensor.sparse.basic.sparse_formats = ['csc', 'csr'][source]#

Types of sparse matrices to use for testing.

pytensor.sparse.basic.structured_dot(x, y)[source]#

Structured Dot is like dot, except that only the gradient wrt non-zero elements of the sparse matrix a are calculated and propagated.

The output is presumed to be a dense matrix, and is represented by a TensorType instance.

Parameters:
• a – A sparse matrix.

• b – A sparse or dense matrix.

Returns:

The dot product of a and b.

Return type:

A sparse matrix

Notes

pytensor.sparse.basic.sub(x, y)[source]#

Subtract two matrices, at least one of which is sparse.

This method will provide the right op according to the inputs.

Parameters:
• x – A matrix variable.

• y – A matrix variable.

Returns:

x - y

Return type:

A sparse matrix

Notes

At least one of x and y must be a sparse matrix.

The grad will be structured only when one of the variable will be a dense matrix.

Operation for efficiently calculating the dot product when one or all operands are sparse. Supported formats are CSC and CSR. The output of the operation is sparse.

Parameters:
• x – Sparse matrix.

• y – Sparse matrix or 2d tensor variable.

• grad_preserves_dense (bool) – If True (default), makes the grad of dense inputs dense. Otherwise the grad is always sparse.

Returns:

• The dot product x.y in a sparse format.

• Notex

• —–

• The grad implemented is regular, i.e. not structured.

pytensor.sparse.basic.vstack(blocks, format=None, dtype=None)[source]#

Stack sparse matrices vertically (row wise).

This wrap the method vstack from scipy.

Parameters:
• blocks – List of sparse array of compatible shape.

• format – String representing the output format. Default is csc.

• dtype – Output dtype.

Returns:

The concatenation of the sparse array row wise.

Return type:

array

Notes

The number of column of the sparse matrix must agree.

The grad implemented is regular, i.e. not structured.