block_zoo.math package

Submodules

block_zoo.math.Add2D module

class block_zoo.math.Add2D.Add2D(layer_conf)

Bases: torch.nn.modules.module.Module

Add2D layer to get sum of two sequences(2D representation)

Parameters:layer_conf (Add2DConf) – configuration of a layer

forward(*args)

process input

Parameters:*args – (Tensor): string, string_len, string2, string2_len e.g. string (Tensor): [batch_size, dim], string_len (Tensor): [batch_size] Returns:[batch_size, output_dim], [batch_size] Return type:Tensor

class block_zoo.math.Add2D.Add2DConf(**kwargs)

Bases: block_zoo.BaseLayer.BaseConf

Configuration of Add2D layer

declare()

Define things like “input_ranks” and “num_of_inputs”, which are certain with regard to your layer

num_of_input is N(N>0) means this layer accepts N inputs;

num_of_input is -1 means this layer accepts any number of inputs;

The rank here is not the same as matrix rank:

For a scalar, its rank is 0;

For a vector, its rank is 1;

For a matrix, its rank is 2;

For a cube of numbers, its rank is 3.

… For instance, the rank of (batch size, sequence length, hidden_dim) is 3.

if num_of_input > 0:

len(input_ranks) should be equal to num_of_input

elif num_of_input == -1:

input_ranks should be a list with only one element and the rank of all the inputs should be equal to that element.

NOTE: when we build the model, if num_of_input is -1, we would replace it with the real number of inputs and replace input_ranks with a list of real input_ranks.

Returns:None

inference()

Inference things like output_dim, which may relies on defined hyper parameter such as hidden dim and input_dim

Returns:None

verify()

Define some necessary varification for your layer when we define the model.

If you define your own layer and rewrite this funciton, please add “super(YourLayerConf, self).verify()” at the beginning

Returns:None

block_zoo.math.Add3D module

class block_zoo.math.Add3D.Add3D(layer_conf)

Bases: torch.nn.modules.module.Module

Add3D layer to get sum of two sequences(3D representation)

Parameters:layer_conf (Add3DConf) – configuration of a layer

forward(*args)

process input

Parameters:*args – (Tensor): string, string_len, string2, string2_len e.g. string (Tensor): [batch_size, seq_len, dim], string_len (Tensor): [batch_size] Returns:[batch_size, seq_len, output_dim], [batch_size] Return type:Tensor

class block_zoo.math.Add3D.Add3DConf(**kwargs)

Bases: block_zoo.BaseLayer.BaseConf

Configuration of Add3D layer

declare()

Define things like “input_ranks” and “num_of_inputs”, which are certain with regard to your layer

num_of_input is N(N>0) means this layer accepts N inputs;

num_of_input is -1 means this layer accepts any number of inputs;

The rank here is not the same as matrix rank:

For a scalar, its rank is 0;

For a vector, its rank is 1;

For a matrix, its rank is 2;

For a cube of numbers, its rank is 3.

… For instance, the rank of (batch size, sequence length, hidden_dim) is 3.

if num_of_input > 0:

len(input_ranks) should be equal to num_of_input

elif num_of_input == -1:

input_ranks should be a list with only one element and the rank of all the inputs should be equal to that element.

NOTE: when we build the model, if num_of_input is -1, we would replace it with the real number of inputs and replace input_ranks with a list of real input_ranks.

Returns:None

inference()

Inference things like output_dim, which may relies on defined hyper parameter such as hidden dim and input_dim

Returns:None

verify()

Define some necessary varification for your layer when we define the model.

If you define your own layer and rewrite this funciton, please add “super(YourLayerConf, self).verify()” at the beginning

Returns:None

block_zoo.math.ElementWisedMultiply2D module

class block_zoo.math.ElementWisedMultiply2D.ElementWisedMultiply2D(layer_conf)

Bases: torch.nn.modules.module.Module

ElementWisedMultiply2D layer to do Element-Wised Multiply of two sequences(2D representation)

Parameters:layer_conf (ElementWisedMultiply2DConf) – configuration of a layer

forward(*args)

process input

Parameters:*args – (Tensor): string, string_len, string2, string2_len e.g. string (Tensor): [batch_size, dim], string_len (Tensor): [batch_size] Returns:[batch_size, output_dim], [batch_size] Return type:Tensor

class block_zoo.math.ElementWisedMultiply2D.ElementWisedMultiply2DConf(**kwargs)

Bases: block_zoo.BaseLayer.BaseConf

Configuration of ElementWisedMultiply2D layer

declare()

Define things like “input_ranks” and “num_of_inputs”, which are certain with regard to your layer

num_of_input is N(N>0) means this layer accepts N inputs;

num_of_input is -1 means this layer accepts any number of inputs;

The rank here is not the same as matrix rank:

For a scalar, its rank is 0;

For a vector, its rank is 1;

For a matrix, its rank is 2;

For a cube of numbers, its rank is 3.

… For instance, the rank of (batch size, sequence length, hidden_dim) is 3.

if num_of_input > 0:

len(input_ranks) should be equal to num_of_input

elif num_of_input == -1:

input_ranks should be a list with only one element and the rank of all the inputs should be equal to that element.

NOTE: when we build the model, if num_of_input is -1, we would replace it with the real number of inputs and replace input_ranks with a list of real input_ranks.

Returns:None

inference()

Inference things like output_dim, which may relies on defined hyper parameter such as hidden dim and input_dim

Returns:None

verify()

Define some necessary varification for your layer when we define the model.

If you define your own layer and rewrite this funciton, please add “super(YourLayerConf, self).verify()” at the beginning

Returns:None

block_zoo.math.ElementWisedMultiply3D module

class block_zoo.math.ElementWisedMultiply3D.ElementWisedMultiply3D(layer_conf)

Bases: torch.nn.modules.module.Module

ElementWisedMultiply3D layer to do Element-Wised Multiply of two sequences(3D representation)

Parameters:layer_conf (ElementWisedMultiply3DConf) – configuration of a layer

forward(*args)

process input

Parameters:*args – (Tensor): string, string_len, string2, string2_len e.g. string (Tensor): [batch_size, seq_len, dim], string_len (Tensor): [batch_size] Returns:[batch_size, seq_len, output_dim], [batch_size] Return type:Tensor

class block_zoo.math.ElementWisedMultiply3D.ElementWisedMultiply3DConf(**kwargs)

Bases: block_zoo.BaseLayer.BaseConf

Configuration of ElementWisedMultiply3D layer

declare()

Define things like “input_ranks” and “num_of_inputs”, which are certain with regard to your layer

num_of_input is N(N>0) means this layer accepts N inputs;

num_of_input is -1 means this layer accepts any number of inputs;

The rank here is not the same as matrix rank:

For a scalar, its rank is 0;

For a vector, its rank is 1;

For a matrix, its rank is 2;

For a cube of numbers, its rank is 3.

… For instance, the rank of (batch size, sequence length, hidden_dim) is 3.

if num_of_input > 0:

len(input_ranks) should be equal to num_of_input

elif num_of_input == -1:

input_ranks should be a list with only one element and the rank of all the inputs should be equal to that element.

NOTE: when we build the model, if num_of_input is -1, we would replace it with the real number of inputs and replace input_ranks with a list of real input_ranks.

Returns:None

inference()

Inference things like output_dim, which may relies on defined hyper parameter such as hidden dim and input_dim

Returns:None

verify()

Define some necessary varification for your layer when we define the model.

If you define your own layer and rewrite this funciton, please add “super(YourLayerConf, self).verify()” at the beginning

Returns:None

block_zoo.math.MatrixMultiply module

class block_zoo.math.MatrixMultiply.MatrixMultiply(layer_conf)

Bases: torch.nn.modules.module.Module

MatrixMultiply layer to multiply two matrix

Parameters:layer_conf (MatrixMultiplyConf) – configuration of a layer

forward(*args)

process input

Parameters:*args – (Tensor): string, string_len, string2, string2_len e.g. string (Tensor): [batch_size, seq_len, dim], string_len (Tensor): [batch_size] Returns:[batch_size, seq_len, output_dim], [batch_size] Return type:Tensor

class block_zoo.math.MatrixMultiply.MatrixMultiplyConf(**kwargs)

Bases: block_zoo.BaseLayer.BaseConf

Configuration of MatrixMultiply layer

Parameters:operation (String) – a element of [‘common’, ‘seq_based’, ‘dim_based’], default is ‘dim_based’ ‘common’ means (batch_size, seq_len, dim)*(batch_size, seq_len, dim) ‘seq_based’ means (batch_size, dim, seq_len)*(batch_size, seq_len, dim) ‘dim_based’ means (batch_size, seq_len, dim)*(batch_size, dim, seq_len)

declare()

Define things like “input_ranks” and “num_of_inputs”, which are certain with regard to your layer

num_of_input is N(N>0) means this layer accepts N inputs;

num_of_input is -1 means this layer accepts any number of inputs;

The rank here is not the same as matrix rank:

For a scalar, its rank is 0;

For a vector, its rank is 1;

For a matrix, its rank is 2;

For a cube of numbers, its rank is 3.

… For instance, the rank of (batch size, sequence length, hidden_dim) is 3.

if num_of_input > 0:

len(input_ranks) should be equal to num_of_input

elif num_of_input == -1:

input_ranks should be a list with only one element and the rank of all the inputs should be equal to that element.

NOTE: when we build the model, if num_of_input is -1, we would replace it with the real number of inputs and replace input_ranks with a list of real input_ranks.

Returns:None

default()

Define the default hyper parameters here. You can define these hyper parameters in your configuration file as well.

Returns:None

inference()

Inference things like output_dim, which may relies on defined hyper parameter such as hidden dim and input_dim

Returns:None

varify

Docstring inheriting method descriptor The class itself is also used as a decorator doc_inherit decorator

block_zoo.math.Minus2D module

class block_zoo.math.Minus2D.Minus2D(layer_conf)

Bases: torch.nn.modules.module.Module

Minus2D layer to get subtraction of two sequences(2D representation)

Parameters:layer_conf (Minus2DConf) – configuration of a layer

forward(*args)

process inputs

Parameters:*args – (Tensor): string, string_len, string2, string2_len e.g. string (Tensor): [batch_size, dim], string_len (Tensor): [batch_size] Returns:[batch_size, output_dim], [batch_size] Return type:Tensor

class block_zoo.math.Minus2D.Minus2DConf(**kwargs)

Bases: block_zoo.BaseLayer.BaseConf

Configuration of Minus2D layer

Parameters:abs_flag – if the result of the Minus2D is abs, default is False

declare()

Define things like “input_ranks” and “num_of_inputs”, which are certain with regard to your layer

num_of_input is N(N>0) means this layer accepts N inputs;

num_of_input is -1 means this layer accepts any number of inputs;

The rank here is not the same as matrix rank:

For a scalar, its rank is 0;

For a vector, its rank is 1;

For a matrix, its rank is 2;

For a cube of numbers, its rank is 3.

… For instance, the rank of (batch size, sequence length, hidden_dim) is 3.

if num_of_input > 0:

len(input_ranks) should be equal to num_of_input

elif num_of_input == -1:

input_ranks should be a list with only one element and the rank of all the inputs should be equal to that element.

NOTE: when we build the model, if num_of_input is -1, we would replace it with the real number of inputs and replace input_ranks with a list of real input_ranks.

Returns:None

default()

Define the default hyper parameters here. You can define these hyper parameters in your configuration file as well.

Returns:None

inference()

Inference things like output_dim, which may relies on defined hyper parameter such as hidden dim and input_dim

Returns:None

verify()

Define some necessary varification for your layer when we define the model.

If you define your own layer and rewrite this funciton, please add “super(YourLayerConf, self).verify()” at the beginning

Returns:None

block_zoo.math.Minus3D module

class block_zoo.math.Minus3D.Minus3D(layer_conf)

Bases: torch.nn.modules.module.Module

Minus3D layer to get subtraction of two sequences(3D representation)

Parameters:layer_conf (Minus3DConf) – configuration of a layer

forward(*args)

process input

Parameters:*args – (Tensor): string, string_len, string2, string2_len e.g. string (Tensor): [batch_size, seq_len, dim], string_len (Tensor): [batch_size] Returns:[batch_size, seq_len, output_dim], [batch_size] Return type:Tensor

class block_zoo.math.Minus3D.Minus3DConf(**kwargs)

Bases: block_zoo.BaseLayer.BaseConf

Configuration of Minus3D layer

Parameters:abs_flag – if the result of the Minus3D is abs, default is False

declare()

Define things like “input_ranks” and “num_of_inputs”, which are certain with regard to your layer

num_of_input is N(N>0) means this layer accepts N inputs;

num_of_input is -1 means this layer accepts any number of inputs;

The rank here is not the same as matrix rank:

For a scalar, its rank is 0;

For a vector, its rank is 1;

For a matrix, its rank is 2;

For a cube of numbers, its rank is 3.

… For instance, the rank of (batch size, sequence length, hidden_dim) is 3.

if num_of_input > 0:

len(input_ranks) should be equal to num_of_input

elif num_of_input == -1:

input_ranks should be a list with only one element and the rank of all the inputs should be equal to that element.

NOTE: when we build the model, if num_of_input is -1, we would replace it with the real number of inputs and replace input_ranks with a list of real input_ranks.

Returns:None

default()

Define the default hyper parameters here. You can define these hyper parameters in your configuration file as well.

Returns:None

inference()

Inference things like output_dim, which may relies on defined hyper parameter such as hidden dim and input_dim

Returns:None

verify()

Define some necessary varification for your layer when we define the model.

If you define your own layer and rewrite this funciton, please add “super(YourLayerConf, self).verify()” at the beginning

Returns:None

Module contents