Tensor¶

class dragon.Tensor(
shape=None,
dtype='float32',
name=None,
symbolic=False,
**kwargs
)[source]¶: A multi-dimensional array for computation.

init¶

Tensor.__init__(
shape=None,
dtype='float32',
name=None,
symbolic=False,
**kwargs
)[source]¶

Create a Tensor.

Parameters:

shape (Sequence[int], optional) – The tensor shape.
dtype (str, optional, default='float32') – The data type.
name (str, optional) – The tensor name.
symbolic (bool, optional, default=False) – Whether to initialize as a symbolic tensor.

Properties¶

device¶

Tensor.device¶

Return the tensor device.

Returns:: dragon.DeviceSpec – The device.

dtype¶

Tensor.dtype¶

Return the data type.

Returns:: str – The data type.

id¶

Tensor.id¶

Return the tensor identity.

Returns:: str – The tensor identity.

name¶

Tensor.name¶

Return the tensor name.

Returns:: str – The tensor name.

ndim¶

Tensor.ndim¶

Return the number of dimensions.

Returns:: int – The number of dimensions.

requires_grad¶

Tensor.requires_grad¶

Return whether the gradient will be recorded.

Returns:: bool – True to record gradient otherwise False.

shape¶

Tensor.shape¶

Return the tensor shape.

Returns:: Tuple[int] – The tensor shape.

size¶

Tensor.size¶

Return the total number of elements in this tensor.

Returns:: int – The total count of elements.

T¶

Tensor.T¶

Return a tensor with axes reversed.

Returns:: dragon.Tensor – The output tensor.

Methods¶

astype¶

Tensor.astype( dtype, copy=True )[source]¶

Convert the data type to a specific one.

Parameters:

dtype (str) – The specific data type.
copy (bool, optional, default=True) – Return a new tensor or converted in-place.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.cast(…)

copy¶

Tensor.copy()[source]¶

Return a tensor with containing data copied.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.identity(…)

fill¶

Tensor.fill(value)[source]¶

Fill self with a scalar value.

\text{self} \leftarrow \text{value}

Parameters:

value (number) – The value to fill.

Returns:: dragon.Tensor – The self.

See also

dragon.fill(…)

glorot_normal¶

Tensor.glorot_normal( mode='fan_in', scale=2.0 )[source]¶

Fill self from a glorot normal distribution.

\text{self} \sim \mathcal{N}(0, \frac{scale}{\text{fan}})

Parameters:

mode ({'fan_in, 'fan_out', 'fan_avg'}, optional) – The mode to compute fans.
scale (float, optional, default=2.0) – The scale factor of distribution.

Returns:: dragon.Tensor – The self.

See also

dragon.random.glorot_normal(…)

glorot_uniform¶

Tensor.glorot_uniform( mode='fan_in', scale=3.0 )[source]¶

Fill self from a glorot uniform distribution.

\text{self} \sim \mathcal{U}(-\sqrt{\frac{scale}{\text{fan}}}, \sqrt{\frac{scale}{\text{fan}}})

Parameters:

mode ({'fan_in, 'fan_out', 'fan_avg'}, optional) – The mode to compute fans.
scale (float, optional, default=3.0) – The scale factor of distribution.

Returns:: dragon.Tensor – The self.

See also

dragon.random.glorot_uniform(…)

item¶

Tensor.item()[source]¶

Return the value as a python number.

Returns:: number – The value.

normal¶

Tensor.normal( mean=0, std=1 )[source]¶

Fill self from a normal distribution.

\text{self} \sim \mathcal{N}(\mu, \sigma^{2})

Parameters:

mean (number, optional, default=0) – The value to $\mu$ .
std (number, optional, default=1) – The value to $\sigma$ .

Returns:: dragon.Tensor – The self.

See also

dragon.random.normal(…)

numpy¶

Tensor.numpy(copy=False)[source]¶

Convert tensor into a numpy array.

Parameters:

copy (bool, optional, default=False) – Whether to copy the data.

Returns:: numpy.ndarray – The value array.

reshape¶

Tensor.reshape( shape, copy=True )[source]¶

Return a tensor containing the same data with new shape.

Parameters:

shape (Union[Sequence[int], dragon.Tensor]) – The output shape.
copy (bool, optional, default=True) – Return a new tensor or reshape in-place.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.reshape(…)

tolist¶

Tensor.tolist()[source]¶

Return the value as a python list.

Returns:: list – The value.

transpose¶

Tensor.transpose( perm=None, copy=True )[source]¶

Return a tensor with permuted axes.

Parameters:

perm (Union[Sequence[int], dragon.Tensor]], optional) – The output permutation.
copy (bool, optional, default=True) – Return a new tensor or transpose in-place.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.transpose(…)

truncated_normal¶

Tensor.truncated_normal( mean=0, std=1 )[source]¶

Fill self from a truncated normal distribution.

\text{self} \sim \mathcal{TN}(\mu, \sigma^{2}, \mu - 2\sigma, \mu + 2\sigma)

Parameters:

mean (number, optional, default=0) – The value to $\mu$ .
std (number, optional, default=1) – The value to $\sigma$ .

Returns:: dragon.Tensor – The self.

See also

dragon.random.truncated_normal(…)

uniform¶

Tensor.uniform( low=0, high=1 )[source]¶

Fill self from a uniform distribution.

\text{self} \sim \mathcal{U}(\alpha, \beta)

Parameters:

low (number, optional, default=0) – The value to $\alpha$ .
high (number, optional, default=1) – The value to $\beta$ .

Returns:: dragon.Tensor – The self.

See also

dragon.random.uniform(…)

Overrides¶

add¶

Tensor.__add__(other)[source]¶

Compute the element-wise addition.

Parameters:

other (Union[dragon.Tensor, number]) – The value to add.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.add(…)

and¶

Tensor.__and__(other)[source]¶

Compute the element-wise AND bitwise operation.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compute with.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.bitwise.bitwise_and(…)

eq¶

Tensor.__eq__(other)[source]¶

Compute element-wise equal comparison.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compare.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.equal(…)

float¶

Tensor.__float__()[source]¶

Return the value as a python number.

Returns:: float – The float value.

ge¶

Tensor.__ge__(other)[source]¶

Compute element-wise greater-equal comparison.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compare.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.greater_equal(…)

getitem¶

Tensor.__getitem__(item)[source]¶

Select elements at the specific index.

Parameters:

item (Union[slice, int, dragon.Tensor]) – The index.

Returns:: dragon.Tensor – The output tensor.

gt¶

Tensor.__gt__(other)[source]¶

Compute element-wise greater comparison.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compare.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.greater(…)

iadd¶

Tensor.__iadd__(other)[source]¶

Compute the element-wise addition.

Parameters:

other (Union[dragon.Tensor, number]) – The value to add.

Returns:: dragon.Tensor – The self.

See also

dragon.math.add(…)

iand¶

Tensor.__iand__(other)[source]¶

Compute the element-wise AND bitwise operation.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compute with.

Returns:: dragon.Tensor – The self.

See also

dragon.bitwise.bitwise_and(…)

imul¶

Tensor.__imul__(other)[source]¶

Compute the element-wise multiplication.

Parameters:

other (Union[dragon.Tensor, number]) – The value to multiply.

Returns:: dragon.Tensor – The self.

See also

dragon.math.mul(…)

int¶

Tensor.__int__()[source]¶

Return the value as a python number.

Returns:: int – The integer value.

invert¶

Tensor.__invert__()[source]¶

Compute the element-wise NOT bitwise operation.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.bitwise.invert(…)

ior¶

Tensor.__ior__(other)[source]¶

Compute the element-wise OR bitwise operation.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compute with.

Returns:: dragon.Tensor – The self.

See also

dragon.bitwise.bitwise_or(…)

isub¶

Tensor.__isub__(other)[source]¶

Compute the element-wise subtraction.

Parameters:

other (Union[dragon.Tensor, number]) – The value to subtract.

Returns:: dragon.Tensor – The self.

See also

dragon.math.sub(…)

itruediv¶

Tensor.__itruediv__(other)[source]¶

Compute the element-wise division.

Parameters:

other (Union[dragon.Tensor, number]) – The value to divide.

Returns:: dragon.Tensor – The self.

See also

dragon.math.div(…)

ixor¶

Tensor.__ixor__(other)[source]¶

Compute the element-wise XOR bitwise operation.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compute with.

Returns:: dragon.Tensor – The self.

See also

dragon.bitwise.bitwise_xor(…)

le¶

Tensor.__le__(other)[source]¶

Compute element-wise less-equal comparison.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compare.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.less_equal(…)

lt¶

Tensor.__lt__(other)[source]¶

Compute element-wise less comparison.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compare.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.less(…)

matmul¶

Tensor.__matmul__(other)[source]¶

Compute the matrix multiplication.

Parameters:

other (dragon.Tensor) – The value to multiply.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.matmul(…)

mul¶

Tensor.__mul__(other)[source]¶

Compute the element-wise multiplication.

Parameters:

other (Union[dragon.Tensor, number]) – The value to multiply.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.mul(…)

ne¶

Tensor.__ne__(other)[source]¶

Compute element-wise not-equal comparison.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compare.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.not_equal(…)

neg¶

Tensor.__neg__()[source]¶

Compute the element-wise negative.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.negative(…)

or¶

Tensor.__or__(other)[source]¶

Compute the element-wise OR bitwise operation.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compute with.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.bitwise.bitwise_or(…)

radd¶

Tensor.__radd__(other)[source]¶

Compute the element-wise addition.

Parameters:

other (Union[dragon.Tensor, number]) – The value to add.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.add(…)

rand¶

Tensor.__rand__(other)[source]¶

Compute the element-wise AND bitwise operation.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compute with.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.bitwise.bitwise_and(…)

rmul¶

Tensor.__rmul__(other)[source]¶

Compute the element-wise multiplication.

Parameters:

other (Union[dragon.Tensor, number]) – The value to multiply.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.mul(…)

ror¶

Tensor.__ror__(other)[source]¶

Compute the element-wise OR bitwise operation.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compute with.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.bitwise.bitwise_or(…)

rsub¶

Tensor.__rsub__(other)[source]¶

Compute the element-wise subtraction.

Parameters:

other (Union[dragon.Tensor, number]) – The value to be subtracted.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.sub(…)

rxor¶

Tensor.__rxor__(other)[source]¶

Compute the element-wise XOR bitwise operation.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compute with.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.bitwise.bitwise_xor(…)

setitem¶

Tensor.__setitem__( key, value )[source]¶

Set elements at the specific index.

Parameters:

key (Union[slice, int, dragon.Tensor]) – The index.
value (Union[dragon.Tensor, number]) – The value to set.

sub¶

Tensor.__sub__(other)[source]¶

Compute the element-wise subtraction.

Parameters:

other (Union[dragon.Tensor, number]) – The value to subtract.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.sub(…)

rtruediv¶

Tensor.__rtruediv__(other)[source]¶

Compute the element-wise division.

Parameters:

other (Union[dragon.Tensor, number]) – The value to be divided.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.div(…)

truediv¶

Tensor.__truediv__(other)[source]¶

Compute the element-wise division.

Parameters:

other (Union[dragon.Tensor, number]) – The value to divide.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.math.div(…)

xor¶

Tensor.__xor__(other)[source]¶

Compute the element-wise XOR bitwise operation.

Parameters:

other (Union[dragon.Tensor, number]) – The value to compute with.

Returns:: dragon.Tensor – The output tensor.

See also

dragon.bitwise.bitwise_xor(…)

Tensor¶

__init__¶

Properties¶

device¶

dtype¶

id¶

name¶

ndim¶

requires_grad¶

shape¶

size¶

T¶

Methods¶

astype¶

copy¶

fill¶

glorot_normal¶

glorot_uniform¶

item¶

normal¶

numpy¶

reshape¶

tolist¶

transpose¶

truncated_normal¶

uniform¶

Overrides¶

__add__¶

__and__¶

__eq__¶

__float__¶

__ge__¶

__getitem__¶

__gt__¶

__iadd__¶

__iand__¶

__imul__¶

__int__¶

__invert__¶

__ior__¶

__isub__¶

__itruediv__¶

__ixor__¶

__le__¶

__lt__¶

__matmul__¶

__mul__¶

__ne__¶

__neg__¶

__or__¶

__radd__¶

__rand__¶

__rmul__¶

__ror__¶

__rsub__¶

__rxor__¶

__setitem__¶

__sub__¶

__rtruediv__¶

__truediv__¶

__xor__¶

init¶

add¶

and¶

eq¶

float¶

ge¶

getitem¶

gt¶

iadd¶

iand¶

imul¶

int¶

invert¶

ior¶

isub¶

itruediv¶

ixor¶

le¶

lt¶

matmul¶

mul¶

ne¶

neg¶

or¶

radd¶

rand¶

rmul¶

ror¶

rsub¶

rxor¶

setitem¶

sub¶

rtruediv¶

truediv¶

xor¶