PyLayerContext¶

class paddle.autograd. PyLayerContext [source]

The object of this class is a context that is used in PyLayer to enhance the function.

Examples

import paddle
from paddle.autograd import PyLayer

class cus_tanh(PyLayer):
    @staticmethod
    def forward(ctx, x):
        # ctx is a object of PyLayerContext.
        y = paddle.tanh(x)
        ctx.save_for_backward(y)
        return y

    @staticmethod
    def backward(ctx, dy):
        # ctx is a object of PyLayerContext.
        y, = ctx.saved_tensor()
        grad = dy * (1 - paddle.square(y))
        return grad

save_for_backward ( *tensors )

save_for_backward¶

Saves given tensors that backward need. Use saved_tensor in the backward to get the saved tensors.

Note

This API should be called at most once, and only inside forward.

Parameters

tensors (list of Tensors) – Tensors to be stored.

Returns

None

Examples

import paddle
from paddle.autograd import PyLayer

class cus_tanh(PyLayer):
    @staticmethod
    def forward(ctx, x):
        # ctx is a context object that store some objects for backward.
        y = paddle.tanh(x)
        # Pass tensors to backward.
        ctx.save_for_backward(y)
        return y

    @staticmethod
    def backward(ctx, dy):
        # Get the tensors passed by forward.
        y, = ctx.saved_tensor()
        grad = dy * (1 - paddle.square(y))
        return grad

saved_tensor ( )

saved_tensor¶

Get the tensors stored by save_for_backward.

Returns

If context contains tensors stored by save_for_backward, then return these tensors, otherwise return None.

Return type

list of Tensors or None

Examples

import paddle
from paddle.autograd import PyLayer

class cus_tanh(PyLayer):
    @staticmethod
    def forward(ctx, x):
        # ctx is a context object that store some objects for backward.
        y = paddle.tanh(x)
        # Pass tensors to backward.
        ctx.save_for_backward(y)
        return y

    @staticmethod
    def backward(ctx, dy):
        # Get the tensors passed by forward.
        y, = ctx.saved_tensor()
        grad = dy * (1 - paddle.square(y))
        return grad