mindspore.nn.Dropout

class mindspore.nn.Dropout(keep_prob=0.5, dtype=mstype.float32)[source]

Dropout layer for the input.

Randomly set some elements of the input tensor to zero with probability \(1 - keep\_prob\) during training using samples from a Bernoulli distribution.

The outputs are scaled by a factor of \(\frac{1}{keep\_prob}\) during training so that the output layer remains at a similar scale. During inference, this layer returns the same tensor as the input.

This technique is proposed in paper Dropout: A Simple Way to Prevent Neural Networks from Overfitting and proved to be effective to reduce over-fitting and prevents neurons from co-adaptation. See more details in Improving neural networks by preventing co-adaptation of feature detectors.

Note

Each channel will be zeroed out independently on every construct call.

Parameters

  • keep_prob (float) – The keep rate, greater than 0 and less equal than 1. E.g. rate=0.9, dropping out 10% of input units. Default: 0.5.

  • dtype (mindspore.dtype) – Data type of input. Default: mindspore.float32.

Inputs:

  • input (Tensor) - The input of Dropout with data type of float16 or float32.

Outputs:

Tensor, output tensor with the same shape as the input.

Raises

  • TypeError – If keep_prob is not a float.

  • TypeError – If dtype of input is not neither float16 nor float32.

  • ValueError – If keep_prob is not in range (0, 1].

  • ValueError – If length of shape of input is less than 1.

Supported Platforms:

Ascend GPU CPU

Examples

>>> x = Tensor(np.ones([2, 2, 3]), mindspore.float32)
>>> net = nn.Dropout(keep_prob=0.8)
>>> net.set_train()
Dropout<keep_prob=0.8>
>>> output = net(x)
>>> print(output.shape)
(2, 2, 3)