dragon.nn¶
Classes¶
class GRU : Apply a multi-layer gated recurrent unit (GRU) RNN. [Cho et.al, 2014].
class LSTM : Apply a multi-layer long short-term memory (LSTM) RNN. [Hochreiter & Schmidhuber, 1997].
class RNN : Apply a multi-layer Elman RNN. [Elman, 1990].
Functions¶
batch_norm(…) : Apply the batch normalization. [Ioffe & Szegedy, 2015].
bias_add(…) : Add the bias across channels to input.
channel_norm(…) : Apply the normalization to each channel of input.
channel_shuffle(…) : Apply the group shuffle to each channel of input. [Zhang et.al, 2017].
conv(…) : Apply the n-dimension convolution.
conv_transpose(…) : Apply the n-dimension deconvolution.
conv1d(…) : Apply the 1d convolution.
conv1d_transpose(…) : Apply the 1d deconvolution.
conv2d(…) : Apply the 2d convolution.
conv2d_transpose(…) : Apply the 2d deconvolution.
conv3d(…) : Apply the 3d convolution.
conv3d_transpose(…) : Apply the 3d deconvolution.
depthwise_conv2d(…) : Apply the 2d depthwise convolution.
depth_to_space(…) : Rearrange depth data into spatial blocks.
dropout(…) : Set the elements of input to zero randomly. [Srivastava et.al, 2014].
drop_block(…) : Set the blocks over input to zero randomly. [Ghiasi et.al, 2018].
drop_path(…) : Set the examples over input to zero randomly. [Larsson et.al, 2016].
elu(…) : Apply the exponential linear unit. [Clevert et.al, 2015].
gelu(…) : Apply the gaussian error linear unit. [Hendrycks & Gimpel, 2016].
group_norm(…) : Apply the group normalization. [Wu & He, 2018].
hardsigmoid(…) : Apply the hard sigmoid function.
hardswish(…) : Apply the hard swish function. [Howard et.al, 2019].
instance_norm(…) : Apply the instance normalization. [Ulyanov et.al, 2016]
layer_norm(…) : Apply the layer normalization. [Ba et.al, 2016]
leaky_relu(…) : Apply the leaky rectified linear unit.
local_response_norm(…) : Apply the local response normalization. [Krizhevsky et.al, 2012].
log_softmax(…) : Compute the composite of logarithm and softmax.
lp_norm(…) : Apply the lp normalization.
moments(…) : Compute the mean and variance of input along the given axis.
prelu(…) : Apply the parametric rectified linear unit. [He et.al, 2015].
pool1d(…) : Apply the 1d pooling.
pool2d(…) : Apply the 2d pooling.
pool3d(…) : Apply the 3d pooling.
relu(…) : Apply the rectified linear unit. [Nair & Hinton, 2010].
relu6(…) : Apply the clipped-6 rectified linear unit. [Krizhevsky, 2010].
selu(…) : Apply the scaled exponential linear unit. [Klambauer et.al, 2017].
silu(…) : Apply the sigmoid linear unit. [Hendrycks & Gimpel, 2016].
softmax(…) : Compute the softmax result.
space_to_depth(…) : Rearrange blocks of spatial data into depth.
sync_batch_norm(…) : Apply the batch normalization with synced statistics. [Ioffe & Szegedy, 2015].