Skip to contents

1D Convolution

Source: R/api.R
nv_conv1d.Rd

Torch-style 1D convolution in NCW layout: x is [batch, in_channels, width], weight is [out_channels, in_channels / groups, kW], output is [batch, out_channels, out_w]. Symmetric zero padding.

Usage

nv_conv1d(
  x,
  weight,
  stride = 1L,
  padding = 0L,
  dilation = 1L,
  groups = 1L,
  precision = "highest"
)

Arguments

x

(arrayish)
[N, C_in, W].

weight

(arrayish)
[C_out, C_in / groups, kW].

stride, padding, dilation

(integer())
Length 1.

groups

(integer(1))
Grouped/depthwise convolution.

precision

(character(1))
"highest", "high" or "default".

Value

arrayish [N, C_out, out_W].

See also

nv_conv2d(), nv_conv3d(), prim_convolution().