conv1d¶

dragon.vm.tensorflow.nn.conv1d(
input,
filters,
strides=1,
data_format='NHWC',
dilations=None,
name=None
)[source]

Apply the 1d convolution.

• If data_format is 'NCHW', excepts input shape $$(N, C_{\text{in}}, H)$$, filters shape $$(C_{\text{out}}, C_{\text{in}}, H_{\text{f}})$$, and output shape is $$(N, C_{\text{out}}, H_{\text{out}})$$.
• If data_format is 'NHWC', excepts input shape $$(N, H, C_{\text{in}})$$, filters shape $$(C_{\text{out}}, H_{\text{f}}, C_{\text{in}})$$, and output shape is $$(N, H_{\text{out}}, C_{\text{out}})$$.
• padding could be 'VALID', 'SAME' or explicit padding size.

Examples:

x = tf.ones((1, 2, 2))
filters = tf.ones((3, 1, 2))
y = tf.nn.conv1d(x, filters)
assert y.shape == (1, 2, 3)

Parameters:
• input (dragon.Tensor) – The input tensor.
• filters (dragon.Tensor) – The filters tensor.
• strides (Union[int, Sequence[int]], optional, default=1) – The stride of convolution window.
• padding (Union[int, Sequence[int], str], optional, default='VALID') – The padding algorithm or size.
• data_format (str, optional, default='NHWC') – 'NCHW' or 'NHWC'.
• dilations (Union[int, Sequence[int]], optional, default=1) – The rate of dilated filters.
• name (str, optional) – The operation name.
Returns:

dragon.Tensor – The output tensor.