From b3db8ef871d3e2cdf13b2583de44866dc6e1b5b8 Mon Sep 17 00:00:00 2001 From: peterzhang2029 Date: Tue, 9 Jan 2018 11:34:27 +0800 Subject: [PATCH] refine the description --- python/paddle/trainer_config_helpers/layers.py | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/python/paddle/trainer_config_helpers/layers.py b/python/paddle/trainer_config_helpers/layers.py index df4a630077b..eac2cb31683 100644 --- a/python/paddle/trainer_config_helpers/layers.py +++ b/python/paddle/trainer_config_helpers/layers.py @@ -2545,12 +2545,18 @@ def img_conv_layer(input, num_filters. There are several groups of filters in PaddlePaddle implementation. - Each group will process some channels of the input. For example, if - num_channel = 256, group = 4, num_filter=32, the PaddlePaddle will create - 32 filters to process the input. The input channels will be split into 4 - pieces. First 256/4 = 64 channels will be processed by first 32/4 = 8 filters. - The rest channels will be processed by the rest groups of filters. - + If the groups attribute is greater than 1, for example groups=2, + the input will be splitted into 2 parts along the channel axis, and + the filters will also be splitted into 2 parts. The first half of the filters + is only connected to the first half of the input channels, while the second + half of the filters is only connected to the second half of the input. After + the computation of convolution for each part of input, + the output will be obtained by concatenating the two results. + + The details of grouped convolution, please refer to: + `ImageNet Classification with Deep Convolutional Neural Networks + `_ + The example usage is: .. code-block:: python -- GitLab