大家好,又见面了,我是你们的朋友全栈君。
Conv2D:图像空间的2维卷积
keras.layers.Conv2D(filters, kernel_size,
strides=(1, 1),
padding='valid',
data_format=None,
dilation_rate=(1, 1),
activation=None,
use_bias=True,
kernel_initializer='glorot_uniform',
bias_initializer='zeros',
kernel_regularizer=None,
bias_regularizer=None,
activity_regularizer=None,
kernel_constraint=None,
bias_constraint=None)
该层创建了一个卷积内核。如果将该图层用作模型中的第一个图层时,需要提供关键参数input_shape
(整数元组),如input_shape=(128,128,3)
对应于128×128 的RGB图片。
“valid”
或“same”
channels_last
或channels_first
,前者对应的输入shape是(batch, height, width, channels)
,后者对应的shape是(batch, channels, height, width)
。默认的是“channels_last”kernel
权重矩阵的初始化器kernel
权重矩阵的正则化函数Input shape
4D tensor with shape: (batch, channels, rows, cols) if data_format is “channels_first” or 4D tensor with shape: (batch, rows, cols, channels) if data_format is “channels_last”.
Output shape
4D tensor with shape: (batch, filters, new_rows, new_cols) if data_format is “channels_first” or 4D tensor with shape: (batch, new_rows, new_cols, filters) if data_format is “channels_last”. rows and cols values might have changed due to padding.
详细内容请查看 原文档
发布者:全栈程序员栈长,转载请注明出处:https://javaforall.cn/128883.html原文链接:https://javaforall.cn