當前位置: 首頁>>代碼示例>>Python>>正文


Python blocks.CompositionOperator方法代碼示例

本文整理匯總了Python中entropy_coder.lib.blocks.CompositionOperator方法的典型用法代碼示例。如果您正苦於以下問題:Python blocks.CompositionOperator方法的具體用法?Python blocks.CompositionOperator怎麽用?Python blocks.CompositionOperator使用的例子?那麽, 這裏精選的方法代碼示例或許可以為您提供幫助。您也可以進一步了解該方法所在entropy_coder.lib.blocks的用法示例。


在下文中一共展示了blocks.CompositionOperator方法的1個代碼示例,這些例子默認根據受歡迎程度排序。您可以為喜歡或者感覺有用的代碼點讚,您的評價將有助於係統推薦出更棒的Python代碼示例。

示例1: __init__

# 需要導入模塊: from entropy_coder.lib import blocks [as 別名]
# 或者: from entropy_coder.lib.blocks import CompositionOperator [as 別名]
def __init__(self, code_depth, name=None):
    super(BrnnPredictor, self).__init__(name)

    with self._BlockScope():
      hidden_depth = 2 * code_depth

      # What is coming from the previous layer/iteration
      # is going through a regular Conv2D layer as opposed to the binary codes
      # of the current layer/iteration which are going through a masked
      # convolution.
      self._adaptation0 = blocks.RasterScanConv2D(
          hidden_depth, [7, 7], [1, 1], 'SAME',
          strict_order=True,
          bias=blocks.Bias(0), act=tf.tanh)
      self._adaptation1 = blocks.Conv2D(
          hidden_depth, [3, 3], [1, 1], 'SAME',
          bias=blocks.Bias(0), act=tf.tanh)
      self._predictor = blocks.CompositionOperator([
          blocks.LineOperator(
              blocks.RasterScanConv2DLSTM(
                  depth=hidden_depth,
                  filter_size=[1, 3],
                  hidden_filter_size=[1, 3],
                  strides=[1, 1],
                  padding='SAME')),
          blocks.Conv2D(hidden_depth, [1, 1], [1, 1], 'SAME',
                        bias=blocks.Bias(0), act=tf.tanh),
          blocks.Conv2D(code_depth, [1, 1], [1, 1], 'SAME',
                        bias=blocks.Bias(0), act=tf.tanh)
      ]) 
開發者ID:ringringyi,項目名稱:DOTA_models,代碼行數:32,代碼來源:progressive.py


注:本文中的entropy_coder.lib.blocks.CompositionOperator方法示例由純淨天空整理自Github/MSDocs等開源代碼及文檔管理平台,相關代碼片段篩選自各路編程大神貢獻的開源項目,源碼版權歸原作者所有,傳播和使用請參考對應項目的License;未經允許,請勿轉載。