提交 f14dc05d authored 作者: James Bergstra's avatar James Bergstra

removed perform from nll grad to work better with debugmode

上级 23535836
......@@ -186,9 +186,6 @@ class GpuCrossentropySoftmax1HotWithBiasDx (Op):
return self.__class__.__name__
def make_node(self, dy, sm, y_idx):
return Apply(self, [dy, sm, y_idx],[sm.type()])
def perform(self, node, input_storage, output_storage):
raise NotImplementedError('only C is implemented')
def c_code_cache_version(self):
return ()
def c_code(self, node, nodename, (dnll, sm, y_idx), (dx,), sub):
......
Markdown 格式
0%
您添加了 0 到此讨论。请谨慎行事。
请先完成此评论的编辑!
注册 或者 后发表评论