zhangximing closed issue OpenModelZoo/feedback_jijiehao#2
算子nn.Conv2d中group参数与dilation参数不兼容问题2 years ago
zhangximing commented on issue OpenModelZoo/feedback_jijiehao#2
算子nn.Conv2d中group参数与dilation参数不兼容问题mindspore r1.5版本已修复此bug,issue关闭
2 years ago
zhangximing pushed to master at OpenModelZoo/AutoDeeplab
2 years ago
zhangximing pushed to master at OpenModelZoo/AutoDeeplab
2 years ago
zhangximing deleted branch dev from OpenModelZoo/AutoDeeplab
2 years ago
zhangximing pushed to master at OpenModelZoo/AutoDeeplab
2 years ago
zhangximing pushed to master at OpenModelZoo/AutoDeeplab
2 years ago
zhangximing commented on issue OpenModelZoo/feedback_jijiehao#2
算子nn.Conv2d中group参数与dilation参数不兼容问题问题尚未解决,bug尚未修复。目前使用替代方案来取代原来的nn.Conv2d,但是时间复杂度有点高。 ``` class DilSepCov(nn.Cell): def __init__(self, channels, kernel_size, stride, pad_mode='pad', padding=0, dilation=1, has_bias=False, weight_init='HeNormal' ): super(DilSepCov, self).__init__() self.channels = channels self.ops = nn.CellList([]) for i in range(channels): self.ops.append(nn.Conv2d(1, 1, kernel_size=kernel_size, stride=stride, pad_mode=pad_mode, padding=padding, dilation=dilation, has_bias=has_bias, weight_init=weight_init)) self.concat = ops.Concat(axis=1) def construct(self, x): output = () for i in range(1, self.channels): output += (self.ops[i](x[:, i:i+1:1, :, :]), ) output = self.concat(output) return output ```
2 years ago
zhangximing pushed to dev at OpenModelZoo/AutoDeeplab
2 years ago
zhangximing opened issue OpenModelZoo/feedback_jijiehao#2
算子nn.Conv2d中group参数与dilation参数不兼容问题3 years ago
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》