您好，想请问一下网络搜索完之后如何得到需要的网络结构代码？

[FelixZhang7]

在网络搜索完后，得到的最高精度网络结构为genotype = Genotype(normal=[('sep_conv_5x5', 1), ('dil_conv_3x3', 0), ('dil_conv_5x5', 2), ('sep_conv_3x3', 1), ('avg_pool_3x3', 3), ('avg_pool_3x3', 1), ('sep_conv_5x5', 1), ('dil_conv_5x5', 4)], normal_concat=range(2, 6), reduce=[('sep_conv_5x5', 1), ('dil_conv_3x3', 0), ('sep_conv_5x5', 2), ('sep_conv_5x5', 1), ('dil_conv_5x5', 2), ('sep_conv_3x3', 3), ('sep_conv_5x5', 4), ('sep_conv_5x5', 3)], reduce_concat=range(2, 6))，如何将这种形式变成可读可加载的网络结构呢？

[LinzheCAI]

在网络搜索完后，得到的最高精度网络结构为genotype = Genotype(normal=[('sep_conv_5x5', 1), ('dil_conv_3x3', 0), ('dil_conv_5x5', 2), ('sep_conv_3x3', 1), ('avg_pool_3x3', 3), ('avg_pool_3x3', 1), ('sep_conv_5x5', 1), ('dil_conv_5x5', 4)], normal_concat=range(2, 6), reduce=[('sep_conv_5x5', 1), ('dil_conv_3x3', 0), ('sep_conv_5x5', 2), ('sep_conv_5x5', 1), ('dil_conv_5x5', 2), ('sep_conv_3x3', 3), ('sep_conv_5x5', 4), ('sep_conv_5x5', 3)], reduce_concat=range(2, 6))，如何将这种形式变成可读可加载的网络结构呢？

问题解决了吗

[myl11045]

config里面是可以配置的，例如 parser.add_argument('--genotype', default="Genotype(normal=[[('max_pool_3x3', 0), ('dil_conv_3x3', 1)], [('max_pool_3x3', 0), ('dil_conv_3x3', 1)], [('max_pool_3x3', 0), ('dil_conv_3x3', 2)], [('dil_conv_3x3', 4), ('max_pool_3x3', 2)]], normal_concat=range(2, 6), reduce=[[('max_pool_3x3', 1), ('sep_conv_3x3', 0)], [('dil_conv_3x3', 2), ('max_pool_3x3', 1)], [('dil_conv_3x3', 3), ('dil_conv_3x3', 2)], [('dil_conv_3x3', 2), ('dil_conv_3x3', 3)]], reduce_concat=range(2, 6))", help='Cell genotype')

[LinzheCAI]

感谢回复，我找到genotype以后，想看看visualization, 但是运行结果如附件所示。

能否告知 visualize.py 里面这三行代码的含义，感谢！

if len(sys.argv) != 2: print("usage:\n python {} ARCH_NAME".format(sys.argv[0])) sys.exit(1)

On Fri, Nov 11, 2022 at 5:36 PM myl11045 @.***> wrote:

config里面是可以配置的，例如 parser.add_argument('--genotype', default="Genotype(normal=[[('max_pool_3x3', 0), ('dil_conv_3x3', 1)], [('max_pool_3x3', 0), ('dil_conv_3x3', 1)], [('max_pool_3x3', 0), ('dil_conv_3x3', 2)], [('dil_conv_3x3', 4), ('max_pool_3x3', 2)]], normal_concat=range(2, 6), reduce=[[('max_pool_3x3', 1), ('sep_conv_3x3', 0)], [('dil_conv_3x3', 2), ('max_pool_3x3', 1)], [('dil_conv_3x3', 3), ('dil_conv_3x3', 2)], [('dil_conv_3x3', 2), ('dil_conv_3x3', 3)]], reduce_concat=range(2, 6))", help='Cell genotype')

— Reply to this email directly, view it on GitHub https://github.com/yuhuixu1993/PC-DARTS/issues/58#issuecomment-1311287729, or unsubscribe https://github.com/notifications/unsubscribe-auth/ATI7NOTNQZFESHTE437IU6LWHXSNXANCNFSM4ZURQGMA . You are receiving this because you commented.Message ID: @.***>