帮我搭建一个resnet50_v1c网络模型

好的，以下是使用MXNet搭建resnet50_v1c网络模型的示例代码：

import mxnet as mx

def conv_block(data, num_filter, name, stride=(1,1), pad=(0,0)):
    conv = mx.sym.Convolution(data=data, num_filter=num_filter, kernel=(3,3), stride=stride, pad=pad, no_bias=True, name=name+'_conv')
    bn = mx.sym.BatchNorm(data=conv, fix_gamma=False, eps=2e-5, momentum=0.9, name=name+'_bn')
    act = mx.sym.Activation(data=bn, act_type='relu', name=name+'_relu')
    return act

def identity_block(data, num_filter, name, stride=(1,1), pad=(0,0)):
    conv1 = conv_block(data=data, num_filter=num_filter, name=name+'_conv1', stride=stride, pad=pad)
    conv2 = conv_block(data=conv1, num_filter=num_filter, name=name+'_conv2', stride=(1,1), pad=pad)
    sum = mx.sym.ElementWiseSum(data, conv2, name=name+'_plus')
    act = mx.sym.Activation(data=sum, act_type='relu', name=name+'_relu')
    return act

def resnet50_v1c():
    data = mx.sym.Variable(name='data')
    conv1 = conv_block(data=data, num_filter=64, name='conv1', stride=(2,2), pad=(1,1))
    pool1 = mx.sym.Pooling(data=conv1, kernel=(3,3), stride=(2,2), pad=(1,1), pool_type='max', name='pool1')
    # stage 2
    block1 = identity_block(data=pool1, num_filter=64, name='stage2_block1', stride=(1,1), pad=(1,1))
    block2 = identity_block(data=block1, num_filter=64, name='stage2_block2', stride=(1,1), pad=(1,1))
    block3 = identity_block(data=block2, num_filter=64, name='stage2_block3', stride=(1,1), pad=(1,1))
    # stage 3
    block4 = identity_block(data=block3, num_filter=128, name='stage3_block1', stride=(2,2), pad=(1,1))
    block5 = identity_block(data=block4, num_filter=128, name='stage3_block2', stride=(1,1), pad=(1,1))
    block6 = identity_block(data=block5, num_filter=128, name='stage3_block3', stride=(1,1), pad=(1,1))
    block7 = identity_block(data=block6, num_filter=128, name='stage3_block4', stride=(1,1), pad=(1,1))
    # stage 4
    block8 = identity_block(data=block7, num_filter=256, name='stage4_block1', stride=(2,2), pad=(1,1))
    block9 = identity_block(data=block8, num_filter=256, name='stage4_block2', stride=(1,1), pad=(1,1))
    block10 = identity_block(data=block9, num_filter=256, name='stage4_block3', stride=(1,1), pad=(1,1))
    block11 = identity_block(data=block10, num_filter=256, name='stage4_block4', stride=(1,1), pad=(1,1))
    block12 = identity_block(data=block11, num_filter=256, name='stage4_block5', stride=(1,1), pad=(1,1))
    block13 = identity_block(data=block12, num_filter=256, name='stage4_block6', stride=(1,1), pad=(1,1))
    # stage 5
    block14 = identity_block(data=block13, num_filter=512, name='stage5_block1', stride=(2,2), pad=(1,1))
    block15 = identity_block(data=block14, num_filter=512, name='stage5_block2', stride=(1,1), pad=(1,1))
    block16 = identity_block(data=block15, num_filter=512, name='stage5_block3', stride=(1,1), pad=(1,1))
    # global average pooling
    pool5 = mx.sym.Pooling(data=block16, global_pool=True, kernel=(7,7), pool_type='avg', name='pool5')
    fc = mx.sym.FullyConnected(data=pool5, num_hidden=1000, name='fc')
    softmax = mx.sym.SoftmaxOutput(data=fc, name='softmax')
    return softmax

这个代码定义了一个resnet50_v1c网络模型，包括卷积块、残差块和全局平均池化层。你可以根据自己的需要修改模型结构中的参数。