How to use the mxnet.symbol.broadcast_add function in mxnet

scale3b1_branch2a = bn3b1_branch2a
        res3b1_branch2a_relu = mx.symbol.Activation(name='res3b1_branch2a_relu', data=scale3b1_branch2a,
                                                    act_type='relu')
        res3b1_branch2b = mx.symbol.Convolution(name='res3b1_branch2b', data=res3b1_branch2a_relu, num_filter=128,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn3b1_branch2b = mx.symbol.BatchNorm(name='bn3b1_branch2b', data=res3b1_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale3b1_branch2b = bn3b1_branch2b
        res3b1_branch2b_relu = mx.symbol.Activation(name='res3b1_branch2b_relu', data=scale3b1_branch2b,
                                                    act_type='relu')
        res3b1_branch2c = mx.symbol.Convolution(name='res3b1_branch2c', data=res3b1_branch2b_relu, num_filter=512,
                                                pad=(0, 0), kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn3b1_branch2c = mx.symbol.BatchNorm(name='bn3b1_branch2c', data=res3b1_branch2c, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale3b1_branch2c = bn3b1_branch2c
        res3b1 = mx.symbol.broadcast_add(name='res3b1', *[res3a_relu, scale3b1_branch2c])
        res3b1_relu = mx.symbol.Activation(name='res3b1_relu', data=res3b1, act_type='relu')
        res3b2_branch2a = mx.symbol.Convolution(name='res3b2_branch2a', data=res3b1_relu, num_filter=128, pad=(0, 0),
                                                kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn3b2_branch2a = mx.symbol.BatchNorm(name='bn3b2_branch2a', data=res3b2_branch2a, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale3b2_branch2a = bn3b2_branch2a
        res3b2_branch2a_relu = mx.symbol.Activation(name='res3b2_branch2a_relu', data=scale3b2_branch2a,
                                                    act_type='relu')
        res3b2_branch2b = mx.symbol.Convolution(name='res3b2_branch2b', data=res3b2_branch2a_relu, num_filter=128,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn3b2_branch2b = mx.symbol.BatchNorm(name='bn3b2_branch2b', data=res3b2_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale3b2_branch2b = bn3b2_branch2b
        res3b2_branch2b_relu = mx.symbol.Activation(name='res3b2_branch2b_relu', data=scale3b2_branch2b,
                                                    act_type='relu')
        res3b2_branch2c = mx.symbol.Convolution(name='res3b2_branch2c', data=res3b2_branch2b_relu, num_filter=512,

scale4b11_branch2a = bn4b11_branch2a
        res4b11_branch2a_relu = mx.symbol.Activation(name='res4b11_branch2a_relu', data=scale4b11_branch2a,
                                                     act_type='relu')
        res4b11_branch2b = mx.symbol.Convolution(name='res4b11_branch2b', data=res4b11_branch2a_relu, num_filter=256,
                                                 pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b11_branch2b = mx.symbol.BatchNorm(name='bn4b11_branch2b', data=res4b11_branch2b, use_global_stats=True,
                                              fix_gamma=False, eps = self.eps)
        scale4b11_branch2b = bn4b11_branch2b
        res4b11_branch2b_relu = mx.symbol.Activation(name='res4b11_branch2b_relu', data=scale4b11_branch2b,
                                                     act_type='relu')
        res4b11_branch2c = mx.symbol.Convolution(name='res4b11_branch2c', data=res4b11_branch2b_relu, num_filter=1024,
                                                 pad=(0, 0), kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b11_branch2c = mx.symbol.BatchNorm(name='bn4b11_branch2c', data=res4b11_branch2c, use_global_stats=True,
                                              fix_gamma=False, eps = self.eps)
        scale4b11_branch2c = bn4b11_branch2c
        res4b11 = mx.symbol.broadcast_add(name='res4b11', *[res4b10_relu, scale4b11_branch2c])
        res4b11_relu = mx.symbol.Activation(name='res4b11_relu', data=res4b11, act_type='relu')
        res4b12_branch2a = mx.symbol.Convolution(name='res4b12_branch2a', data=res4b11_relu, num_filter=256, pad=(0, 0),
                                                 kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b12_branch2a = mx.symbol.BatchNorm(name='bn4b12_branch2a', data=res4b12_branch2a, use_global_stats=True,
                                              fix_gamma=False, eps = self.eps)
        scale4b12_branch2a = bn4b12_branch2a
        res4b12_branch2a_relu = mx.symbol.Activation(name='res4b12_branch2a_relu', data=scale4b12_branch2a,
                                                     act_type='relu')
        res4b12_branch2b = mx.symbol.Convolution(name='res4b12_branch2b', data=res4b12_branch2a_relu, num_filter=256,
                                                 pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b12_branch2b = mx.symbol.BatchNorm(name='bn4b12_branch2b', data=res4b12_branch2b, use_global_stats=True,
                                              fix_gamma=False, eps = self.eps)
        scale4b12_branch2b = bn4b12_branch2b
        res4b12_branch2b_relu = mx.symbol.Activation(name='res4b12_branch2b_relu', data=scale4b12_branch2b,
                                                     act_type='relu')
        res4b12_branch2c = mx.symbol.Convolution(name='res4b12_branch2c', data=res4b12_branch2b_relu, num_filter=1024,

bn4b6_branch2a = mx.symbol.BatchNorm(name='bn4b6_branch2a', data=res4b6_branch2a, use_global_stats=True,
                                             fix_gamma=False, eps=self.eps)
        scale4b6_branch2a = bn4b6_branch2a
        res4b6_branch2a_relu = mx.symbol.Activation(name='res4b6_branch2a_relu', data=scale4b6_branch2a, act_type='relu')
        res4b6_branch2b = mx.symbol.Convolution(name='res4b6_branch2b', data=res4b6_branch2a_relu, num_filter=256,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b6_branch2b = mx.symbol.BatchNorm(name='bn4b6_branch2b', data=res4b6_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps=self.eps)
        scale4b6_branch2b = bn4b6_branch2b
        res4b6_branch2b_relu = mx.symbol.Activation(name='res4b6_branch2b_relu', data=scale4b6_branch2b, act_type='relu')
        res4b6_branch2c = mx.symbol.Convolution(name='res4b6_branch2c', data=res4b6_branch2b_relu, num_filter=1024,
                                                pad=(0, 0), kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b6_branch2c = mx.symbol.BatchNorm(name='bn4b6_branch2c', data=res4b6_branch2c, use_global_stats=True,
                                             fix_gamma=False, eps=self.eps)
        scale4b6_branch2c = bn4b6_branch2c
        res4b6 = mx.symbol.broadcast_add(name='res4b6', *[res4b5_relu, scale4b6_branch2c])
        res4b6_relu = mx.symbol.Activation(name='res4b6_relu', data=res4b6, act_type='relu')
        res4b7_branch2a = mx.symbol.Convolution(name='res4b7_branch2a', data=res4b6_relu, num_filter=256, pad=(0, 0),
                                                kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b7_branch2a = mx.symbol.BatchNorm(name='bn4b7_branch2a', data=res4b7_branch2a, use_global_stats=True,
                                             fix_gamma=False, eps=self.eps)
        scale4b7_branch2a = bn4b7_branch2a
        res4b7_branch2a_relu = mx.symbol.Activation(name='res4b7_branch2a_relu', data=scale4b7_branch2a, act_type='relu')
        res4b7_branch2b = mx.symbol.Convolution(name='res4b7_branch2b', data=res4b7_branch2a_relu, num_filter=256,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b7_branch2b = mx.symbol.BatchNorm(name='bn4b7_branch2b', data=res4b7_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps=self.eps)
        scale4b7_branch2b = bn4b7_branch2b
        res4b7_branch2b_relu = mx.symbol.Activation(name='res4b7_branch2b_relu', data=scale4b7_branch2b, act_type='relu')
        res4b7_branch2c = mx.symbol.Convolution(name='res4b7_branch2c', data=res4b7_branch2b_relu, num_filter=1024,
                                                pad=(0, 0), kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b7_branch2c = mx.symbol.BatchNorm(name='bn4b7_branch2c', data=res4b7_branch2c, use_global_stats=True,

bn4b5_branch2a = mx.symbol.BatchNorm(name='bn4b5_branch2a', data=res4b5_branch2a, use_global_stats=True,
                                             fix_gamma=False, eps=self.eps)
        scale4b5_branch2a = bn4b5_branch2a
        res4b5_branch2a_relu = mx.symbol.Activation(name='res4b5_branch2a_relu', data=scale4b5_branch2a, act_type='relu')
        res4b5_branch2b = mx.symbol.Convolution(name='res4b5_branch2b', data=res4b5_branch2a_relu, num_filter=256,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b5_branch2b = mx.symbol.BatchNorm(name='bn4b5_branch2b', data=res4b5_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps=self.eps)
        scale4b5_branch2b = bn4b5_branch2b
        res4b5_branch2b_relu = mx.symbol.Activation(name='res4b5_branch2b_relu', data=scale4b5_branch2b, act_type='relu')
        res4b5_branch2c = mx.symbol.Convolution(name='res4b5_branch2c', data=res4b5_branch2b_relu, num_filter=1024,
                                                pad=(0, 0), kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b5_branch2c = mx.symbol.BatchNorm(name='bn4b5_branch2c', data=res4b5_branch2c, use_global_stats=True,
                                             fix_gamma=False, eps=self.eps)
        scale4b5_branch2c = bn4b5_branch2c
        res4b5 = mx.symbol.broadcast_add(name='res4b5', *[res4b4_relu, scale4b5_branch2c])
        res4b5_relu = mx.symbol.Activation(name='res4b5_relu', data=res4b5, act_type='relu')
        res4b6_branch2a = mx.symbol.Convolution(name='res4b6_branch2a', data=res4b5_relu, num_filter=256, pad=(0, 0),
                                                kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b6_branch2a = mx.symbol.BatchNorm(name='bn4b6_branch2a', data=res4b6_branch2a, use_global_stats=True,
                                             fix_gamma=False, eps=self.eps)
        scale4b6_branch2a = bn4b6_branch2a
        res4b6_branch2a_relu = mx.symbol.Activation(name='res4b6_branch2a_relu', data=scale4b6_branch2a, act_type='relu')
        res4b6_branch2b = mx.symbol.Convolution(name='res4b6_branch2b', data=res4b6_branch2a_relu, num_filter=256,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b6_branch2b = mx.symbol.BatchNorm(name='bn4b6_branch2b', data=res4b6_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps=self.eps)
        scale4b6_branch2b = bn4b6_branch2b
        res4b6_branch2b_relu = mx.symbol.Activation(name='res4b6_branch2b_relu', data=scale4b6_branch2b, act_type='relu')
        res4b6_branch2c = mx.symbol.Convolution(name='res4b6_branch2c', data=res4b6_branch2b_relu, num_filter=1024,
                                                pad=(0, 0), kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b6_branch2c = mx.symbol.BatchNorm(name='bn4b6_branch2c', data=res4b6_branch2c, use_global_stats=True,

bn2c_branch2a = mx.symbol.BatchNorm(name='bn2c_branch2a', data=res2c_branch2a, use_global_stats=True,
                                            fix_gamma=False, eps=self.eps)
        scale2c_branch2a = bn2c_branch2a
        res2c_branch2a_relu = mx.symbol.Activation(name='res2c_branch2a_relu', data=scale2c_branch2a, act_type='relu')
        res2c_branch2b = mx.symbol.Convolution(name='res2c_branch2b', data=res2c_branch2a_relu, num_filter=64, pad=(1, 1),
                                               kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn2c_branch2b = mx.symbol.BatchNorm(name='bn2c_branch2b', data=res2c_branch2b, use_global_stats=True,
                                            fix_gamma=False, eps=self.eps)
        scale2c_branch2b = bn2c_branch2b
        res2c_branch2b_relu = mx.symbol.Activation(name='res2c_branch2b_relu', data=scale2c_branch2b, act_type='relu')
        res2c_branch2c = mx.symbol.Convolution(name='res2c_branch2c', data=res2c_branch2b_relu, num_filter=256, pad=(0, 0),
                                               kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn2c_branch2c = mx.symbol.BatchNorm(name='bn2c_branch2c', data=res2c_branch2c, use_global_stats=True,
                                            fix_gamma=False, eps=self.eps)
        scale2c_branch2c = bn2c_branch2c
        res2c = mx.symbol.broadcast_add(name='res2c', *[res2b_relu, scale2c_branch2c])
        res2c_relu = mx.symbol.Activation(name='res2c_relu', data=res2c, act_type='relu')
        res3a_branch1 = mx.symbol.Convolution(name='res3a_branch1', data=res2c_relu, num_filter=512, pad=(0, 0),
                                              kernel=(1, 1), stride=(2, 2), no_bias=True)
        bn3a_branch1 = mx.symbol.BatchNorm(name='bn3a_branch1', data=res3a_branch1, use_global_stats=True, fix_gamma=False, eps=self.eps)
        scale3a_branch1 = bn3a_branch1
        res3a_branch2a = mx.symbol.Convolution(name='res3a_branch2a', data=res2c_relu, num_filter=128, pad=(0, 0),
                                               kernel=(1, 1), stride=(2, 2), no_bias=True)
        bn3a_branch2a = mx.symbol.BatchNorm(name='bn3a_branch2a', data=res3a_branch2a, use_global_stats=True,
                                            fix_gamma=False, eps=self.eps)
        scale3a_branch2a = bn3a_branch2a
        res3a_branch2a_relu = mx.symbol.Activation(name='res3a_branch2a_relu', data=scale3a_branch2a, act_type='relu')
        res3a_branch2b = mx.symbol.Convolution(name='res3a_branch2b', data=res3a_branch2a_relu, num_filter=128, pad=(1, 1),
                                               kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn3a_branch2b = mx.symbol.BatchNorm(name='bn3a_branch2b', data=res3a_branch2b, use_global_stats=True,
                                            fix_gamma=False, eps=self.eps)
        scale3a_branch2b = bn3a_branch2b

scale4b1_branch2a = bn4b1_branch2a
        res4b1_branch2a_relu = mx.symbol.Activation(name='res4b1_branch2a_relu', data=scale4b1_branch2a,
                                                    act_type='relu')
        res4b1_branch2b = mx.symbol.Convolution(name='res4b1_branch2b', data=res4b1_branch2a_relu, num_filter=256,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b1_branch2b = mx.symbol.BatchNorm(name='bn4b1_branch2b', data=res4b1_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b1_branch2b = bn4b1_branch2b
        res4b1_branch2b_relu = mx.symbol.Activation(name='res4b1_branch2b_relu', data=scale4b1_branch2b,
                                                    act_type='relu')
        res4b1_branch2c = mx.symbol.Convolution(name='res4b1_branch2c', data=res4b1_branch2b_relu, num_filter=1024,
                                                pad=(0, 0), kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b1_branch2c = mx.symbol.BatchNorm(name='bn4b1_branch2c', data=res4b1_branch2c, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b1_branch2c = bn4b1_branch2c
        res4b1 = mx.symbol.broadcast_add(name='res4b1', *[res4a_relu, scale4b1_branch2c])
        res4b1_relu = mx.symbol.Activation(name='res4b1_relu', data=res4b1, act_type='relu')
        res4b2_branch2a = mx.symbol.Convolution(name='res4b2_branch2a', data=res4b1_relu, num_filter=256, pad=(0, 0),
                                                kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b2_branch2a = mx.symbol.BatchNorm(name='bn4b2_branch2a', data=res4b2_branch2a, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b2_branch2a = bn4b2_branch2a
        res4b2_branch2a_relu = mx.symbol.Activation(name='res4b2_branch2a_relu', data=scale4b2_branch2a,
                                                    act_type='relu')
        res4b2_branch2b = mx.symbol.Convolution(name='res4b2_branch2b', data=res4b2_branch2a_relu, num_filter=256,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b2_branch2b = mx.symbol.BatchNorm(name='bn4b2_branch2b', data=res4b2_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b2_branch2b = bn4b2_branch2b
        res4b2_branch2b_relu = mx.symbol.Activation(name='res4b2_branch2b_relu', data=scale4b2_branch2b,
                                                    act_type='relu')
        res4b2_branch2c = mx.symbol.Convolution(name='res4b2_branch2c', data=res4b2_branch2b_relu, num_filter=1024,

bn5c_branch2a = mx.symbol.BatchNorm(name='bn5c_branch2a', data=res5c_branch2a, use_global_stats=True,
                                            fix_gamma=False, eps = self.eps)
        scale5c_branch2a = bn5c_branch2a
        res5c_branch2a_relu = mx.symbol.Activation(name='res5c_branch2a_relu', data=scale5c_branch2a, act_type='relu')
        res5c_branch2b = mx.symbol.Convolution(name='res5c_branch2b', data=res5c_branch2a_relu, num_filter=512,
                                               pad=(2, 2), kernel=(3, 3), dilate=(2, 2), stride=(1, 1), no_bias=True)
        bn5c_branch2b = mx.symbol.BatchNorm(name='bn5c_branch2b', data=res5c_branch2b, use_global_stats=True,
                                            fix_gamma=False, eps = self.eps)
        scale5c_branch2b = bn5c_branch2b
        res5c_branch2b_relu = mx.symbol.Activation(name='res5c_branch2b_relu', data=scale5c_branch2b, act_type='relu')
        res5c_branch2c = mx.symbol.Convolution(name='res5c_branch2c', data=res5c_branch2b_relu, num_filter=2048,
                                               pad=(0, 0), kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn5c_branch2c = mx.symbol.BatchNorm(name='bn5c_branch2c', data=res5c_branch2c, use_global_stats=True,
                                            fix_gamma=False, eps = self.eps)
        scale5c_branch2c = bn5c_branch2c
        res5c = mx.symbol.broadcast_add(name='res5c', *[res5b_relu, scale5c_branch2c])
        res5c_relu = mx.symbol.Activation(name='res5c_relu', data=res5c, act_type='relu')
        return res5c_relu

scale4b10_branch2a = bn4b10_branch2a
        res4b10_branch2a_relu = mx.symbol.Activation(name='res4b10_branch2a_relu', data=scale4b10_branch2a,
                                                     act_type='relu')
        res4b10_branch2b = mx.symbol.Convolution(name='res4b10_branch2b', data=res4b10_branch2a_relu, num_filter=256,
                                                 pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b10_branch2b = mx.symbol.BatchNorm(name='bn4b10_branch2b', data=res4b10_branch2b, use_global_stats=True,
                                              fix_gamma=False, eps=eps)
        scale4b10_branch2b = bn4b10_branch2b
        res4b10_branch2b_relu = mx.symbol.Activation(name='res4b10_branch2b_relu', data=scale4b10_branch2b,
                                                     act_type='relu')
        res4b10_branch2c = mx.symbol.Convolution(name='res4b10_branch2c', data=res4b10_branch2b_relu, num_filter=1024,
                                                 pad=(0, 0), kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b10_branch2c = mx.symbol.BatchNorm(name='bn4b10_branch2c', data=res4b10_branch2c, use_global_stats=True,
                                              fix_gamma=False, eps=eps)
        scale4b10_branch2c = bn4b10_branch2c
        res4b10 = mx.symbol.broadcast_add(name='res4b10', *[res4b9_relu, scale4b10_branch2c])
        res4b10_relu = mx.symbol.Activation(name='res4b10_relu', data=res4b10, act_type='relu')
        res4b11_branch2a = mx.symbol.Convolution(name='res4b11_branch2a', data=res4b10_relu, num_filter=256, pad=(0, 0),
                                                 kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b11_branch2a = mx.symbol.BatchNorm(name='bn4b11_branch2a', data=res4b11_branch2a, use_global_stats=True,
                                              fix_gamma=False, eps=eps)
        scale4b11_branch2a = bn4b11_branch2a
        res4b11_branch2a_relu = mx.symbol.Activation(name='res4b11_branch2a_relu', data=scale4b11_branch2a,
                                                     act_type='relu')
        res4b11_branch2b = mx.symbol.Convolution(name='res4b11_branch2b', data=res4b11_branch2a_relu, num_filter=256,
                                                 pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b11_branch2b = mx.symbol.BatchNorm(name='bn4b11_branch2b', data=res4b11_branch2b, use_global_stats=True,
                                              fix_gamma=False, eps=eps)
        scale4b11_branch2b = bn4b11_branch2b
        res4b11_branch2b_relu = mx.symbol.Activation(name='res4b11_branch2b_relu', data=scale4b11_branch2b,
                                                     act_type='relu')
        res4b11_branch2c = mx.symbol.Convolution(name='res4b11_branch2c', data=res4b11_branch2b_relu, num_filter=1024,

scale4b5_branch2a = bn4b5_branch2a
        res4b5_branch2a_relu = mx.symbol.Activation(name='res4b5_branch2a_relu', data=scale4b5_branch2a,
                                                    act_type='relu')
        res4b5_branch2b = mx.symbol.Convolution(name='res4b5_branch2b', data=res4b5_branch2a_relu, num_filter=256,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b5_branch2b = mx.symbol.BatchNorm(name='bn4b5_branch2b', data=res4b5_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b5_branch2b = bn4b5_branch2b
        res4b5_branch2b_relu = mx.symbol.Activation(name='res4b5_branch2b_relu', data=scale4b5_branch2b,
                                                    act_type='relu')
        res4b5_branch2c = mx.symbol.Convolution(name='res4b5_branch2c', data=res4b5_branch2b_relu, num_filter=1024,
                                                pad=(0, 0), kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b5_branch2c = mx.symbol.BatchNorm(name='bn4b5_branch2c', data=res4b5_branch2c, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b5_branch2c = bn4b5_branch2c
        res4b5 = mx.symbol.broadcast_add(name='res4b5', *[res4b4_relu, scale4b5_branch2c])
        res4b5_relu = mx.symbol.Activation(name='res4b5_relu', data=res4b5, act_type='relu')
        res4b6_branch2a = mx.symbol.Convolution(name='res4b6_branch2a', data=res4b5_relu, num_filter=256, pad=(0, 0),
                                                kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b6_branch2a = mx.symbol.BatchNorm(name='bn4b6_branch2a', data=res4b6_branch2a, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b6_branch2a = bn4b6_branch2a
        res4b6_branch2a_relu = mx.symbol.Activation(name='res4b6_branch2a_relu', data=scale4b6_branch2a,
                                                    act_type='relu')
        res4b6_branch2b = mx.symbol.Convolution(name='res4b6_branch2b', data=res4b6_branch2a_relu, num_filter=256,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b6_branch2b = mx.symbol.BatchNorm(name='bn4b6_branch2b', data=res4b6_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b6_branch2b = bn4b6_branch2b
        res4b6_branch2b_relu = mx.symbol.Activation(name='res4b6_branch2b_relu', data=scale4b6_branch2b,
                                                    act_type='relu')
        res4b6_branch2c = mx.symbol.Convolution(name='res4b6_branch2c', data=res4b6_branch2b_relu, num_filter=1024,

scale4b2_branch2a = bn4b2_branch2a
        res4b2_branch2a_relu = mx.symbol.Activation(name='res4b2_branch2a_relu', data=scale4b2_branch2a,
                                                    act_type='relu')
        res4b2_branch2b = mx.symbol.Convolution(name='res4b2_branch2b', data=res4b2_branch2a_relu, num_filter=256,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b2_branch2b = mx.symbol.BatchNorm(name='bn4b2_branch2b', data=res4b2_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b2_branch2b = bn4b2_branch2b
        res4b2_branch2b_relu = mx.symbol.Activation(name='res4b2_branch2b_relu', data=scale4b2_branch2b,
                                                    act_type='relu')
        res4b2_branch2c = mx.symbol.Convolution(name='res4b2_branch2c', data=res4b2_branch2b_relu, num_filter=1024,
                                                pad=(0, 0), kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b2_branch2c = mx.symbol.BatchNorm(name='bn4b2_branch2c', data=res4b2_branch2c, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b2_branch2c = bn4b2_branch2c
        res4b2 = mx.symbol.broadcast_add(name='res4b2', *[res4b1_relu, scale4b2_branch2c])
        res4b2_relu = mx.symbol.Activation(name='res4b2_relu', data=res4b2, act_type='relu')
        res4b3_branch2a = mx.symbol.Convolution(name='res4b3_branch2a', data=res4b2_relu, num_filter=256, pad=(0, 0),
                                                kernel=(1, 1), stride=(1, 1), no_bias=True)
        bn4b3_branch2a = mx.symbol.BatchNorm(name='bn4b3_branch2a', data=res4b3_branch2a, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b3_branch2a = bn4b3_branch2a
        res4b3_branch2a_relu = mx.symbol.Activation(name='res4b3_branch2a_relu', data=scale4b3_branch2a,
                                                    act_type='relu')
        res4b3_branch2b = mx.symbol.Convolution(name='res4b3_branch2b', data=res4b3_branch2a_relu, num_filter=256,
                                                pad=(1, 1), kernel=(3, 3), stride=(1, 1), no_bias=True)
        bn4b3_branch2b = mx.symbol.BatchNorm(name='bn4b3_branch2b', data=res4b3_branch2b, use_global_stats=True,
                                             fix_gamma=False, eps = self.eps)
        scale4b3_branch2b = bn4b3_branch2b
        res4b3_branch2b_relu = mx.symbol.Activation(name='res4b3_branch2b_relu', data=scale4b3_branch2b,
                                                    act_type='relu')
        res4b3_branch2c = mx.symbol.Convolution(name='res4b3_branch2c', data=res4b3_branch2b_relu, num_filter=1024,