Python init.uniform方法代码示例

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def assert_and_infer_cfg(make_immutable=True):
    """Call this function in your script after you have finished setting all cfg
    values that are necessary (e.g., merging a config from a file, merging
    command line config options, etc.). By default, this function will also
    mark the global cfg as immutable to prevent changing the global cfg settings
    during script execution (which can lead to hard to debug errors or code
    that's harder to understand than is necessary).
    """
    if __C.MODEL.LOAD_IMAGENET_PRETRAINED_WEIGHTS:
        assert __C.VGG.IMAGENET_PRETRAINED_WEIGHTS, \
            "Path to the weight file must not be empty to load imagenet pertrained resnets."
    if version.parse(torch.__version__) < version.parse('0.4.0'):
        __C.PYTORCH_VERSION_LESS_THAN_040 = True
        # create alias for PyTorch version less than 0.4.0
        init.uniform_ = init.uniform
        init.normal_ = init.normal
        init.constant_ = init.constant
        nn.GroupNorm = mynn.GroupNorm
    if make_immutable:
        cfg.immutable(True) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def __init__(self, input_size, activation=nn.Tanh(),
                 method="dot"):
        super(AttnScore, self).__init__()
        self.activation = activation
        self.input_size = input_size
        self.method = method
        if method == "general":
            self.linear = nn.Linear(input_size, input_size)
            init.uniform(self.linear.weight.data, -0.005, 0.005)
        elif method == "concat":
            self.linear_1 = nn.Linear(input_size*2, input_size)
            self.linear_2 = nn.Linear(input_size, 1)
            init.uniform(self.linear_1.weight.data, -0.005, 0.005)
            init.uniform(self.linear_2.weight.data, -0.005, 0.005)
        elif method == "tri_concat":
            self.linear = nn.Linear(input_size*3, 1)
            init.uniform(self.linear.weight.data, -0.005, 0.005) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def forward_test(self, input_features, adj):
        self.max_num_nodes = 4
        adj_data = torch.zeros(self.max_num_nodes, self.max_num_nodes)
        adj_data[:4, :4] = torch.FloatTensor([[1,1,0,0], [1,1,1,0], [0,1,1,1], [0,0,1,1]])
        adj_features = torch.Tensor([2,3,3,2])

        adj_data1 = torch.zeros(self.max_num_nodes, self.max_num_nodes)
        adj_data1 = torch.FloatTensor([[1,1,1,0], [1,1,0,1], [1,0,1,0], [0,1,0,1]])
        adj_features1 = torch.Tensor([3,3,2,2])
        S = self.edge_similarity_matrix(adj_data, adj_data1, adj_features, adj_features1,
                self.deg_feature_similarity)

        # initialization strategies
        init_corr = 1 / self.max_num_nodes
        init_assignment = torch.ones(self.max_num_nodes, self.max_num_nodes) * init_corr
        #init_assignment = torch.FloatTensor(4, 4)
        #init.uniform(init_assignment)
        assignment = self.mpm(init_assignment, S)
        #print('Assignment: ', assignment)

        # matching
        row_ind, col_ind = scipy.optimize.linear_sum_assignment(-assignment.numpy())
        print('row: ', row_ind)
        print('col: ', col_ind)

        permuted_adj = self.permute_adj(adj_data, row_ind, col_ind)
        print('permuted: ', permuted_adj)

        adj_recon_loss = self.adj_recon_loss(permuted_adj, adj_data1)
        print(adj_data1)
        print('diff: ', adj_recon_loss) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def weights_init_normal(m):
    classname = m.__class__.__name__
    # print(classname)
    if classname.find('Conv') != -1:
        init.uniform(m.weight.data, 0.0, 0.02)
    elif classname.find('Linear') != -1:
        init.uniform(m.weight.data, 0.0, 0.02)
    elif classname.find('BatchNorm2d') != -1:
        init.uniform(m.weight.data, 1.0, 0.02)
        init.constant(m.bias.data, 0.0) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def weights_init_xavier(m):
    classname = m.__class__.__name__
    # print(classname)
    if classname.find('Conv') != -1:
        init.xavier_normal(m.weight.data, gain=1)
    elif classname.find('Linear') != -1:
        init.xavier_normal(m.weight.data, gain=1)
    elif classname.find('BatchNorm2d') != -1:
        init.uniform(m.weight.data, 1.0, 0.02)
        init.constant(m.bias.data, 0.0) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def weights_init_kaiming(m):
    classname = m.__class__.__name__
    # print(classname)
    if classname.find('Conv') != -1:
        init.kaiming_normal(m.weight.data, a=0, mode='fan_in')
    elif classname.find('Linear') != -1:
        init.kaiming_normal(m.weight.data, a=0, mode='fan_in')
    elif classname.find('BatchNorm2d') != -1:
        init.uniform(m.weight.data, 1.0, 0.02)
        init.constant(m.bias.data, 0.0) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def weights_init_orthogonal(m):
    classname = m.__class__.__name__
    print(classname)
    if classname.find('Conv') != -1:
        init.orthogonal(m.weight.data, gain=1)
    elif classname.find('Linear') != -1:
        init.orthogonal(m.weight.data, gain=1)
    elif classname.find('BatchNorm2d') != -1:
        init.uniform(m.weight.data, 1.0, 0.02)
        init.constant(m.bias.data, 0.0) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def assert_and_infer_cfg(make_immutable=True):
    """Call this function in your script after you have finished setting all cfg
    values that are necessary (e.g., merging a config from a file, merging
    command line config options, etc.). By default, this function will also
    mark the global cfg as immutable to prevent changing the global cfg settings
    during script execution (which can lead to hard to debug errors or code
    that's harder to understand than is necessary).
    """
    if __C.MODEL.RPN_ONLY or __C.MODEL.FASTER_RCNN:
        __C.RPN.RPN_ON = True
    if __C.RPN.RPN_ON or __C.RETINANET.RETINANET_ON:
        __C.TEST.PRECOMPUTED_PROPOSALS = False
    if __C.MODEL.LOAD_IMAGENET_PRETRAINED_WEIGHTS:
        assert __C.RESNETS.IMAGENET_PRETRAINED_WEIGHTS, \
            "Path to the weight file must not be empty to load imagenet pertrained resnets."
    if set([__C.MRCNN.ROI_MASK_HEAD, __C.KRCNN.ROI_KEYPOINTS_HEAD]) & _SHARE_RES5_HEADS:
        __C.MODEL.SHARE_RES5 = True
    if version.parse(torch.__version__) < version.parse('0.4.0'):
        __C.PYTORCH_VERSION_LESS_THAN_040 = True
        # create alias for PyTorch version less than 0.4.0
        init.uniform_ = init.uniform
        init.normal_ = init.normal
        init.constant_ = init.constant
        nn.GroupNorm = mynn.GroupNorm
    if make_immutable:
        cfg.immutable(True) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def __init__(self, d_in, d_out, bias=True, out_norm=False):
        super().__init__(d_in, d_out, bias)
        self.out_norm = out_norm
        stdv = 1. / math.sqrt(self.weight.size(1))
        init.uniform(self.weight, -stdv, stdv)
        if bias:
            self.bias.data.zero_() 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def init_weights(self):
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                if m.bias is not None:
                    init.uniform(m.bias)
                init.xavier_uniform(m.weight)

            if isinstance(m, nn.ConvTranspose2d):
                if m.bias is not None:
                    init.uniform(m.bias)
                init.xavier_uniform(m.weight)
                # init_deconv_bilinear(m.weight) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def init_weights(self):
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                if m.bias is not None:
                    init.uniform(m.bias)
                init.xavier_uniform(m.weight)

            if isinstance(m, nn.ConvTranspose2d):
                if m.bias is not None:
                    init.uniform(m.bias)
                init.xavier_uniform(m.weight) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def __init__(self, with_bn=False, fp16=False, rgb_max=255., div_flow=20):
        super(FlowNet2CS, self).__init__()
        self.with_bn = with_bn
        self.fp16 = fp16
        self.rgb_max = rgb_max
        self.div_flow = div_flow

        self.channelnorm = ChannelNorm()

        # First Block (FlowNetC)
        self.flownetc = FlowNetC(with_bn=with_bn, fp16=fp16)
        self.upsample1 = nn.Upsample(scale_factor=4, mode='bilinear')

        self.resample1 = (nn.Sequential(tofp32(), Resample2d(), tofp16())
                          if fp16 else Resample2d())

        # Block (FlowNetS1)
        self.flownets_1 = FlowNetS(with_bn=with_bn)
        self.upsample2 = nn.Upsample(scale_factor=4, mode='bilinear')

        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                if m.bias is not None:
                    nn_init.uniform(m.bias)
                nn_init.xavier_uniform(m.weight)

            if isinstance(m, nn.ConvTranspose2d):
                if m.bias is not None:
                    nn_init.uniform(m.bias)
                nn_init.xavier_uniform(m.weight) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def weights_init_normal(m):
  classname = m.__class__.__name__
  # print(classname)
  if classname.find('Conv') != -1:
    init.uniform(m.weight.data, 0.0, 0.02)
  elif classname.find('Linear') != -1:
    init.uniform(m.weight.data, 0.0, 0.02)
  elif classname.find('BatchNorm2d') != -1:
    init.uniform(m.weight.data, 1.0, 0.02)
    init.constant(m.bias.data, 0.0) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def weights_init_xavier(m):
  classname = m.__class__.__name__
  # print(classname)
  if classname.find('Conv') != -1:
    init.xavier_normal(m.weight.data, gain=1)
  elif classname.find('Linear') != -1:
    init.xavier_normal(m.weight.data, gain=1)
  elif classname.find('BatchNorm2d') != -1:
    init.uniform(m.weight.data, 1.0, 0.02)
    init.constant(m.bias.data, 0.0) 

# 需要导入模块: from torch.nn import init [as 别名]
# 或者: from torch.nn.init import uniform [as 别名]
def weights_init_kaiming(m):
  classname = m.__class__.__name__
  # print(classname)
  if classname.find('Conv') != -1:
    init.kaiming_normal(m.weight.data, a=0, mode='fan_in')
  elif classname.find('Linear') != -1:
    init.kaiming_normal(m.weight.data, a=0, mode='fan_in')
  elif classname.find('BatchNorm2d') != -1:
    init.uniform(m.weight.data, 1.0, 0.02)
    init.constant(m.bias.data, 0.0)