def init_weights(self):
# initialize weight and bias
for m in self.modules():
if isinstance(m, nn.Conv2d):
xavier_init(m)
elif isinstance(m, nn.BatchNorm2d):
uniform_init(m)
def test_uniform_init():
conv_module = nn.Conv2d(3, 16, 3)
uniform_init(conv_module, bias=0.1)
# TODO: sanity check of weight distribution, e.g. mean, std
assert conv_module.bias.allclose(torch.full_like(conv_module.bias, 0.1))
conv_module_no_bias = nn.Conv2d(3, 16, 3, bias=False)
uniform_init(conv_module_no_bias)
def init_weights(self, pretrained=None):
# initialize weight and bias
for m in self.modules():
if isinstance(m, nn.Conv2d):
kaiming_init(m)
elif isinstance(m, nn.BatchNorm2d):
uniform_init(m)
def init_weights(self, pretrained=None):
if pretrained is not None:
logger = get_root_logger()
load_checkpoint(self, pretrained, strict=False, logger=logger)
else:
for m in self.modules():
if isinstance(m, nn.Conv2d):
xavier_init(m)
elif isinstance(m, nn.BatchNorm2d):
uniform_init(m)
def init_weights(self, pretrained=None):
for m in self.modules():
if isinstance(m, nn.Conv2d):
xavier_init(m)
elif isinstance(m, nn.BatchNorm2d):
uniform_init(m)
def init_weights(self):
"""Initialize the learnable weights."""
uniform_init(self.row_embed)
uniform_init(self.col_embed)
