myBN Class Reference

Inheritance diagram for myBN:

Public Member Functions
	__init__ (self, num_channels, eps=1e-5, momentum=0.1)
	Constructor.
	reset_stats (self)
	reset standing stats
	forward (self, x, gain, bias)
	forward

Public Attributes
	momentum = momentum
	momentum for updating running stats
	eps = eps
	epsilon to avoid dividing by 0
bool	accumulate_standing = False
	Accumulate running means and vars.
bool	training
	Training mode flag (inherited from nn.Module).

Detailed Description

My batchnorm, supports standing stats

Definition at line 419 of file ieagan.py.

Constructor & Destructor Documentation

__init__()

__init__	(		self,
			num_channels,
			eps = 1e-5,
			momentum = 0.1 )

Constructor.

Definition at line 425 of file ieagan.py.

    def __init__(self, num_channels, eps=1e-5, momentum=0.1):
        super(myBN, self).__init__()
        
        self.momentum = momentum
        
        self.eps = eps
        # Momentum
        self.momentum = momentum
        # Register buffers
        self.register_buffer("stored_mean", torch.zeros(num_channels))
        self.register_buffer("stored_var", torch.ones(num_channels))
        self.register_buffer("accumulation_counter", torch.zeros(1))
        
        self.accumulate_standing = False
        
        self.training: bool
 

Member Function Documentation

forward()

forward	(		self,
			x,
			gain,
			bias )

forward

Definition at line 449 of file ieagan.py.

    def forward(self, x, gain, bias):
        if self.training:
            out, mean, var = manual_bn(
                x, gain, bias, return_mean_var=True, eps=self.eps
            )
            # If accumulating standing stats, increment them
            if self.accumulate_standing:
                self.stored_mean[:] = self.stored_mean + mean.data
                self.stored_var[:] = self.stored_var + var.data
                self.accumulation_counter += 1.0
            # If not accumulating standing stats, take running averages
            else:
                self.stored_mean[:] = (
                    self.stored_mean * (1 - self.momentum) + mean * self.momentum
                )
                self.stored_var[:] = (
                    self.stored_var * (1 - self.momentum) + var * self.momentum
                )
            return out
        # If not in training mode, use the stored statistics
        else:
            mean = self.stored_mean.view(1, -1, 1, 1)
            var = self.stored_var.view(1, -1, 1, 1)
            # If using standing stats, divide them by the accumulation counter
            if self.accumulate_standing:
                mean = mean / self.accumulation_counter
                var = var / self.accumulation_counter
            return fused_bn(x, mean, var, gain, bias, self.eps)
 
 

reset_stats()

reset_stats

(

self

)

reset standing stats

Definition at line 443 of file ieagan.py.

    def reset_stats(self):
        self.stored_mean[:] = 0
        self.stored_var[:] = 0
        self.accumulation_counter[:] = 0
 

Member Data Documentation

accumulate_standing

bool accumulate_standing = False

Accumulate running means and vars.

Definition at line 438 of file ieagan.py.

eps

eps = eps

epsilon to avoid dividing by 0

Definition at line 430 of file ieagan.py.

momentum

momentum = momentum

momentum for updating running stats

Definition at line 428 of file ieagan.py.

training

bool training

Training mode flag (inherited from nn.Module).

True if the module is in training mode.

Definition at line 440 of file ieagan.py.

---

The documentation for this class was generated from the following file:

• pxd/scripts/pxd/background_generator/models/ieagan.py