Module flowcon.CNF.neural_odes.diffeq_layers.resnet

Classes

class BasicBlock (dim, conv_block=None)

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes::

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call :meth:to, etc.

Note

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

:ivar training: Boolean represents whether this module is in training or evaluation mode. :vartype training: bool

Initialize internal Module state, shared by both nn.Module and ScriptModule.

Expand source code 
class BasicBlock(nn.Module):
    expansion = 1

    def __init__(self, dim, conv_block=None):
        super(BasicBlock, self).__init__()

        if conv_block is None:
            conv_block = basic.ConcatCoordConv2d

        self.norm1 = nn.GroupNorm(NGROUPS, dim, eps=1e-4)
        self.relu1 = nn.ReLU(inplace=True)
        self.conv1 = conv_block(dim, dim, ksize=3, stride=1, padding=1, bias=False)
        self.norm2 = nn.GroupNorm(NGROUPS, dim, eps=1e-4)
        self.relu2 = nn.ReLU(inplace=True)
        self.conv2 = conv_block(dim, dim, ksize=3, stride=1, padding=1, bias=False)

    def forward(self, t, x):
        residual = x

        out = self.norm1(x)
        out = self.relu1(out)
        out = self.conv1(t, out)

        out = self.norm2(out)
        out = self.relu2(out)
        out = self.conv2(t, out)

        out += residual

        return out

Ancestors

  • torch.nn.modules.module.Module

Class variables

var call_super_init : bool
var dump_patches : bool
var expansion
var training : bool

Methods

def forward(self, t, x) ‑> Callable[..., Any]

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the :class:Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class ResNet (dim, intermediate_dim, n_resblocks, conv_block=None)

A container for a sequential chain of layers. Supports both regular and diffeq layers.

Initialize internal Module state, shared by both nn.Module and ScriptModule.

Expand source code 
class ResNet(container.SequentialDiffEq):
    def __init__(self, dim, intermediate_dim, n_resblocks, conv_block=None):
        super(ResNet, self).__init__()

        if conv_block is None:
            conv_block = basic.ConcatCoordConv2d

        self.dim = dim
        self.intermediate_dim = intermediate_dim
        self.n_resblocks = n_resblocks

        layers = []
        layers.append(conv_block(dim, intermediate_dim, ksize=3, stride=1, padding=1, bias=False))
        for _ in range(n_resblocks):
            layers.append(BasicBlock(intermediate_dim, conv_block))
        layers.append(nn.GroupNorm(NGROUPS, intermediate_dim, eps=1e-4))
        layers.append(nn.ReLU(inplace=True))
        layers.append(conv_block(intermediate_dim, dim, ksize=1, bias=False))

        super(ResNet, self).__init__(*layers)

    def __repr__(self):
        return (
            '{name}({dim}, intermediate_dim={intermediate_dim}, n_resblocks={n_resblocks})'.format(
                name=self.__class__.__name__, **self.__dict__
            )
        )

Ancestors

Class variables

var call_super_init : bool
var dump_patches : bool
var training : bool

Inherited members