module
def module(
*flds, **defaults
):Decorator to create an nn.Module using f as forward method
Layers
Custom fastai layers and basic functions to grab them.
Basic manipulations and resize
test_eq(Identity()(1), 1)Lambda
def Lambda(
*args, **kwargs
):An easy way to create a pytorch layer for a simple func
def _add2(x): return x+2
tst = Lambda(_add2)
x = torch.randn(10,20)
test_eq(tst(x), x+2)
tst2 = pickle.loads(pickle.dumps(tst))
test_eq(tst2(x), x+2)
tstLambda(func=<function _add2>)PartialLambda
def PartialLambda(
func, **kwargs
):Layer that applies partial(func, **kwargs)
def test_func(a,b=2): return a+b
tst = PartialLambda(test_func, b=5)
test_eq(tst(x), x+5)Flatten
def Flatten(
*args, **kwargs
):Flatten x to a single dimension, e.g. at end of a model. full for rank-1 tensor
tst = Flatten()
x = torch.randn(10,5,4)
test_eq(tst(x).shape, [10,20])
tst = Flatten(full=True)
test_eq(tst(x).shape, [200])ttb = ToTensorBase()
timg = TensorImage(torch.rand(1,3,32,32))
test_eq(type(ttb(timg)), TensorBase)tst = View(10,5,4)
test_eq(tst(x).shape, [10,5,4])tst = ResizeBatch(5,4)
test_eq(tst(x).shape, [10,5,4])test = tensor([-10.,0.,10.])
assert torch.allclose(sigmoid_range(test, -1, 2), tensor([-1.,0.5, 2.]), atol=1e-4, rtol=1e-4)
assert torch.allclose(sigmoid_range(test, -5, -1), tensor([-5.,-3.,-1.]), atol=1e-4, rtol=1e-4)
assert torch.allclose(sigmoid_range(test, 2, 4), tensor([2., 3., 4.]), atol=1e-4, rtol=1e-4)tst = SigmoidRange(-1, 2)
assert torch.allclose(tst(test), tensor([-1.,0.5, 2.]), atol=1e-4, rtol=1e-4)Pooling layers
AdaptiveConcatPool1d
def AdaptiveConcatPool1d(
size:NoneType=None
):Layer that concats AdaptiveAvgPool1d and AdaptiveMaxPool1d
AdaptiveConcatPool2d
def AdaptiveConcatPool2d(
size:NoneType=None
):Layer that concats AdaptiveAvgPool2d and AdaptiveMaxPool2d
If the input is bs x nf x h x h, the output will be bs x 2*nf x 1 x 1 if no size is passed or bs x 2*nf x size x size
tst = AdaptiveConcatPool2d()
x = torch.randn(10,5,4,4)
test_eq(tst(x).shape, [10,10,1,1])
max1 = torch.max(x, dim=2, keepdim=True)[0]
maxp = torch.max(max1, dim=3, keepdim=True)[0]
test_eq(tst(x)[:,:5], maxp)
test_eq(tst(x)[:,5:], x.mean(dim=[2,3], keepdim=True))
tst = AdaptiveConcatPool2d(2)
test_eq(tst(x).shape, [10,10,2,2])PoolType
def PoolType(
*args, **kwargs
):Initialize self. See help(type(self)) for accurate signature.
PoolFlatten
def PoolFlatten(
pool_type:str='Avg'
):Combine nn.AdaptiveAvgPool2d and Flatten.
tst = PoolFlatten()
test_eq(tst(x).shape, [10,5])
test_eq(tst(x), x.mean(dim=[2,3]))BatchNorm layers
BatchNorm
def BatchNorm(
nf, ndim:int=2, norm_type:NormType=<NormType.Batch: 1>, eps:float=1e-05, momentum:float | None=0.1,
affine:bool=True, track_running_stats:bool=True, device:NoneType=None, dtype:NoneType=None, bias:bool=True
):BatchNorm layer with nf features and ndim initialized depending on norm_type.
InstanceNorm
def InstanceNorm(
nf, ndim:int=2, norm_type:NormType=<NormType.Instance: 5>, affine:bool=True, eps:float=1e-05, momentum:float=0.1,
track_running_stats:bool=False, device:NoneType=None, dtype:NoneType=None,
bias:bool=True, # for backward compatibility
):InstanceNorm layer with nf features and ndim initialized depending on norm_type.
kwargs are passed to nn.BatchNorm and can be eps, momentum, affine and track_running_stats.
tst = BatchNorm(15)
assert isinstance(tst, nn.BatchNorm2d)
test_eq(tst.weight, torch.ones(15))
tst = BatchNorm(15, norm_type=NormType.BatchZero)
test_eq(tst.weight, torch.zeros(15))
tst = BatchNorm(15, ndim=1)
assert isinstance(tst, nn.BatchNorm1d)
tst = BatchNorm(15, ndim=3)
assert isinstance(tst, nn.BatchNorm3d)tst = InstanceNorm(15)
assert isinstance(tst, nn.InstanceNorm2d)
test_eq(tst.weight, torch.ones(15))
tst = InstanceNorm(15, norm_type=NormType.InstanceZero)
test_eq(tst.weight, torch.zeros(15))
tst = InstanceNorm(15, ndim=1)
assert isinstance(tst, nn.InstanceNorm1d)
tst = InstanceNorm(15, ndim=3)
assert isinstance(tst, nn.InstanceNorm3d)If affine is false the weight should be None
test_eq(BatchNorm(15, affine=False).weight, None)
test_eq(InstanceNorm(15, affine=False).weight, None)BatchNorm1dFlat
def BatchNorm1dFlat(
num_features:int, eps:float=1e-05, momentum:float | None=0.1, affine:bool=True, track_running_stats:bool=True,
device:NoneType=None, dtype:NoneType=None, bias:bool=True
)->None:nn.BatchNorm1d, but first flattens leading dimensions
tst = BatchNorm1dFlat(15)
x = torch.randn(32, 64, 15)
y = tst(x)
mean = x.mean(dim=[0,1])
test_close(tst.running_mean, 0*0.9 + mean*0.1)
var = (x-mean).pow(2).mean(dim=[0,1])
test_close(tst.running_var, 1*0.9 + var*0.1, eps=1e-4)
test_close(y, (x-mean)/torch.sqrt(var+1e-5) * tst.weight + tst.bias, eps=1e-4)LinBnDrop
def LinBnDrop(
n_in, n_out, bn:bool=True, p:float=0.0, act:NoneType=None, lin_first:bool=False
):Module grouping BatchNorm1d, Dropout and Linear layers
The BatchNorm layer is skipped if bn=False, as is the dropout if p=0.. Optionally, you can add an activation for after the linear layer with act.
tst = LinBnDrop(10, 20)
mods = list(tst.children())
test_eq(len(mods), 2)
assert isinstance(mods[0], nn.BatchNorm1d)
assert isinstance(mods[1], nn.Linear)
tst = LinBnDrop(10, 20, p=0.1)
mods = list(tst.children())
test_eq(len(mods), 3)
assert isinstance(mods[0], nn.BatchNorm1d)
assert isinstance(mods[1], nn.Dropout)
assert isinstance(mods[2], nn.Linear)
tst = LinBnDrop(10, 20, act=nn.ReLU(), lin_first=True)
mods = list(tst.children())
test_eq(len(mods), 3)
assert isinstance(mods[0], nn.Linear)
assert isinstance(mods[1], nn.ReLU)
assert isinstance(mods[2], nn.BatchNorm1d)
tst = LinBnDrop(10, 20, bn=False)
mods = list(tst.children())
test_eq(len(mods), 1)
assert isinstance(mods[0], nn.Linear)Inits
sigmoid
def sigmoid(
input, eps:float=1e-07
):Same as torch.sigmoid, plus clamping to `(eps,1-eps)
sigmoid_
def sigmoid_(
input, eps:float=1e-07
):Same as torch.sigmoid_, plus clamping to `(eps,1-eps)
init_default
def init_default(
m, func:function=kaiming_normal_
):Initialize m weights with func and set bias to 0.
init_linear
def init_linear(
m, act_func:NoneType=None, init:str='auto', bias_std:float=0.01
):Call self as a function.
Convolutions
ConvLayer
def ConvLayer(
ni, nf, ks:int=3, stride:int=1, padding:NoneType=None, bias:NoneType=None, ndim:int=2,
norm_type:NormType=<NormType.Batch: 1>, bn_1st:bool=True, act_cls:type=ReLU, transpose:bool=False,
init:str='auto', xtra:NoneType=None, bias_std:float=0.01, dilation:Union=1, groups:int=1,
padding_mode:Literal='zeros', device:NoneType=None, dtype:NoneType=None
):Create a sequence of convolutional (ni to nf), ReLU (if use_activ) and norm_type layers.
The convolution uses ks (kernel size) stride, padding and bias. padding will default to the appropriate value ((ks-1)//2 if it’s not a transposed conv) and bias will default to True the norm_type is Spectral or Weight, False if it’s Batch or BatchZero. Note that if you don’t want any normalization, you should pass norm_type=None.
This defines a conv layer with ndim (1,2 or 3) that will be a ConvTranspose if transpose=True. act_cls is the class of the activation function to use (instantiated inside). Pass act=None if you don’t want an activation function. If you quickly want to change your default activation, you can change the value of defaults.activation.
init is used to initialize the weights (the bias are initialized to 0) and xtra is an optional layer to add at the end.
tst = ConvLayer(16, 32)
mods = list(tst.children())
test_eq(len(mods), 3)
test_eq(mods[1].weight, torch.ones(32))
test_eq(mods[0].padding, (1,1))x = torch.randn(64, 16, 8, 8)#.cuda()#Padding is selected to make the shape the same if stride=1
test_eq(tst(x).shape, [64,32,8,8])#Padding is selected to make the shape half if stride=2
tst = ConvLayer(16, 32, stride=2)
test_eq(tst(x).shape, [64,32,4,4])#But you can always pass your own padding if you want
tst = ConvLayer(16, 32, padding=0)
test_eq(tst(x).shape, [64,32,6,6])#No bias by default for Batch NormType
assert mods[0].bias is None
#But can be overridden with `bias=True`
tst = ConvLayer(16, 32, bias=True)
assert first(tst.children()).bias is not None
#For no norm, or spectral/weight, bias is True by default
for t in [None, NormType.Spectral, NormType.Weight]:
tst = ConvLayer(16, 32, norm_type=t)
assert first(tst.children()).bias is not None#Various n_dim/tranpose
tst = ConvLayer(16, 32, ndim=3)
assert isinstance(list(tst.children())[0], nn.Conv3d)
tst = ConvLayer(16, 32, ndim=1, transpose=True)
assert isinstance(list(tst.children())[0], nn.ConvTranspose1d)#No activation/leaky
tst = ConvLayer(16, 32, ndim=3, act_cls=None)
mods = list(tst.children())
test_eq(len(mods), 2)
tst = ConvLayer(16, 32, ndim=3, act_cls=partial(nn.LeakyReLU, negative_slope=0.1))
mods = list(tst.children())
test_eq(len(mods), 3)
assert isinstance(mods[2], nn.LeakyReLU)# #export
# def linear(in_features, out_features, bias=True, act_cls=None, init='auto'):
# "Linear layer followed by optional activation, with optional auto-init"
# res = nn.Linear(in_features, out_features, bias=bias)
# if act_cls: act_cls = act_cls()
# init_linear(res, act_cls, init=init)
# if act_cls: res = nn.Sequential(res, act_cls)
# return res# #export
# @delegates(ConvLayer)
# def conv1d(ni, nf, ks, stride=1, ndim=1, norm_type=None, **kwargs):
# "Convolutional layer followed by optional activation, with optional auto-init"
# return ConvLayer(ni, nf, ks, stride=stride, ndim=ndim, norm_type=norm_type, **kwargs)# #export
# @delegates(ConvLayer)
# def conv2d(ni, nf, ks, stride=1, ndim=2, norm_type=None, **kwargs):
# "Convolutional layer followed by optional activation, with optional auto-init"
# return ConvLayer(ni, nf, ks, stride=stride, ndim=ndim, norm_type=norm_type, **kwargs)# #export
# @delegates(ConvLayer)
# def conv3d(ni, nf, ks, stride=1, ndim=3, norm_type=None, **kwargs):
# "Convolutional layer followed by optional activation, with optional auto-init"
# return ConvLayer(ni, nf, ks, stride=stride, ndim=ndim, norm_type=norm_type, **kwargs)AdaptiveAvgPool
def AdaptiveAvgPool(
sz:int=1, ndim:int=2
):nn.AdaptiveAvgPool layer for ndim
MaxPool
def MaxPool(
ks:int=2, stride:NoneType=None, padding:int=0, ndim:int=2, ceil_mode:bool=False
):nn.MaxPool layer for ndim
AvgPool
def AvgPool(
ks:int=2, stride:NoneType=None, padding:int=0, ndim:int=2, ceil_mode:bool=False
):nn.AvgPool layer for ndim
Embeddings
trunc_normal_
def trunc_normal_(
x, mean:float=0.0, std:float=1.0
):Truncated normal initialization (approximation)
Embedding
def Embedding(
ni, nf, std:float=0.01
):Embedding layer with truncated normal initialization
Truncated normal initialization bounds the distribution to avoid large value. For a given standard deviation std, the bounds are roughly -2*std, 2*std.
std = 0.02
tst = Embedding(10, 30, std)
assert tst.weight.min() > -2*std
assert tst.weight.max() < 2*std
test_close(tst.weight.mean(), 0, 1e-2)
test_close(tst.weight.std(), std, 0.1)Self attention
Self-attention layer as introduced in Self-Attention Generative Adversarial Networks.
Initially, no change is done to the input. This is controlled by a trainable parameter named gamma as we return x + gamma * out.
tst = SelfAttention(16)
x = torch.randn(32, 16, 8, 8)
test_eq(tst(x),x)Then during training gamma will probably change since it’s a trainable parameter. Let’s see what’s happening when it gets a nonzero value.
tst.gamma.data.fill_(1.)
y = tst(x)
test_eq(y.shape, [32,16,8,8])The attention mechanism requires three matrix multiplications (here represented by 1x1 convs). The multiplications are done on the channel level (the second dimension in our tensor) and we flatten the feature map (which is 8x8 here). As in the paper, we note f, g and h the results of those multiplications.
q,k,v = tst.query[0].weight.data,tst.key[0].weight.data,tst.value[0].weight.data
test_eq([q.shape, k.shape, v.shape], [[2, 16, 1], [2, 16, 1], [16, 16, 1]])
f,g,h = map(lambda m: x.view(32, 16, 64).transpose(1,2) @ m.squeeze().t(), [q,k,v])
test_eq([f.shape, g.shape, h.shape], [[32,64,2], [32,64,2], [32,64,16]])The key part of the attention layer is to compute attention weights for each of our location in the feature map (here 8x8 = 64). Those are positive numbers that sum to 1 and tell the model to pay attention to this or that part of the picture. We make the product of f and the transpose of g (to get something of size bs by 64 by 64) then apply a softmax on the first dimension (to get the positive numbers that sum up to 1). The result can then be multiplied with h transposed to get an output of size bs by channels by 64, which we can then be viewed as an output the same size as the original input.
The final result is then x + gamma * out as we saw before.
beta = F.softmax(torch.bmm(f, g.transpose(1,2)), dim=1)
test_eq(beta.shape, [32, 64, 64])
out = torch.bmm(h.transpose(1,2), beta)
test_eq(out.shape, [32, 16, 64])
test_close(y, x + out.view(32, 16, 8, 8), eps=1e-4)PooledSelfAttention2d
def PooledSelfAttention2d(
n_channels
):Pooled self attention layer for 2d.
Self-attention layer used in the Big GAN paper.
It uses the same attention as in SelfAttention but adds a max pooling of stride 2 before computing the matrices g and h: the attention is ported on one of the 2x2 max-pooled window, not the whole feature map. There is also a final matrix product added at the end to the output, before retuning gamma * out + x.
SimpleSelfAttention
def SimpleSelfAttention(
n_in:int, ks:int=1, sym:bool=False
):Same as nn.Module, but no need for subclasses to call super().__init__
PixelShuffle
PixelShuffle introduced in this article to avoid checkerboard artifacts when upsampling images. If we want an output with ch_out filters, we use a convolution with ch_out * (r**2) filters, where r is the upsampling factor. Then we reorganize those filters like in the picture below:
icnr_init
def icnr_init(
x, scale:int=2, init:function=kaiming_normal_
):ICNR init of x, with scale and init function
ICNR init was introduced in this article. It suggests to initialize the convolution that will be used in PixelShuffle so that each of the r**2 channels get the same weight (so that in the picture above, the 9 colors in a 3 by 3 window are initially the same).
Note
This is done on the first dimension because PyTorch stores the weights of a convolutional layer in this format: ch_out x ch_in x ks x ks.
tst = torch.randn(16*4, 32, 1, 1)
tst = icnr_init(tst)
for i in range(0,16*4,4):
test_eq(tst[i],tst[i+1])
test_eq(tst[i],tst[i+2])
test_eq(tst[i],tst[i+3])PixelShuffle_ICNR
def PixelShuffle_ICNR(
ni, nf:NoneType=None, scale:int=2, blur:bool=False, norm_type:NormType=<NormType.Weight: 3>, act_cls:type=ReLU
):Upsample by scale from ni filters to nf (default ni), using nn.PixelShuffle.
The convolutional layer is initialized with icnr_init and passed act_cls and norm_type (the default of weight normalization seemed to be what’s best for super-resolution problems, in our experiments).
The blur option comes from Super-Resolution using Convolutional Neural Networks without Any Checkerboard Artifacts where the authors add a little bit of blur to completely get rid of checkerboard artifacts.
psfl = PixelShuffle_ICNR(16)
x = torch.randn(64, 16, 8, 8)
y = psfl(x)
test_eq(y.shape, [64, 16, 16, 16])
#ICNR init makes every 2x2 window (stride 2) have the same elements
for i in range(0,16,2):
for j in range(0,16,2):
test_eq(y[:,:,i,j],y[:,:,i+1,j])
test_eq(y[:,:,i,j],y[:,:,i ,j+1])
test_eq(y[:,:,i,j],y[:,:,i+1,j+1])psfl = PixelShuffle_ICNR(16, norm_type=None)
x = torch.randn(64, 16, 8, 8)
y = psfl(x)
test_eq(y.shape, [64, 16, 16, 16])
#ICNR init makes every 2x2 window (stride 2) have the same elements
for i in range(0,16,2):
for j in range(0,16,2):
test_eq(y[:,:,i,j],y[:,:,i+1,j])
test_eq(y[:,:,i,j],y[:,:,i ,j+1])
test_eq(y[:,:,i,j],y[:,:,i+1,j+1])psfl = PixelShuffle_ICNR(16, norm_type=NormType.Spectral)
x = torch.randn(64, 16, 8, 8)
y = psfl(x)
test_eq(y.shape, [64, 16, 16, 16])
#ICNR init makes every 2x2 window (stride 2) have the same elements
for i in range(0,16,2):
for j in range(0,16,2):
test_eq(y[:,:,i,j],y[:,:,i+1,j])
test_eq(y[:,:,i,j],y[:,:,i ,j+1])
test_eq(y[:,:,i,j],y[:,:,i+1,j+1])Sequential extensions
sequential
def sequential(
*args
):Create an nn.Sequential, wrapping items with Lambda if needed
SequentialEx
def SequentialEx(
*layers
):Like nn.Sequential, but with ModuleList semantics, and can access module input
This is useful to write layers that require to remember the input (like a resnet block) in a sequential way.
MergeLayer
def MergeLayer(
dense:bool=False
):Merge a shortcut with the result of the module by adding them or concatenating them if dense=True.
res_block = SequentialEx(ConvLayer(16, 16), ConvLayer(16,16))
res_block.append(MergeLayer()) # just to test append - normally it would be in init params
x = torch.randn(32, 16, 8, 8)
y = res_block(x)
test_eq(y.shape, [32, 16, 8, 8])
test_eq(y, x + res_block[1](res_block[0](x)))x = TensorBase(torch.randn(32, 16, 8, 8))
y = res_block(x)
test_is(y.orig, None)Concat
Equivalent to keras.layers.Concatenate, it will concat the outputs of a ModuleList over a given dimension (default the filter dimension)
layers = [ConvLayer(2,4), ConvLayer(2,4), ConvLayer(2,4)]
x = torch.rand(1,2,8,8)
cat = Cat(layers)
test_eq(cat(x).shape, [1,12,8,8])
test_eq(cat(x), torch.cat([l(x) for l in layers], dim=1))Ready-to-go models
SimpleCNN
def SimpleCNN(
filters, kernel_szs:NoneType=None, strides:NoneType=None, bn:bool=True
):Create a simple CNN with filters.
The model is a succession of convolutional layers from (filters[0],filters[1]) to (filters[n-2],filters[n-1]) (if n is the length of the filters list) followed by a PoolFlatten. kernel_szs and strides defaults to a list of 3s and a list of 2s. If bn=True the convolutional layers are successions of conv-relu-batchnorm, otherwise conv-relu.
tst = SimpleCNN([8,16,32])
mods = list(tst.children())
test_eq(len(mods), 3)
test_eq([[m[0].in_channels, m[0].out_channels] for m in mods[:2]], [[8,16], [16,32]])Test kernel sizes
tst = SimpleCNN([8,16,32], kernel_szs=[1,3])
mods = list(tst.children())
test_eq([m[0].kernel_size for m in mods[:2]], [(1,1), (3,3)])Test strides
tst = SimpleCNN([8,16,32], strides=[1,2])
mods = list(tst.children())
test_eq([m[0].stride for m in mods[:2]], [(1,1),(2,2)])ProdLayer
def ProdLayer():Merge a shortcut with the result of the module by multiplying them.
ResBlock
def ResBlock(
expansion, ni, nf, stride:int=1, groups:int=1, reduction:NoneType=None, nh1:NoneType=None, nh2:NoneType=None,
dw:bool=False, g2:int=1, sa:bool=False, sym:bool=False, norm_type:NormType=<NormType.Batch: 1>,
act_cls:type=ReLU, ndim:int=2, ks:int=3, pool:function=AvgPool, pool_first:bool=True, padding:NoneType=None,
bias:NoneType=None, bn_1st:bool=True, transpose:bool=False, init:str='auto', xtra:NoneType=None,
bias_std:float=0.01, dilation:Union=1, padding_mode:Literal='zeros', device:NoneType=None, dtype:NoneType=None
):Resnet block from ni to nh with stride
This is a resnet block (normal or bottleneck depending on expansion, 1 for the normal block and 4 for the traditional bottleneck) that implements the tweaks from Bag of Tricks for Image Classification with Convolutional Neural Networks. In particular, the last batchnorm layer (if that is the selected norm_type) is initialized with a weight (or gamma) of zero to facilitate the flow from the beginning to the end of the network. It also implements optional Squeeze and Excitation and grouped convs for ResNeXT and similar models (use dw=True for depthwise convs).
The kwargs are passed to ConvLayer along with norm_type.
SEBlock
def SEBlock(
expansion, ni, nf, groups:int=1, reduction:int=16, stride:int=1, **kwargs
):Call self as a function.
SEResNeXtBlock
def SEResNeXtBlock(
expansion, ni, nf, groups:int=32, reduction:int=16, stride:int=1, base_width:int=4, **kwargs
):Call self as a function.
SeparableBlock
def SeparableBlock(
expansion, ni, nf, reduction:int=16, stride:int=1, base_width:int=4, **kwargs
):Call self as a function.
Time Distributed Layer
Equivalent to Keras TimeDistributed Layer, enables computing pytorch Module over an axis.
bs, seq_len = 2, 5
x, y = torch.rand(bs,seq_len,3,2,2), torch.rand(bs,seq_len,3,2,2)tconv = TimeDistributed(nn.Conv2d(3,4,1))
test_eq(tconv(x).shape, (2,5,4,2,2))
tconv.low_mem=True
test_eq(tconv(x).shape, (2,5,4,2,2))class Mod(Module):
def __init__(self):
self.conv = nn.Conv2d(3,4,1)
def forward(self, x, y):
return self.conv(x) + self.conv(y)
tmod = TimeDistributed(Mod())out = tmod(x,y)
test_eq(out.shape, (2,5,4,2,2))
tmod.low_mem=True
out_low_mem = tmod(x,y)
test_eq(out_low_mem.shape, (2,5,4,2,2))
test_eq(out, out_low_mem)class Mod2(Module):
def __init__(self):
self.conv = nn.Conv2d(3,4,1)
def forward(self, x, y):
return self.conv(x), self.conv(y)
tmod2 = TimeDistributed(Mod2())out = tmod2(x,y)
test_eq(len(out), 2)
test_eq(out[0].shape, (2,5,4,2,2))
tmod2.low_mem=True
out_low_mem = tmod2(x,y)
test_eq(out_low_mem[0].shape, (2,5,4,2,2))
test_eq(out, out_low_mem)TimeDistributed
def TimeDistributed(
module, low_mem:bool=False, tdim:int=1
):Applies module over tdim identically for each step, use low_mem to compute one at a time.
This module is equivalent to Keras TimeDistributed Layer. This wrapper allows to apply a layer to every temporal slice of an input. By default it is assumed the time axis (tdim) is the 1st one (the one after the batch size). A typical usage would be to encode a sequence of images using an image encoder.
The forward function of TimeDistributed supports *args and **kkwargs but only args will be split and passed to the underlying module independently for each timestep, kwargs will be passed as they are. This is useful when you have module that take multiple arguments as inputs, this way, you can put all tensors you need spliting as args and other arguments that don’t need split as kwargs.
This module is heavy on memory, as it will try to pass mutiple timesteps at the same time on the batch dimension, if you get out of memorey errors, try first reducing your batch size by the number of timesteps.
from fastai.vision.all import *encoder = create_body(resnet18())A resnet18 will encode a feature map of 512 channels. Height and Width will be divided by 32.
time_resnet = TimeDistributed(encoder)a synthetic batch of 2 image-sequences of lenght 5. (bs, seq_len, ch, w, h)
image_sequence = torch.rand(2, 5, 3, 64, 64)time_resnet(image_sequence).shapetorch.Size([2, 5, 512, 2, 2])This way, one can encode a sequence of images on feature space. There is also a low_mem_forward that will pass images one at a time to reduce GPU memory consumption.
time_resnet.low_mem_forward(image_sequence).shapetorch.Size([2, 5, 512, 2, 2])Swish and Mish
SwishJit
def SwishJit():Same as nn.Module, but no need for subclasses to call super().__init__
MishJitAutoFn
def MishJitAutoFn(
*args, **kwargs
):Base class to create custom autograd.Function.
To create a custom autograd.Function, subclass this class and implement the :meth:forward and :meth:backward static methods. Then, to use your custom op in the forward pass, call the class method [apply](https://docs.fast.ai/torch_core.html#apply). Do not call :meth:forward directly.
To ensure correctness and best performance, make sure you are calling the correct methods on ctx and validating your backward function using :func:torch.autograd.gradcheck.
See :ref:extending-autograd for more details on how to use this class.
Examples::
>>> # xdoctest: +REQUIRES(env:TORCH_DOCTEST_AUTOGRAD)
>>> class Exp(Function):
>>> @staticmethod
>>> def forward(ctx, i):
>>> result = i.exp()
>>> ctx.save_for_backward(result)
>>> return result
>>>
>>> @staticmethod
>>> def backward(ctx, grad_output):
>>> result, = ctx.saved_tensors
>>> return grad_output * result
>>>
>>> # Use it by calling the apply method:
>>> # xdoctest: +SKIP
>>> output = Exp.apply(input)MishJit
def MishJit():Same as nn.Module, but no need for subclasses to call super().__init__
Helper functions for submodules
It’s easy to get the list of all parameters of a given model. For when you want all submodules (like linear/conv layers) without forgetting lone parameters, the following class wraps those in fake modules.
children_and_parameters
def children_and_parameters(
m
):Return the children of m and its direct parameters not registered in modules.
class TstModule(Module):
def __init__(self): self.a,self.lin = nn.Parameter(torch.randn(1)),nn.Linear(5,10)
tst = TstModule()
children = children_and_parameters(tst)
test_eq(len(children), 2)
test_eq(children[0], tst.lin)
assert isinstance(children[1], ParameterModule)
test_eq(children[1].val, tst.a)class A(Module): pass
assert not has_children(A())
assert has_children(TstModule())tst = nn.Sequential(TstModule(), TstModule())
children = flatten_model(tst)
test_eq(len(children), 4)
assert isinstance(children[1], ParameterModule)
assert isinstance(children[3], ParameterModule)NoneReduce
def NoneReduce(
loss_func
):A context manager to evaluate loss_func with none reduce.
x,y = torch.randn(5),torch.randn(5)
loss_fn = nn.MSELoss()
with NoneReduce(loss_fn) as loss_func:
loss = loss_func(x,y)
test_eq(loss.shape, [5])
test_eq(loss_fn.reduction, 'mean')
loss_fn = F.mse_loss
with NoneReduce(loss_fn) as loss_func:
loss = loss_func(x,y)
test_eq(loss.shape, [5])
test_eq(loss_fn, F.mse_loss)test_eq(in_channels(nn.Sequential(nn.Conv2d(5,4,3), nn.Conv2d(4,3,3))), 5)
test_eq(in_channels(nn.Sequential(nn.AvgPool2d(4), nn.Conv2d(4,3,3))), 4)
test_eq(in_channels(nn.Sequential(BatchNorm(4), nn.Conv2d(4,3,3))), 4)
test_eq(in_channels(nn.Sequential(InstanceNorm(4), nn.Conv2d(4,3,3))), 4)
test_eq(in_channels(nn.Sequential(InstanceNorm(4, affine=False), nn.Conv2d(4,3,3))), 4)
with expect_fail(): in_channels(nn.Sequential(nn.AvgPool2d(4)))