jdyjjj
946 lines
39 KiB
Python
946 lines
39 KiB
Python
import torch
|
|
import torch.nn as nn
|
|
from ..base_model import BaseModel
|
|
from ..modules import HorizontalPoolingPyramid, PackSequenceWrapper, SeparateFCs, SeparateBNNecks, SetBlockWrapper, ParallelBN1d
|
|
|
|
# ******* Copy from https://github.com/haofanwang/video-swin-transformer-pytorch/blob/main/video_swin_transformer.py *******
|
|
from functools import reduce, lru_cache
|
|
from operator import mul
|
|
from einops import rearrange
|
|
|
|
import torch.nn.functional as F
|
|
class Mlp(nn.Module):
|
|
""" Multilayer perceptron."""
|
|
|
|
def __init__(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.):
|
|
super().__init__()
|
|
out_features = out_features or in_features
|
|
hidden_features = hidden_features or in_features
|
|
self.fc1 = nn.Linear(in_features, hidden_features)
|
|
self.act = act_layer()
|
|
self.fc2 = nn.Linear(hidden_features, out_features)
|
|
self.drop = nn.Dropout(drop)
|
|
|
|
def forward(self, x):
|
|
x = self.fc1(x)
|
|
x = self.act(x)
|
|
x = self.drop(x)
|
|
x = self.fc2(x)
|
|
x = self.drop(x)
|
|
return x
|
|
|
|
|
|
def window_partition(x, window_size):
|
|
"""
|
|
Args:
|
|
x: (B, D, H, W, C)
|
|
window_size (tuple[int]): window size
|
|
Returns:
|
|
windows: (B*num_windows, window_size*window_size, C)
|
|
"""
|
|
B, D, H, W, C = x.shape
|
|
x = x.view(B, D // window_size[0], window_size[0], H // window_size[1], window_size[1], W // window_size[2], window_size[2], C)
|
|
windows = x.permute(0, 1, 3, 5, 2, 4, 6, 7).contiguous().view(-1, reduce(mul, window_size), C)
|
|
return windows
|
|
|
|
|
|
def window_reverse(windows, window_size, B, D, H, W):
|
|
"""
|
|
Args:
|
|
windows: (B*num_windows, window_size, window_size, C)
|
|
window_size (tuple[int]): Window size
|
|
H (int): Height of image
|
|
W (int): Width of image
|
|
Returns:
|
|
x: (B, D, H, W, C)
|
|
"""
|
|
x = windows.view(B, D // window_size[0], H // window_size[1], W // window_size[2], window_size[0], window_size[1], window_size[2], -1)
|
|
x = x.permute(0, 1, 4, 2, 5, 3, 6, 7).contiguous().view(B, D, H, W, -1)
|
|
return x
|
|
|
|
|
|
def get_window_size(x_size, window_size, shift_size=None):
|
|
use_window_size = list(window_size)
|
|
if shift_size is not None:
|
|
use_shift_size = list(shift_size)
|
|
for i in range(len(x_size)):
|
|
if x_size[i] <= window_size[i]:
|
|
use_window_size[i] = x_size[i]
|
|
if shift_size is not None:
|
|
use_shift_size[i] = 0
|
|
|
|
if shift_size is None:
|
|
return tuple(use_window_size)
|
|
else:
|
|
return tuple(use_window_size), tuple(use_shift_size)
|
|
|
|
|
|
from torch.nn.init import _calculate_fan_in_and_fan_out
|
|
|
|
|
|
import math
|
|
import warnings
|
|
# Copy from https://github.com/rwightman/pytorch-image-models/blob/8ff45e41f7a6aba4d5fdadee7dc3b7f2733df045/timm/models/layers/weight_init.py
|
|
def _trunc_normal_(tensor, mean, std, a, b):
|
|
# Cut & paste from PyTorch official master until it's in a few official releases - RW
|
|
# Method based on https://people.sc.fsu.edu/~jburkardt/presentations/truncated_normal.pdf
|
|
def norm_cdf(x):
|
|
# Computes standard normal cumulative distribution function
|
|
return (1. + math.erf(x / math.sqrt(2.))) / 2.
|
|
|
|
if (mean < a - 2 * std) or (mean > b + 2 * std):
|
|
warnings.warn("mean is more than 2 std from [a, b] in nn.init.trunc_normal_. "
|
|
"The distribution of values may be incorrect.",
|
|
stacklevel=2)
|
|
|
|
# Values are generated by using a truncated uniform distribution and
|
|
# then using the inverse CDF for the normal distribution.
|
|
# Get upper and lower cdf values
|
|
l = norm_cdf((a - mean) / std)
|
|
u = norm_cdf((b - mean) / std)
|
|
|
|
# Uniformly fill tensor with values from [l, u], then translate to
|
|
# [2l-1, 2u-1].
|
|
tensor.uniform_(2 * l - 1, 2 * u - 1)
|
|
|
|
# Use inverse cdf transform for normal distribution to get truncated
|
|
# standard normal
|
|
tensor.erfinv_()
|
|
|
|
# Transform to proper mean, std
|
|
tensor.mul_(std * math.sqrt(2.))
|
|
tensor.add_(mean)
|
|
|
|
# Clamp to ensure it's in the proper range
|
|
tensor.clamp_(min=a, max=b)
|
|
return tensor
|
|
|
|
# Copy from https://github.com/rwightman/pytorch-image-models/blob/8ff45e41f7a6aba4d5fdadee7dc3b7f2733df045/timm/models/layers/weight_init.py
|
|
def trunc_normal_(tensor, mean=0., std=1., a=-2., b=2.):
|
|
# type: (Tensor, float, float, float, float) -> Tensor
|
|
r"""Fills the input Tensor with values drawn from a truncated
|
|
normal distribution. The values are effectively drawn from the
|
|
normal distribution :math:`\mathcal{N}(\text{mean}, \text{std}^2)`
|
|
with values outside :math:`[a, b]` redrawn until they are within
|
|
the bounds. The method used for generating the random values works
|
|
best when :math:`a \leq \text{mean} \leq b`.
|
|
NOTE: this impl is similar to the PyTorch trunc_normal_, the bounds [a, b] are
|
|
applied while sampling the normal with mean/std applied, therefore a, b args
|
|
should be adjusted to match the range of mean, std args.
|
|
Args:
|
|
tensor: an n-dimensional `torch.Tensor`
|
|
mean: the mean of the normal distribution
|
|
std: the standard deviation of the normal distribution
|
|
a: the minimum cutoff value
|
|
b: the maximum cutoff value
|
|
Examples:
|
|
>>> w = torch.empty(3, 5)
|
|
>>> nn.init.trunc_normal_(w)
|
|
"""
|
|
with torch.no_grad():
|
|
return _trunc_normal_(tensor, mean, std, a, b)
|
|
|
|
class WindowAttention3D(nn.Module):
|
|
""" Window based multi-head self attention (W-MSA) module with relative position bias.
|
|
It supports both of shifted and non-shifted window.
|
|
Args:
|
|
dim (int): Number of input channels.
|
|
window_size (tuple[int]): The temporal length, height and width of the window.
|
|
num_heads (int): Number of attention heads.
|
|
qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True
|
|
qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set
|
|
attn_drop (float, optional): Dropout ratio of attention weight. Default: 0.0
|
|
proj_drop (float, optional): Dropout ratio of output. Default: 0.0
|
|
"""
|
|
|
|
def __init__(self, dim, window_size, num_heads, qkv_bias=False, qk_scale=None, attn_drop=0., proj_drop=0.):
|
|
|
|
super().__init__()
|
|
self.dim = dim
|
|
self.window_size = window_size # Wd, Wh, Ww
|
|
self.num_heads = num_heads
|
|
head_dim = dim // num_heads
|
|
self.scale = qk_scale or head_dim ** -0.5
|
|
|
|
# define a parameter table of relative position bias
|
|
self.relative_position_bias_table = nn.Parameter(
|
|
torch.zeros((2 * window_size[0] - 1) * (2 * window_size[1] - 1) * (2 * window_size[2] - 1), num_heads)) # 2*Wd-1 * 2*Wh-1 * 2*Ww-1, nH
|
|
|
|
# get pair-wise relative position index for each token inside the window
|
|
coords_d = torch.arange(self.window_size[0])
|
|
coords_h = torch.arange(self.window_size[1])
|
|
coords_w = torch.arange(self.window_size[2])
|
|
coords = torch.stack(torch.meshgrid(coords_d, coords_h, coords_w)) # 3, Wd, Wh, Ww
|
|
coords_flatten = torch.flatten(coords, 1) # 3, Wd*Wh*Ww
|
|
relative_coords = coords_flatten[:, :, None] - coords_flatten[:, None, :] # 3, Wd*Wh*Ww, Wd*Wh*Ww
|
|
relative_coords = relative_coords.permute(1, 2, 0).contiguous() # Wd*Wh*Ww, Wd*Wh*Ww, 3
|
|
relative_coords[:, :, 0] += self.window_size[0] - 1 # shift to start from 0
|
|
relative_coords[:, :, 1] += self.window_size[1] - 1
|
|
relative_coords[:, :, 2] += self.window_size[2] - 1
|
|
|
|
relative_coords[:, :, 0] *= (2 * self.window_size[1] - 1) * (2 * self.window_size[2] - 1)
|
|
relative_coords[:, :, 1] *= (2 * self.window_size[2] - 1)
|
|
relative_position_index = relative_coords.sum(-1) # Wd*Wh*Ww, Wd*Wh*Ww
|
|
self.register_buffer("relative_position_index", relative_position_index)
|
|
|
|
self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias)
|
|
self.attn_drop = nn.Dropout(attn_drop)
|
|
self.proj = nn.Linear(dim, dim)
|
|
self.proj_drop = nn.Dropout(proj_drop)
|
|
|
|
trunc_normal_(self.relative_position_bias_table, std=.02)
|
|
self.softmax = nn.Softmax(dim=-1)
|
|
|
|
def forward(self, x, mask=None):
|
|
""" Forward function.
|
|
Args:
|
|
x: input features with shape of (num_windows*B, N, C)
|
|
mask: (0/-inf) mask with shape of (num_windows, N, N) or None
|
|
"""
|
|
B_, N, C = x.shape
|
|
qkv = self.qkv(x).reshape(B_, N, 3, self.num_heads, C // self.num_heads).permute(2, 0, 3, 1, 4)
|
|
q, k, v = qkv[0], qkv[1], qkv[2] # B_, nH, N, C
|
|
|
|
q = q * self.scale
|
|
attn = q @ k.transpose(-2, -1)
|
|
|
|
relative_position_bias = self.relative_position_bias_table[self.relative_position_index[:N, :N].reshape(-1)].reshape(
|
|
N, N, -1) # Wd*Wh*Ww,Wd*Wh*Ww,nH
|
|
relative_position_bias = relative_position_bias.permute(2, 0, 1).contiguous() # nH, Wd*Wh*Ww, Wd*Wh*Ww
|
|
attn = attn + relative_position_bias.unsqueeze(0) # B_, nH, N, N
|
|
|
|
if mask is not None:
|
|
nW = mask.shape[0]
|
|
attn = attn.view(B_ // nW, nW, self.num_heads, N, N) + mask.unsqueeze(1).unsqueeze(0)
|
|
attn = attn.view(-1, self.num_heads, N, N)
|
|
attn = self.softmax(attn)
|
|
else:
|
|
attn = self.softmax(attn)
|
|
|
|
attn = self.attn_drop(attn)
|
|
|
|
x = (attn @ v).transpose(1, 2).reshape(B_, N, C)
|
|
x = self.proj(x)
|
|
x = self.proj_drop(x)
|
|
return x
|
|
|
|
# Copy from https://github.com/rwightman/pytorch-image-models/blob/8ff45e41f7a6aba4d5fdadee7dc3b7f2733df045/timm/models/layers/drop.py
|
|
def drop_path(x, drop_prob: float = 0., training: bool = False, scale_by_keep: bool = True):
|
|
"""Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).
|
|
This is the same as the DropConnect impl I created for EfficientNet, etc networks, however,
|
|
the original name is misleading as 'Drop Connect' is a different form of dropout in a separate paper...
|
|
See discussion: https://github.com/tensorflow/tpu/issues/494#issuecomment-532968956 ... I've opted for
|
|
changing the layer and argument names to 'drop path' rather than mix DropConnect as a layer name and use
|
|
'survival rate' as the argument.
|
|
"""
|
|
if drop_prob == 0. or not training:
|
|
return x
|
|
keep_prob = 1 - drop_prob
|
|
shape = (x.shape[0],) + (1,) * (x.ndim - 1) # work with diff dim tensors, not just 2D ConvNets
|
|
random_tensor = x.new_empty(shape).bernoulli_(keep_prob)
|
|
if keep_prob > 0.0 and scale_by_keep:
|
|
random_tensor.div_(keep_prob)
|
|
return x * random_tensor
|
|
|
|
# Copy from https://github.com/rwightman/pytorch-image-models/blob/8ff45e41f7a6aba4d5fdadee7dc3b7f2733df045/timm/models/layers/drop.py
|
|
class DropPath(nn.Module):
|
|
"""Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).
|
|
"""
|
|
def __init__(self, drop_prob: float = 0., scale_by_keep: bool = True):
|
|
super(DropPath, self).__init__()
|
|
self.drop_prob = drop_prob
|
|
self.scale_by_keep = scale_by_keep
|
|
|
|
def forward(self, x):
|
|
return drop_path(x, self.drop_prob, self.training, self.scale_by_keep)
|
|
|
|
def extra_repr(self):
|
|
return f'drop_prob={round(self.drop_prob,3):0.3f}'
|
|
|
|
class SwinTransformerBlock3D(nn.Module):
|
|
""" Swin Transformer Block.
|
|
Args:
|
|
dim (int): Number of input channels.
|
|
num_heads (int): Number of attention heads.
|
|
window_size (tuple[int]): Window size.
|
|
shift_size (tuple[int]): Shift size for SW-MSA.
|
|
mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.
|
|
qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True
|
|
qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set.
|
|
drop (float, optional): Dropout rate. Default: 0.0
|
|
attn_drop (float, optional): Attention dropout rate. Default: 0.0
|
|
drop_path (float, optional): Stochastic depth rate. Default: 0.0
|
|
act_layer (nn.Module, optional): Activation layer. Default: nn.GELU
|
|
norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm
|
|
"""
|
|
|
|
def __init__(self, dim, num_heads, window_size=(2,7,7), shift_size=(0,0,0),
|
|
mlp_ratio=4., qkv_bias=True, qk_scale=None, drop=0., attn_drop=0., drop_path=0.,
|
|
act_layer=nn.GELU, norm_layer=nn.LayerNorm, use_checkpoint=False):
|
|
super().__init__()
|
|
self.dim = dim
|
|
self.num_heads = num_heads
|
|
self.window_size = window_size
|
|
self.shift_size = shift_size
|
|
self.mlp_ratio = mlp_ratio
|
|
self.use_checkpoint=use_checkpoint
|
|
|
|
assert 0 <= self.shift_size[0] < self.window_size[0], "shift_size must in 0-window_size"
|
|
assert 0 <= self.shift_size[1] < self.window_size[1], "shift_size must in 0-window_size"
|
|
assert 0 <= self.shift_size[2] < self.window_size[2], "shift_size must in 0-window_size"
|
|
|
|
self.norm1 = norm_layer(dim)
|
|
self.attn = WindowAttention3D(
|
|
dim, window_size=self.window_size, num_heads=num_heads,
|
|
qkv_bias=qkv_bias, qk_scale=qk_scale, attn_drop=attn_drop, proj_drop=drop)
|
|
|
|
self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
|
|
self.norm2 = norm_layer(dim)
|
|
mlp_hidden_dim = int(dim * mlp_ratio)
|
|
self.mlp = Mlp(in_features=dim, hidden_features=mlp_hidden_dim, act_layer=act_layer, drop=drop)
|
|
|
|
def forward_part1(self, x, mask_matrix):
|
|
B, D, H, W, C = x.shape
|
|
window_size, shift_size = get_window_size((D, H, W), self.window_size, self.shift_size)
|
|
|
|
x = self.norm1(x)
|
|
# pad feature maps to multiples of window size
|
|
pad_l = pad_t = pad_d0 = 0
|
|
pad_d1 = (window_size[0] - D % window_size[0]) % window_size[0]
|
|
pad_b = (window_size[1] - H % window_size[1]) % window_size[1]
|
|
pad_r = (window_size[2] - W % window_size[2]) % window_size[2]
|
|
x = F.pad(x, (0, 0, pad_l, pad_r, pad_t, pad_b, pad_d0, pad_d1))
|
|
_, Dp, Hp, Wp, _ = x.shape
|
|
# cyclic shift
|
|
if any(i > 0 for i in shift_size):
|
|
shifted_x = torch.roll(x, shifts=(-shift_size[0], -shift_size[1], -shift_size[2]), dims=(1, 2, 3))
|
|
attn_mask = mask_matrix
|
|
else:
|
|
shifted_x = x
|
|
attn_mask = None
|
|
# partition windows
|
|
x_windows = window_partition(shifted_x, window_size) # B*nW, Wd*Wh*Ww, C
|
|
# W-MSA/SW-MSA
|
|
attn_windows = self.attn(x_windows, mask=attn_mask) # B*nW, Wd*Wh*Ww, C
|
|
# merge windows
|
|
attn_windows = attn_windows.view(-1, *(window_size+(C,)))
|
|
shifted_x = window_reverse(attn_windows, window_size, B, Dp, Hp, Wp) # B D' H' W' C
|
|
# reverse cyclic shift
|
|
if any(i > 0 for i in shift_size):
|
|
x = torch.roll(shifted_x, shifts=(shift_size[0], shift_size[1], shift_size[2]), dims=(1, 2, 3))
|
|
else:
|
|
x = shifted_x
|
|
|
|
if pad_d1 >0 or pad_r > 0 or pad_b > 0:
|
|
x = x[:, :D, :H, :W, :].contiguous()
|
|
return x
|
|
|
|
def forward_part2(self, x):
|
|
return self.drop_path(self.mlp(self.norm2(x)))
|
|
|
|
def forward(self, x, mask_matrix):
|
|
""" Forward function.
|
|
Args:
|
|
x: Input feature, tensor size (B, D, H, W, C).
|
|
mask_matrix: Attention mask for cyclic shift.
|
|
"""
|
|
|
|
shortcut = x
|
|
if self.use_checkpoint:
|
|
x = checkpoint.checkpoint(self.forward_part1, x, mask_matrix)
|
|
else:
|
|
x = self.forward_part1(x, mask_matrix)
|
|
x = shortcut + self.drop_path(x)
|
|
|
|
if self.use_checkpoint:
|
|
x = x + checkpoint.checkpoint(self.forward_part2, x)
|
|
else:
|
|
x = x + self.forward_part2(x)
|
|
|
|
return x
|
|
|
|
|
|
class PatchMerging(nn.Module):
|
|
""" Patch Merging Layer
|
|
Args:
|
|
dim (int): Number of input channels.
|
|
norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm
|
|
"""
|
|
def __init__(self, dim, norm_layer=nn.LayerNorm):
|
|
super().__init__()
|
|
self.dim = dim
|
|
self.reduction = nn.Linear(4 * dim, 2 * dim, bias=False)
|
|
self.norm = norm_layer(4 * dim)
|
|
|
|
def forward(self, x):
|
|
""" Forward function.
|
|
Args:
|
|
x: Input feature, tensor size (B, D, H, W, C).
|
|
"""
|
|
B, D, H, W, C = x.shape
|
|
|
|
# padding
|
|
pad_input = (H % 2 == 1) or (W % 2 == 1)
|
|
if pad_input:
|
|
x = F.pad(x, (0, 0, 0, W % 2, 0, H % 2))
|
|
|
|
x0 = x[:, :, 0::2, 0::2, :] # B D H/2 W/2 C
|
|
x1 = x[:, :, 1::2, 0::2, :] # B D H/2 W/2 C
|
|
x2 = x[:, :, 0::2, 1::2, :] # B D H/2 W/2 C
|
|
x3 = x[:, :, 1::2, 1::2, :] # B D H/2 W/2 C
|
|
x = torch.cat([x0, x1, x2, x3], -1) # B D H/2 W/2 4*C
|
|
|
|
x = self.norm(x)
|
|
x = self.reduction(x)
|
|
|
|
return x
|
|
|
|
|
|
# cache each stage results
|
|
@lru_cache()
|
|
def compute_mask(D, H, W, window_size, shift_size, device):
|
|
img_mask = torch.zeros((1, D, H, W, 1), device=device) # 1 Dp Hp Wp 1
|
|
cnt = 0
|
|
for d in slice(-window_size[0]), slice(-window_size[0], -shift_size[0]), slice(-shift_size[0],None):
|
|
for h in slice(-window_size[1]), slice(-window_size[1], -shift_size[1]), slice(-shift_size[1],None):
|
|
for w in slice(-window_size[2]), slice(-window_size[2], -shift_size[2]), slice(-shift_size[2],None):
|
|
img_mask[:, d, h, w, :] = cnt
|
|
cnt += 1
|
|
mask_windows = window_partition(img_mask, window_size) # nW, ws[0]*ws[1]*ws[2], 1
|
|
mask_windows = mask_windows.squeeze(-1) # nW, ws[0]*ws[1]*ws[2]
|
|
attn_mask = mask_windows.unsqueeze(1) - mask_windows.unsqueeze(2)
|
|
attn_mask = attn_mask.masked_fill(attn_mask != 0, float(-100.0)).masked_fill(attn_mask == 0, float(0.0))
|
|
return attn_mask
|
|
|
|
import numpy as np
|
|
class BasicLayer(nn.Module):
|
|
""" A basic Swin Transformer layer for one stage.
|
|
Args:
|
|
dim (int): Number of feature channels
|
|
depth (int): Depths of this stage.
|
|
num_heads (int): Number of attention head.
|
|
window_size (tuple[int]): Local window size. Default: (1,7,7).
|
|
mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. Default: 4.
|
|
qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True
|
|
qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set.
|
|
drop (float, optional): Dropout rate. Default: 0.0
|
|
attn_drop (float, optional): Attention dropout rate. Default: 0.0
|
|
drop_path (float | tuple[float], optional): Stochastic depth rate. Default: 0.0
|
|
norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm
|
|
downsample (nn.Module | None, optional): Downsample layer at the end of the layer. Default: None
|
|
"""
|
|
|
|
def __init__(self,
|
|
dim,
|
|
depth,
|
|
num_heads,
|
|
window_size=(1,7,7),
|
|
mlp_ratio=4.,
|
|
qkv_bias=False,
|
|
qk_scale=None,
|
|
drop=0.,
|
|
attn_drop=0.,
|
|
drop_path=0.,
|
|
norm_layer=nn.LayerNorm,
|
|
downsample=None,
|
|
use_checkpoint=False):
|
|
super().__init__()
|
|
self.window_size = window_size
|
|
self.shift_size = tuple(i // 2 for i in window_size)
|
|
self.depth = depth
|
|
self.use_checkpoint = use_checkpoint
|
|
|
|
# build blocks
|
|
self.blocks = nn.ModuleList([
|
|
SwinTransformerBlock3D(
|
|
dim=dim,
|
|
num_heads=num_heads,
|
|
window_size=window_size,
|
|
shift_size=(0,0,0) if (i % 2 == 0) else self.shift_size,
|
|
mlp_ratio=mlp_ratio,
|
|
qkv_bias=qkv_bias,
|
|
qk_scale=qk_scale,
|
|
drop=drop,
|
|
attn_drop=attn_drop,
|
|
drop_path=drop_path[i] if isinstance(drop_path, list) else drop_path,
|
|
norm_layer=norm_layer,
|
|
use_checkpoint=use_checkpoint,
|
|
)
|
|
for i in range(depth)]
|
|
)
|
|
|
|
self.downsample = downsample
|
|
if self.downsample == False:
|
|
self.downsample = lambda x: x
|
|
else:
|
|
if self.downsample is not None:
|
|
self.downsample = downsample(dim=dim, norm_layer=norm_layer)
|
|
else:
|
|
self.downsample = nn.Sequential(
|
|
norm_layer(dim),
|
|
nn.Linear(dim, 2 * dim, bias=False)
|
|
)
|
|
|
|
def forward(self, x):
|
|
""" Forward function.
|
|
Args:
|
|
x: Input feature, tensor size (B, C, D, H, W).
|
|
"""
|
|
# calculate attention mask for SW-MSA
|
|
B, C, D, H, W = x.shape
|
|
window_size, shift_size = get_window_size((D,H,W), self.window_size, self.shift_size)
|
|
x = rearrange(x, 'b c d h w -> b d h w c')
|
|
Dp = int(np.ceil(D / window_size[0])) * window_size[0]
|
|
Hp = int(np.ceil(H / window_size[1])) * window_size[1]
|
|
Wp = int(np.ceil(W / window_size[2])) * window_size[2]
|
|
attn_mask = compute_mask(Dp, Hp, Wp, window_size, shift_size, x.device)
|
|
for blk in self.blocks:
|
|
x = blk(x, attn_mask)
|
|
x = x.view(B, D, H, W, -1)
|
|
|
|
if self.downsample is not None:
|
|
x = self.downsample(x)
|
|
x = rearrange(x, 'b d h w c -> b c d h w')
|
|
return x
|
|
|
|
|
|
class PatchEmbed3D(nn.Module):
|
|
""" Video to Patch Embedding.
|
|
Args:
|
|
patch_size (int): Patch token size. Default: (2,4,4).
|
|
in_chans (int): Number of input video channels. Default: 3.
|
|
embed_dim (int): Number of linear projection output channels. Default: 96.
|
|
norm_layer (nn.Module, optional): Normalization layer. Default: None
|
|
"""
|
|
def __init__(self, patch_size=(2,4,4), in_chans=3, embed_dim=96, norm_layer=None):
|
|
super().__init__()
|
|
self.patch_size = patch_size
|
|
|
|
self.in_chans = in_chans
|
|
self.embed_dim = embed_dim
|
|
|
|
self.proj = nn.Conv3d(in_chans, embed_dim, kernel_size=patch_size, stride=patch_size)
|
|
if norm_layer is not None:
|
|
self.norm = norm_layer(embed_dim)
|
|
else:
|
|
self.norm = None
|
|
|
|
def forward(self, x):
|
|
"""Forward function."""
|
|
# padding
|
|
_, _, D, H, W = x.size()
|
|
if W % self.patch_size[2] != 0:
|
|
x = F.pad(x, (0, self.patch_size[2] - W % self.patch_size[2]))
|
|
if H % self.patch_size[1] != 0:
|
|
x = F.pad(x, (0, 0, 0, self.patch_size[1] - H % self.patch_size[1]))
|
|
if D % self.patch_size[0] != 0:
|
|
x = F.pad(x, (0, 0, 0, 0, 0, self.patch_size[0] - D % self.patch_size[0]))
|
|
|
|
x = self.proj(x) # B C D Wh Ww
|
|
if self.norm is not None:
|
|
D, Wh, Ww = x.size(2), x.size(3), x.size(4)
|
|
x = x.flatten(2).transpose(1, 2)
|
|
x = self.norm(x)
|
|
x = x.transpose(1, 2).view(-1, self.embed_dim, D, Wh, Ww)
|
|
|
|
return x
|
|
|
|
class SwinTransformer3D(nn.Module):
|
|
""" Swin Transformer backbone.
|
|
A PyTorch impl of : `Swin Transformer: Hierarchical Vision Transformer using Shifted Windows` -
|
|
https://arxiv.org/pdf/2103.14030
|
|
Args:
|
|
patch_size (int | tuple(int)): Patch size. Default: (4,4,4).
|
|
in_chans (int): Number of input image channels. Default: 3.
|
|
embed_dim (int): Number of linear projection output channels. Default: 96.
|
|
depths (tuple[int]): Depths of each Swin Transformer stage.
|
|
num_heads (tuple[int]): Number of attention head of each stage.
|
|
window_size (int): Window size. Default: 7.
|
|
mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. Default: 4.
|
|
qkv_bias (bool): If True, add a learnable bias to query, key, value. Default: Truee
|
|
qk_scale (float): Override default qk scale of head_dim ** -0.5 if set.
|
|
drop_rate (float): Dropout rate.
|
|
attn_drop_rate (float): Attention dropout rate. Default: 0.
|
|
drop_path_rate (float): Stochastic depth rate. Default: 0.2.
|
|
norm_layer: Normalization layer. Default: nn.LayerNorm.
|
|
patch_norm (bool): If True, add normalization after patch embedding. Default: False.
|
|
frozen_stages (int): Stages to be frozen (stop grad and set eval mode).
|
|
-1 means not freezing any parameters.
|
|
"""
|
|
|
|
def __init__(self,
|
|
pretrained=None,
|
|
pretrained2d=True,
|
|
patch_size=(4,4,4),
|
|
in_chans=3,
|
|
embed_dim=96,
|
|
depths=[2, 2, 6, 2],
|
|
num_heads=[3, 6, 12, 24],
|
|
window_size=(2,7,7),
|
|
mlp_ratio=4.,
|
|
qkv_bias=True,
|
|
qk_scale=None,
|
|
drop_rate=0.,
|
|
attn_drop_rate=0.,
|
|
drop_path_rate=0.2,
|
|
norm_layer=nn.LayerNorm,
|
|
patch_norm=False,
|
|
frozen_stages=-1,
|
|
use_checkpoint=False,
|
|
downsample=[1, 2, 2, 1]):
|
|
super().__init__()
|
|
|
|
self.pretrained = pretrained
|
|
self.pretrained2d = pretrained2d
|
|
self.num_layers = len(depths)
|
|
self.embed_dim = embed_dim
|
|
self.patch_norm = patch_norm
|
|
self.frozen_stages = frozen_stages
|
|
self.window_size = window_size
|
|
self.patch_size = patch_size
|
|
|
|
# split image into non-overlapping patches
|
|
self.patch_embed = PatchEmbed3D(
|
|
patch_size=patch_size, in_chans=in_chans, embed_dim=embed_dim,
|
|
norm_layer=norm_layer if self.patch_norm else None)
|
|
|
|
self.pos_drop = nn.Dropout(p=drop_rate)
|
|
|
|
# stochastic depth
|
|
dpr = [x.item() for x in torch.linspace(0, drop_path_rate, sum(depths))] # stochastic depth decay rule
|
|
|
|
# build layers
|
|
self.layers = nn.ModuleList()
|
|
for i_layer in range(self.num_layers):
|
|
if downsample[i_layer]== 2:
|
|
dfunc = PatchMerging
|
|
elif downsample[i_layer] == 1:
|
|
dfunc = None
|
|
elif downsample[i_layer] == 0:
|
|
dfunc = False
|
|
else:
|
|
raise ValueError('xxx')
|
|
|
|
layer = BasicLayer(
|
|
dim=int(embed_dim * 2**i_layer),
|
|
depth=depths[i_layer],
|
|
num_heads=num_heads[i_layer],
|
|
window_size=window_size,
|
|
mlp_ratio=mlp_ratio,
|
|
qkv_bias=qkv_bias,
|
|
qk_scale=qk_scale,
|
|
drop=drop_rate,
|
|
attn_drop=attn_drop_rate,
|
|
drop_path=dpr[sum(depths[:i_layer]):sum(depths[:i_layer + 1])],
|
|
norm_layer=norm_layer,
|
|
downsample=dfunc,
|
|
use_checkpoint=use_checkpoint)
|
|
|
|
self.layers.append(layer)
|
|
|
|
# self.num_features = int(embed_dim * 2**self.num_layers)
|
|
self.num_features = 512
|
|
|
|
# add a norm layer for each output
|
|
self.norm = norm_layer(self.num_features)
|
|
|
|
self._freeze_stages()
|
|
|
|
def _freeze_stages(self):
|
|
if self.frozen_stages >= 0:
|
|
self.patch_embed.eval()
|
|
for param in self.patch_embed.parameters():
|
|
param.requires_grad = False
|
|
|
|
if self.frozen_stages >= 1:
|
|
self.pos_drop.eval()
|
|
for i in range(0, self.frozen_stages):
|
|
m = self.layers[i]
|
|
m.eval()
|
|
for param in m.parameters():
|
|
param.requires_grad = False
|
|
|
|
def inflate_weights(self, logger):
|
|
"""Inflate the swin2d parameters to swin3d.
|
|
The differences between swin3d and swin2d mainly lie in an extra
|
|
axis. To utilize the pretrained parameters in 2d model,
|
|
the weight of swin2d models should be inflated to fit in the shapes of
|
|
the 3d counterpart.
|
|
Args:
|
|
logger (logging.Logger): The logger used to print
|
|
debugging infomation.
|
|
"""
|
|
checkpoint = torch.load(self.pretrained, map_location='cpu')
|
|
state_dict = checkpoint['model']
|
|
|
|
# delete relative_position_index since we always re-init it
|
|
relative_position_index_keys = [k for k in state_dict.keys() if "relative_position_index" in k]
|
|
for k in relative_position_index_keys:
|
|
del state_dict[k]
|
|
|
|
# delete attn_mask since we always re-init it
|
|
attn_mask_keys = [k for k in state_dict.keys() if "attn_mask" in k]
|
|
for k in attn_mask_keys:
|
|
del state_dict[k]
|
|
|
|
state_dict['patch_embed.proj.weight'] = state_dict['patch_embed.proj.weight'].unsqueeze(2).repeat(1,1,self.patch_size[0],1,1) / self.patch_size[0]
|
|
|
|
# bicubic interpolate relative_position_bias_table if not match
|
|
relative_position_bias_table_keys = [k for k in state_dict.keys() if "relative_position_bias_table" in k]
|
|
for k in relative_position_bias_table_keys:
|
|
relative_position_bias_table_pretrained = state_dict[k]
|
|
relative_position_bias_table_current = self.state_dict()[k]
|
|
L1, nH1 = relative_position_bias_table_pretrained.size()
|
|
L2, nH2 = relative_position_bias_table_current.size()
|
|
L2 = (2*self.window_size[1]-1) * (2*self.window_size[2]-1)
|
|
wd = self.window_size[0]
|
|
if nH1 != nH2:
|
|
logger.warning(f"Error in loading {k}, passing")
|
|
else:
|
|
if L1 != L2:
|
|
S1 = int(L1 ** 0.5)
|
|
relative_position_bias_table_pretrained_resized = torch.nn.functional.interpolate(
|
|
relative_position_bias_table_pretrained.permute(1, 0).view(1, nH1, S1, S1), size=(2*self.window_size[1]-1, 2*self.window_size[2]-1),
|
|
mode='bicubic')
|
|
relative_position_bias_table_pretrained = relative_position_bias_table_pretrained_resized.view(nH2, L2).permute(1, 0)
|
|
state_dict[k] = relative_position_bias_table_pretrained.repeat(2*wd-1,1)
|
|
|
|
msg = self.load_state_dict(state_dict, strict=False)
|
|
logger.info(msg)
|
|
logger.info(f"=> loaded successfully '{self.pretrained}'")
|
|
del checkpoint
|
|
torch.cuda.empty_cache()
|
|
|
|
def init_weights(self, pretrained=None):
|
|
"""Initialize the weights in backbone.
|
|
Args:
|
|
pretrained (str, optional): Path to pre-trained weights.
|
|
Defaults to None.
|
|
"""
|
|
def _init_weights(m):
|
|
if isinstance(m, nn.Linear):
|
|
trunc_normal_(m.weight, std=.02)
|
|
if isinstance(m, nn.Linear) and m.bias is not None:
|
|
nn.init.constant_(m.bias, 0)
|
|
elif isinstance(m, nn.LayerNorm):
|
|
nn.init.constant_(m.bias, 0)
|
|
nn.init.constant_(m.weight, 1.0)
|
|
|
|
if pretrained:
|
|
self.pretrained = pretrained
|
|
if isinstance(self.pretrained, str):
|
|
self.apply(_init_weights)
|
|
logger = get_root_logger()
|
|
logger.info(f'load model from: {self.pretrained}')
|
|
|
|
if self.pretrained2d:
|
|
# Inflate 2D model into 3D model.
|
|
self.inflate_weights(logger)
|
|
else:
|
|
# Directly load 3D model.
|
|
load_checkpoint(self, self.pretrained, strict=False, logger=logger)
|
|
elif self.pretrained is None:
|
|
self.apply(_init_weights)
|
|
else:
|
|
raise TypeError('pretrained must be a str or None')
|
|
|
|
def forward(self, x):
|
|
"""Forward function."""
|
|
x = self.patch_embed(x)
|
|
|
|
x = self.pos_drop(x)
|
|
|
|
for layer in self.layers:
|
|
x = layer(x.contiguous())
|
|
|
|
x = rearrange(x, 'n c d h w -> n d h w c')
|
|
x = self.norm(x)
|
|
x = rearrange(x, 'n d h w c -> n c d h w')
|
|
|
|
return x
|
|
|
|
def train(self, mode=True):
|
|
"""Convert the model into training mode while keep layers freezed."""
|
|
super(SwinTransformer3D, self).train(mode)
|
|
self._freeze_stages()
|
|
|
|
# ******* Copy from https://github.com/haofanwang/video-swin-transformer-pytorch/blob/main/video_swin_transformer.py *******
|
|
|
|
def conv3x3(in_planes, out_planes, stride=1, groups=1, dilation=1):
|
|
"""3x3 convolution with padding"""
|
|
return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride,
|
|
padding=dilation, groups=groups, bias=False, dilation=dilation)
|
|
|
|
def conv1x1(in_planes, out_planes, stride=1):
|
|
"""1x1 convolution"""
|
|
return nn.Conv2d(in_planes, out_planes, kernel_size=1, stride=stride, bias=False)
|
|
|
|
# import copy
|
|
|
|
from ..modules import BasicBlock2D, BasicBlockP3D
|
|
|
|
import torch.optim as optim
|
|
import os.path as osp
|
|
from collections import OrderedDict
|
|
from utils import get_valid_args, get_attr_from
|
|
|
|
class SwinGait(BaseModel):
|
|
def __init__(self, cfgs, training):
|
|
self.T_max_iter = cfgs['trainer_cfg']['T_max_iter']
|
|
super(SwinGait, self).__init__(cfgs, training=training)
|
|
|
|
def build_network(self, model_cfg):
|
|
channels = model_cfg['Backbone']['channels']
|
|
layers = model_cfg['Backbone']['layers']
|
|
in_c = model_cfg['Backbone']['in_channels']
|
|
|
|
self.inplanes = channels[0]
|
|
self.layer0 = SetBlockWrapper(nn.Sequential(
|
|
conv3x3(in_c, self.inplanes, 1),
|
|
nn.BatchNorm2d(self.inplanes),
|
|
nn.ReLU(inplace=True)
|
|
))
|
|
self.layer1 = SetBlockWrapper(self.make_layer(BasicBlock2D, channels[0], stride=[1, 1], blocks_num=layers[0], mode='2d'))
|
|
self.layer2 = self.make_layer(BasicBlockP3D, channels[1], stride=[2, 2], blocks_num=layers[1], mode='p3d')
|
|
|
|
self.ulayer = SetBlockWrapper(nn.UpsamplingBilinear2d(size=(30, 20)))
|
|
|
|
self.transformer = SwinTransformer3D(
|
|
patch_size = [1, 2, 2],
|
|
in_chans = channels[1],
|
|
embed_dim = 256,
|
|
depths = [layers[2], layers[3]],
|
|
num_heads = [16, 32],
|
|
window_size = [3, 3, 5],
|
|
downsample = [1, 0],
|
|
drop_path_rate = 0.1,
|
|
patch_norm = True,
|
|
)
|
|
|
|
self.FCs = SeparateFCs(model_cfg['SeparateBNNecks']['parts_num'], in_channels=512, out_channels=256)
|
|
self.BNNecks = SeparateBNNecks(**model_cfg['SeparateBNNecks'])
|
|
self.TP = PackSequenceWrapper(torch.max)
|
|
self.HPP = HorizontalPoolingPyramid(bin_num=model_cfg['bin_num'])
|
|
|
|
def get_optimizer(self, optimizer_cfg):
|
|
self.msg_mgr.log_info(optimizer_cfg)
|
|
optimizer = get_attr_from([optim], optimizer_cfg['solver'])
|
|
valid_arg = get_valid_args(optimizer, optimizer_cfg, ['solver'])
|
|
|
|
transformer_no_decay = ['patch_embed', 'norm', 'relative_position_bias_table']
|
|
transformer_params = list(self.transformer.named_parameters())
|
|
params_list = [
|
|
{'params': [p for n, p in transformer_params if any(nd in n for nd in transformer_no_decay)], 'lr': optimizer_cfg['lr'], 'weight_decay': 0.},
|
|
{'params': [p for n, p in transformer_params if not any(nd in n for nd in transformer_no_decay)], 'lr': optimizer_cfg['lr'], 'weight_decay': optimizer_cfg['weight_decay']},
|
|
{'params': self.FCs.parameters(), 'lr': optimizer_cfg['lr'] * 0.1, 'weight_decay': optimizer_cfg['weight_decay']},
|
|
{'params': self.BNNecks.parameters(), 'lr': optimizer_cfg['lr'] * 0.1, 'weight_decay': optimizer_cfg['weight_decay']},
|
|
]
|
|
for i in range(5):
|
|
if hasattr(self, 'layer%d'%i):
|
|
params_list.append(
|
|
{'params': getattr(self, 'layer%d'%i).parameters(), 'lr': optimizer_cfg['lr'] * 0.1, 'weight_decay': optimizer_cfg['weight_decay']}
|
|
)
|
|
|
|
optimizer = optimizer(params_list)
|
|
return optimizer
|
|
|
|
def init_parameters(self):
|
|
for m in self.modules():
|
|
if isinstance(m, nn.Linear):
|
|
trunc_normal_(m.weight, std=.02)
|
|
if isinstance(m, nn.Linear) and m.bias is not None:
|
|
nn.init.constant_(m.bias, 0)
|
|
elif isinstance(m, nn.LayerNorm):
|
|
nn.init.constant_(m.bias, 0)
|
|
nn.init.constant_(m.weight, 1.0)
|
|
elif isinstance(m, (nn.Conv3d, nn.Conv2d, nn.Conv1d)):
|
|
nn.init.xavier_uniform_(m.weight.data)
|
|
if m.bias is not None:
|
|
nn.init.constant_(m.bias.data, 0.0)
|
|
elif isinstance(m, (nn.BatchNorm3d, nn.BatchNorm2d, nn.BatchNorm1d)):
|
|
if m.affine:
|
|
nn.init.normal_(m.weight.data, 1.0, 0.02)
|
|
nn.init.constant_(m.bias.data, 0.0)
|
|
|
|
def make_layer(self, block, planes, stride, blocks_num, mode='2d'):
|
|
|
|
if max(stride) > 1 or self.inplanes != planes * block.expansion:
|
|
if mode == '3d':
|
|
downsample = nn.Sequential(nn.Conv3d(self.inplanes, planes * block.expansion, kernel_size=[1, 1, 1], stride=stride, padding=[0, 0, 0], bias=False), nn.BatchNorm3d(planes * block.expansion))
|
|
elif mode == '2d':
|
|
downsample = nn.Sequential(conv1x1(self.inplanes, planes * block.expansion, stride=stride), nn.BatchNorm2d(planes * block.expansion))
|
|
elif mode == 'p3d':
|
|
downsample = nn.Sequential(nn.Conv3d(self.inplanes, planes * block.expansion, kernel_size=[1, 1, 1], stride=[1, *stride], padding=[0, 0, 0], bias=False), nn.BatchNorm3d(planes * block.expansion))
|
|
else:
|
|
raise TypeError('xxx')
|
|
else:
|
|
downsample = lambda x: x
|
|
|
|
layers = [block(self.inplanes, planes, stride=stride, downsample=downsample)]
|
|
self.inplanes = planes * block.expansion
|
|
s = [1, 1] if mode in ['2d', 'p3d'] else [1, 1, 1]
|
|
for i in range(1, blocks_num):
|
|
layers.append(
|
|
block(self.inplanes, planes, stride=s)
|
|
)
|
|
return nn.Sequential(*layers)
|
|
|
|
def forward(self, inputs):
|
|
if self.training:
|
|
adjust_learning_rate(self.optimizer, self.iteration, T_max_iter=self.T_max_iter)
|
|
ipts, labs, _, _, seqL = inputs
|
|
|
|
sils = ipts[0].unsqueeze(1)
|
|
|
|
del ipts
|
|
|
|
out0 = self.layer0(sils)
|
|
out1 = self.layer1(out0)
|
|
out2 = self.layer2(out1) # [n, c, s, h, w]
|
|
out2 = self.ulayer(out2)
|
|
out4 = self.transformer(out2) # [n, 768, s/4, 4, 3]
|
|
|
|
# Temporal Pooling, TP
|
|
outs = self.TP(out4, seqL, options={"dim": 2})[0] # [n, c, h, w]
|
|
# Horizontal Pooling Matching, HPM
|
|
feat = self.HPP(outs) # [n, c, p]
|
|
|
|
feat = torch.cat([feat, feat[:, :, -1].clone().detach().unsqueeze(-1)], dim=-1)
|
|
embed_1 = self.FCs(feat) # [n, c, p]
|
|
embed_2, logits = self.BNNecks(embed_1) # [n, c, p]
|
|
embed_1 = embed_1.contiguous()[:, :, :-1] # [n, p, c]
|
|
embed_2 = embed_2.contiguous()[:, :, :-1] # [n, p, c]
|
|
logits = logits.contiguous()[:, :, :-1] # [n, p, c]
|
|
|
|
embed = embed_1
|
|
|
|
retval = {
|
|
'training_feat': {
|
|
'triplet': {'embeddings': embed_1, 'labels': labs},
|
|
'softmax': {'logits': logits, 'labels': labs}
|
|
},
|
|
'visual_summary': {
|
|
'image/sils': rearrange(sils,'n c s h w -> (n s) c h w')
|
|
},
|
|
'inference_feat': {
|
|
'embeddings': embed
|
|
}
|
|
}
|
|
return retval
|
|
|
|
import math
|
|
def adjust_learning_rate(optimizer, iteration, iteration_per_epoch=1000, T_max_iter=10000, min_lr=1e-6):
|
|
"""Decay the learning rate based on schedule"""
|
|
if iteration < T_max_iter:
|
|
if iteration % iteration_per_epoch == 0 :
|
|
alpha = 0.5 * (1. + math.cos(math.pi * iteration / T_max_iter))
|
|
for param_group in optimizer.param_groups:
|
|
param_group['lr'] = max(param_group['initial_lr'] * alpha, min_lr)
|
|
else:
|
|
pass
|
|
elif iteration == T_max_iter:
|
|
for param_group in optimizer.param_groups:
|
|
param_group['lr'] = min_lr
|
|
else:
|
|
pass
|