1491 lines
65 KiB
Python
1491 lines
65 KiB
Python
# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
|
|
"""
|
|
Common modules
|
|
"""
|
|
|
|
import ast
|
|
import contextlib
|
|
import json
|
|
import math
|
|
import platform
|
|
import warnings
|
|
import zipfile
|
|
from collections import OrderedDict, namedtuple
|
|
from copy import copy
|
|
from pathlib import Path
|
|
from urllib.parse import urlparse
|
|
|
|
import cv2
|
|
import numpy as np
|
|
import pandas as pd
|
|
import requests
|
|
import torch
|
|
import torch.nn as nn
|
|
from PIL import Image
|
|
from torch.cuda import amp
|
|
|
|
from thirdTool.yolo5x_qr.utils import TryExcept
|
|
from thirdTool.yolo5x_qr.utils.dataloaders import exif_transpose, letterbox
|
|
from thirdTool.yolo5x_qr.utils.general import (LOGGER, ROOT, Profile, check_requirements, check_suffix, check_version, colorstr,
|
|
increment_path, is_jupyter, make_divisible, non_max_suppression, scale_boxes, xywh2xyxy,
|
|
xyxy2xywh, yaml_load)
|
|
from thirdTool.yolo5x_qr.utils.plots import Annotator, colors, save_one_box
|
|
from thirdTool.yolo5x_qr.utils.torch_utils import copy_attr, smart_inference_mode
|
|
|
|
from torch.nn import functional as F
|
|
from .Functions import Encoding, Mean, DropPath, Mlp, GroupNorm, ConvBlock
|
|
|
|
|
|
def autopad(k, p=None, d=1): # kernel, padding, dilation
|
|
# Pad to 'same' shape outputs
|
|
if d > 1:
|
|
k = d * (k - 1) + 1 if isinstance(k, int) else [d * (x - 1) + 1 for x in k] # actual kernel-size
|
|
if p is None:
|
|
p = k // 2 if isinstance(k, int) else [x // 2 for x in k] # auto-pad
|
|
return p
|
|
|
|
def export_formats():
|
|
# YOLOv5 export formats
|
|
x = [
|
|
['PyTorch', '-', '.pt', True, True],
|
|
['TorchScript', 'torchscript', '.torchscript', True, True],
|
|
['ONNX', 'onnx', '.onnx', True, True],
|
|
['OpenVINO', 'openvino', '_openvino_model', True, False],
|
|
['TensorRT', 'engine', '.engine', False, True],
|
|
['CoreML', 'coreml', '.mlmodel', True, False],
|
|
['TensorFlow SavedModel', 'saved_model', '_saved_model', True, True],
|
|
['TensorFlow GraphDef', 'pb', '.pb', True, True],
|
|
['TensorFlow Lite', 'tflite', '.tflite', True, False],
|
|
['TensorFlow Edge TPU', 'edgetpu', '_edgetpu.tflite', False, False],
|
|
['TensorFlow.js', 'tfjs', '_web_model', False, False],
|
|
['PaddlePaddle', 'paddle', '_paddle_model', True, True],]
|
|
return pd.DataFrame(x, columns=['Format', 'Argument', 'Suffix', 'CPU', 'GPU'])
|
|
|
|
class Conv(nn.Module):
|
|
# Standard convolution with args(ch_in, ch_out, kernel, stride, padding, groups, dilation, activation)
|
|
default_act = nn.SiLU() # default activation
|
|
|
|
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, d=1, act=True):
|
|
super().__init__()
|
|
self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p, d), groups=g, dilation=d, bias=False)
|
|
self.bn = nn.BatchNorm2d(c2)
|
|
self.act = self.default_act if act is True else act if isinstance(act, nn.Module) else nn.Identity()
|
|
|
|
def forward(self, x):
|
|
return self.act(self.bn(self.conv(x)))
|
|
|
|
def forward_fuse(self, x):
|
|
return self.act(self.conv(x))
|
|
|
|
|
|
class DWConv(Conv):
|
|
# Depth-wise convolution
|
|
def __init__(self, c1, c2, k=1, s=1, d=1, act=True): # ch_in, ch_out, kernel, stride, dilation, activation
|
|
super().__init__(c1, c2, k, s, g=math.gcd(c1, c2), d=d, act=act)
|
|
|
|
|
|
class DWConvTranspose2d(nn.ConvTranspose2d):
|
|
# Depth-wise transpose convolution
|
|
def __init__(self, c1, c2, k=1, s=1, p1=0, p2=0): # ch_in, ch_out, kernel, stride, padding, padding_out
|
|
super().__init__(c1, c2, k, s, p1, p2, groups=math.gcd(c1, c2))
|
|
|
|
|
|
class TransformerLayer(nn.Module):
|
|
# Transformer layer https://arxiv.org/abs/2010.11929 (LayerNorm layers removed for better performance)
|
|
def __init__(self, c, num_heads):
|
|
super().__init__()
|
|
self.q = nn.Linear(c, c, bias=False)
|
|
self.k = nn.Linear(c, c, bias=False)
|
|
self.v = nn.Linear(c, c, bias=False)
|
|
self.ma = nn.MultiheadAttention(embed_dim=c, num_heads=num_heads)
|
|
self.fc1 = nn.Linear(c, c, bias=False)
|
|
self.fc2 = nn.Linear(c, c, bias=False)
|
|
|
|
def forward(self, x):
|
|
x = self.ma(self.q(x), self.k(x), self.v(x))[0] + x
|
|
x = self.fc2(self.fc1(x)) + x
|
|
return x
|
|
|
|
|
|
class TransformerBlock(nn.Module):
|
|
# Vision Transformer https://arxiv.org/abs/2010.11929
|
|
def __init__(self, c1, c2, num_heads, num_layers):
|
|
super().__init__()
|
|
self.conv = None
|
|
if c1 != c2:
|
|
self.conv = Conv(c1, c2)
|
|
self.linear = nn.Linear(c2, c2) # learnable position embedding
|
|
self.tr = nn.Sequential(*(TransformerLayer(c2, num_heads) for _ in range(num_layers)))
|
|
self.c2 = c2
|
|
|
|
def forward(self, x):
|
|
if self.conv is not None:
|
|
x = self.conv(x)
|
|
b, _, w, h = x.shape
|
|
p = x.flatten(2).permute(2, 0, 1)
|
|
return self.tr(p + self.linear(p)).permute(1, 2, 0).reshape(b, self.c2, w, h)
|
|
|
|
|
|
class Bottleneck(nn.Module):
|
|
# Standard bottleneck
|
|
def __init__(self, c1, c2, shortcut=True, g=1, e=0.5): # ch_in, ch_out, shortcut, groups, expansion
|
|
super().__init__()
|
|
c_ = int(c2 * e) # hidden channels
|
|
self.cv1 = Conv(c1, c_, 1, 1)
|
|
self.cv2 = Conv(c_, c2, 3, 1, g=g)
|
|
self.add = shortcut and c1 == c2
|
|
|
|
def forward(self, x):
|
|
return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))
|
|
|
|
|
|
class BottleneckCSP(nn.Module):
|
|
# CSP Bottleneck https://github.com/WongKinYiu/CrossStagePartialNetworks
|
|
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion
|
|
super().__init__()
|
|
c_ = int(c2 * e) # hidden channels
|
|
self.cv1 = Conv(c1, c_, 1, 1)
|
|
self.cv2 = nn.Conv2d(c1, c_, 1, 1, bias=False)
|
|
self.cv3 = nn.Conv2d(c_, c_, 1, 1, bias=False)
|
|
self.cv4 = Conv(2 * c_, c2, 1, 1)
|
|
self.bn = nn.BatchNorm2d(2 * c_) # applied to cat(cv2, cv3)
|
|
self.act = nn.SiLU()
|
|
self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)))
|
|
|
|
def forward(self, x):
|
|
y1 = self.cv3(self.m(self.cv1(x)))
|
|
y2 = self.cv2(x)
|
|
return self.cv4(self.act(self.bn(torch.cat((y1, y2), 1))))
|
|
|
|
|
|
class CrossConv(nn.Module):
|
|
# Cross Convolution Downsample
|
|
def __init__(self, c1, c2, k=3, s=1, g=1, e=1.0, shortcut=False):
|
|
# ch_in, ch_out, kernel, stride, groups, expansion, shortcut
|
|
super().__init__()
|
|
c_ = int(c2 * e) # hidden channels
|
|
self.cv1 = Conv(c1, c_, (1, k), (1, s))
|
|
self.cv2 = Conv(c_, c2, (k, 1), (s, 1), g=g)
|
|
self.add = shortcut and c1 == c2
|
|
|
|
def forward(self, x):
|
|
return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))
|
|
|
|
|
|
class C3(nn.Module):
|
|
# CSP Bottleneck with 3 convolutions
|
|
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion
|
|
super().__init__()
|
|
c_ = int(c2 * e) # hidden channels
|
|
self.cv1 = Conv(c1, c_, 1, 1)
|
|
self.cv2 = Conv(c1, c_, 1, 1)
|
|
self.cv3 = Conv(2 * c_, c2, 1) # optional act=FReLU(c2)
|
|
self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)))
|
|
|
|
def forward(self, x):
|
|
return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), 1))
|
|
|
|
|
|
class C3x(C3):
|
|
# C3 module with cross-convolutions
|
|
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):
|
|
super().__init__(c1, c2, n, shortcut, g, e)
|
|
c_ = int(c2 * e)
|
|
self.m = nn.Sequential(*(CrossConv(c_, c_, 3, 1, g, 1.0, shortcut) for _ in range(n)))
|
|
|
|
|
|
class C3TR(C3):
|
|
# C3 module with TransformerBlock()
|
|
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):
|
|
super().__init__(c1, c2, n, shortcut, g, e)
|
|
c_ = int(c2 * e)
|
|
self.m = TransformerBlock(c_, c_, 4, n)
|
|
|
|
|
|
class C3SPP(C3):
|
|
# C3 module with SPP()
|
|
def __init__(self, c1, c2, k=(5, 9, 13), n=1, shortcut=True, g=1, e=0.5):
|
|
super().__init__(c1, c2, n, shortcut, g, e)
|
|
c_ = int(c2 * e)
|
|
self.m = SPP(c_, c_, k)
|
|
|
|
|
|
class C3Ghost(C3):
|
|
# C3 module with GhostBottleneck()
|
|
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):
|
|
super().__init__(c1, c2, n, shortcut, g, e)
|
|
c_ = int(c2 * e) # hidden channels
|
|
self.m = nn.Sequential(*(GhostBottleneck(c_, c_) for _ in range(n)))
|
|
|
|
|
|
class SPP(nn.Module):
|
|
# Spatial Pyramid Pooling (SPP) layer https://arxiv.org/abs/1406.4729
|
|
def __init__(self, c1, c2, k=(5, 9, 13)):
|
|
super().__init__()
|
|
c_ = c1 // 2 # hidden channels
|
|
self.cv1 = Conv(c1, c_, 1, 1)
|
|
self.cv2 = Conv(c_ * (len(k) + 1), c2, 1, 1)
|
|
self.m = nn.ModuleList([nn.MaxPool2d(kernel_size=x, stride=1, padding=x // 2) for x in k])
|
|
|
|
def forward(self, x):
|
|
x = self.cv1(x)
|
|
with warnings.catch_warnings():
|
|
warnings.simplefilter('ignore') # suppress torch 1.9.0 max_pool2d() warning
|
|
return self.cv2(torch.cat([x] + [m(x) for m in self.m], 1))
|
|
|
|
|
|
class SPPF(nn.Module):
|
|
# Spatial Pyramid Pooling - Fast (SPPF) layer for YOLOv5 by Glenn Jocher
|
|
def __init__(self, c1, c2, k=5): # equivalent to SPP(k=(5, 9, 13))
|
|
super().__init__()
|
|
c_ = c1 // 2 # hidden channels
|
|
self.cv1 = Conv(c1, c_, 1, 1)
|
|
self.cv2 = Conv(c_ * 4, c2, 1, 1)
|
|
self.m = nn.MaxPool2d(kernel_size=k, stride=1, padding=k // 2)
|
|
|
|
def forward(self, x):
|
|
x = self.cv1(x)
|
|
with warnings.catch_warnings():
|
|
warnings.simplefilter('ignore') # suppress torch 1.9.0 max_pool2d() warning
|
|
y1 = self.m(x)
|
|
y2 = self.m(y1)
|
|
return self.cv2(torch.cat((x, y1, y2, self.m(y2)), 1))
|
|
|
|
|
|
class Focus(nn.Module):
|
|
# Focus wh information into c-space
|
|
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups
|
|
super().__init__()
|
|
self.conv = Conv(c1 * 4, c2, k, s, p, g, act=act)
|
|
# self.contract = Contract(gain=2)
|
|
|
|
def forward(self, x): # x(b,c,w,h) -> y(b,4c,w/2,h/2)
|
|
return self.conv(torch.cat((x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]), 1))
|
|
# return self.conv(self.contract(x))
|
|
|
|
|
|
class GhostConv(nn.Module):
|
|
# Ghost Convolution https://github.com/huawei-noah/ghostnet
|
|
def __init__(self, c1, c2, k=1, s=1, g=1, act=True): # ch_in, ch_out, kernel, stride, groups
|
|
super().__init__()
|
|
c_ = c2 // 2 # hidden channels
|
|
self.cv1 = Conv(c1, c_, k, s, None, g, act=act)
|
|
self.cv2 = Conv(c_, c_, 5, 1, None, c_, act=act)
|
|
|
|
def forward(self, x):
|
|
y = self.cv1(x)
|
|
return torch.cat((y, self.cv2(y)), 1)
|
|
|
|
|
|
class GhostBottleneck(nn.Module):
|
|
# Ghost Bottleneck https://github.com/huawei-noah/ghostnet
|
|
def __init__(self, c1, c2, k=3, s=1): # ch_in, ch_out, kernel, stride
|
|
super().__init__()
|
|
c_ = c2 // 2
|
|
self.conv = nn.Sequential(
|
|
GhostConv(c1, c_, 1, 1), # pw
|
|
DWConv(c_, c_, k, s, act=False) if s == 2 else nn.Identity(), # dw
|
|
GhostConv(c_, c2, 1, 1, act=False)) # pw-linear
|
|
self.shortcut = nn.Sequential(DWConv(c1, c1, k, s, act=False), Conv(c1, c2, 1, 1,
|
|
act=False)) if s == 2 else nn.Identity()
|
|
|
|
def forward(self, x):
|
|
return self.conv(x) + self.shortcut(x)
|
|
|
|
|
|
class Contract(nn.Module):
|
|
# Contract width-height into channels, i.e. x(1,64,80,80) to x(1,256,40,40)
|
|
def __init__(self, gain=2):
|
|
super().__init__()
|
|
self.gain = gain
|
|
|
|
def forward(self, x):
|
|
b, c, h, w = x.size() # assert (h / s == 0) and (W / s == 0), 'Indivisible gain'
|
|
s = self.gain
|
|
x = x.view(b, c, h // s, s, w // s, s) # x(1,64,40,2,40,2)
|
|
x = x.permute(0, 3, 5, 1, 2, 4).contiguous() # x(1,2,2,64,40,40)
|
|
return x.view(b, c * s * s, h // s, w // s) # x(1,256,40,40)
|
|
|
|
|
|
class Expand(nn.Module):
|
|
# Expand channels into width-height, i.e. x(1,64,80,80) to x(1,16,160,160)
|
|
def __init__(self, gain=2):
|
|
super().__init__()
|
|
self.gain = gain
|
|
|
|
def forward(self, x):
|
|
b, c, h, w = x.size() # assert C / s ** 2 == 0, 'Indivisible gain'
|
|
s = self.gain
|
|
x = x.view(b, s, s, c // s ** 2, h, w) # x(1,2,2,16,80,80)
|
|
x = x.permute(0, 3, 4, 1, 5, 2).contiguous() # x(1,16,80,2,80,2)
|
|
return x.view(b, c // s ** 2, h * s, w * s) # x(1,16,160,160)
|
|
|
|
|
|
class Concat(nn.Module):
|
|
# Concatenate a list of tensors along dimension
|
|
def __init__(self, dimension=1):
|
|
super().__init__()
|
|
self.d = dimension
|
|
|
|
def forward(self, x):
|
|
return torch.cat(x, self.d)
|
|
|
|
#by AI&CV C2F
|
|
class v8_C2fBottleneck(nn.Module):
|
|
# Standard bottleneck
|
|
def __init__(self, c1, c2, shortcut=True, g=1, k=(3, 3), e=0.5): # ch_in, ch_out, shortcut, groups, kernels, expand
|
|
super().__init__()
|
|
c_ = int(c2 * e) # hidden channels
|
|
self.cv1 = Conv(c1, c_, k[0], 1)
|
|
self.cv2 = Conv(c_, c2, k[1], 1, g=g)
|
|
self.add = shortcut and c1 == c2
|
|
|
|
def forward(self, x):
|
|
return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))
|
|
|
|
|
|
class C2f(nn.Module):
|
|
# CSP Bottleneck with 2 convolutions
|
|
def __init__(self, c1, c2, n=1, shortcut=False, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion
|
|
super().__init__()
|
|
self.c = int(c2 * e) # hidden channels
|
|
self.cv1 = Conv(c1, 2 * self.c, 1, 1)
|
|
self.cv2 = Conv((2 + n) * self.c, c2, 1) # optional act=FReLU(c2)
|
|
self.m = nn.ModuleList(
|
|
v8_C2fBottleneck(self.c, self.c, shortcut, g, k=((3, 3), (3, 3)), e=1.0) for _ in range(n))
|
|
|
|
def forward(self, x):
|
|
y = list(self.cv1(x).split((self.c, self.c), 1))
|
|
y.extend(m(y[-1]) for m in self.m)
|
|
return self.cv2(torch.cat(y, 1))
|
|
|
|
|
|
# by AI&CV EVCBlock ####### start #########################
|
|
|
|
class SiLU(nn.Module):
|
|
"""export-friendly version of nn.SiLU()"""
|
|
|
|
@staticmethod
|
|
def forward(x):
|
|
return x * torch.sigmoid(x)
|
|
|
|
|
|
def get_activation(name="silu", inplace=True):
|
|
if name == "silu":
|
|
module = nn.SiLU(inplace=inplace)
|
|
elif name == "relu":
|
|
module = nn.ReLU(inplace=inplace)
|
|
elif name == "lrelu":
|
|
module = nn.LeakyReLU(0.1, inplace=inplace)
|
|
else:
|
|
raise AttributeError("Unsupported act type: {}".format(name))
|
|
return module
|
|
|
|
class BaseConv(nn.Module):
|
|
"""A Conv2d -> Batchnorm -> silu/leaky relu block""" # CBL
|
|
|
|
def __init__(
|
|
self, in_channels, out_channels, ksize, stride, groups=1, bias=False, act="silu"
|
|
):
|
|
super().__init__()
|
|
# same padding
|
|
pad = (ksize - 1) // 2
|
|
self.conv = nn.Conv2d(
|
|
in_channels,
|
|
out_channels,
|
|
kernel_size=ksize,
|
|
stride=stride,
|
|
padding=pad,
|
|
groups=groups,
|
|
bias=bias,
|
|
)
|
|
self.bn = nn.BatchNorm2d(out_channels)
|
|
self.act = get_activation(act, inplace=True)
|
|
|
|
def forward(self, x):
|
|
return self.act(self.bn(self.conv(x)))
|
|
|
|
def fuseforward(self, x):
|
|
return self.act(self.conv(x))
|
|
|
|
class DWConv(nn.Module):
|
|
"""Depthwise Conv + Conv"""
|
|
def __init__(self, in_channels, out_channels, ksize, stride=1, act="silu"):
|
|
super().__init__()
|
|
self.dconv = BaseConv(
|
|
in_channels,
|
|
in_channels,
|
|
ksize=ksize,
|
|
stride=stride,
|
|
groups=in_channels,
|
|
act=act,
|
|
)
|
|
self.pconv = BaseConv(
|
|
in_channels, out_channels, ksize=1, stride=1, groups=1, act=act
|
|
)
|
|
|
|
def forward(self, x):
|
|
x = self.dconv(x)
|
|
return self.pconv(x)
|
|
|
|
class LVCBlock(nn.Module):
|
|
def __init__(self, c1, c2, num_codes, channel_ratio=0.25, base_channel=64):
|
|
super(LVCBlock, self).__init__()
|
|
self.c2 = c2
|
|
self.num_codes = num_codes
|
|
num_codes = 64
|
|
|
|
self.conv_1 = ConvBlock(c1, c1, res_conv=True, stride=1)
|
|
|
|
self.LVC = nn.Sequential(
|
|
nn.Conv2d(c1, c1, 1, bias=False),
|
|
nn.BatchNorm2d(c1),
|
|
nn.ReLU(inplace=True),
|
|
Encoding(c1, num_codes=num_codes),
|
|
nn.BatchNorm1d(num_codes),
|
|
nn.ReLU(inplace=True),
|
|
Mean(dim=1))
|
|
self.fc = nn.Sequential(nn.Linear(c1, c1), nn.Sigmoid())
|
|
|
|
def forward(self, x):
|
|
x = self.conv_1(x, return_x_2=False)
|
|
en = self.LVC(x)
|
|
gam = self.fc(en)
|
|
b, in_channels, _, _ = x.size()
|
|
y = gam.view(b, in_channels, 1, 1)
|
|
x = F.relu_(x + x * y)
|
|
return x
|
|
|
|
|
|
# LightMLPBlock
|
|
class LightMLPBlock(nn.Module):
|
|
def __init__(self, c1, c2, ksize=1, stride=1, act="silu",
|
|
mlp_ratio=4., drop=0., act_layer=nn.GELU,
|
|
use_layer_scale=True, layer_scale_init_value=1e-5, drop_path=0., norm_layer=GroupNorm): # act_layer=nn.GELU,
|
|
super().__init__()
|
|
self.dw = DWConv(c1, c2, ksize=1, stride=1, act="silu")
|
|
self.linear = nn.Linear(c2, c2) # learnable position embedding
|
|
self.c2 = c2
|
|
|
|
self.norm1 = norm_layer(c1)
|
|
self.norm2 = norm_layer(c1)
|
|
|
|
mlp_hidden_dim = int(c1 * mlp_ratio)
|
|
self.mlp = Mlp(in_features=c1, hidden_features=mlp_hidden_dim, act_layer=nn.GELU,
|
|
drop=drop)
|
|
|
|
self.drop_path = DropPath(drop_path) if drop_path > 0. \
|
|
else nn.Identity()
|
|
|
|
self.use_layer_scale = use_layer_scale
|
|
if use_layer_scale:
|
|
self.layer_scale_1 = nn.Parameter(
|
|
layer_scale_init_value * torch.ones((c2)), requires_grad=True)
|
|
self.layer_scale_2 = nn.Parameter(
|
|
layer_scale_init_value * torch.ones((c2)), requires_grad=True)
|
|
|
|
def forward(self, x):
|
|
if self.use_layer_scale:
|
|
x = x + self.drop_path(self.layer_scale_1.unsqueeze(-1).unsqueeze(-1) * self.dw(self.norm1(x)))
|
|
x = x + self.drop_path(self.layer_scale_2.unsqueeze(-1).unsqueeze(-1) * self.mlp(self.norm2(x)))
|
|
else:
|
|
x = x + self.drop_path(self.dw(self.norm1(x)))
|
|
x = x + self.drop_path(self.mlp(self.norm2(x)))
|
|
return x
|
|
|
|
|
|
# EVCBlock
|
|
class EVCBlock(nn.Module):
|
|
def __init__(self, c1, c2, channel_ratio=4, base_channel=16):
|
|
super().__init__()
|
|
expansion = 2
|
|
ch = c2 * expansion
|
|
# Stem stage: get the feature maps by conv block (copied form resnet.py) 进入conformer框架之前的处理
|
|
self.conv1 = nn.Conv2d(c1, c1, kernel_size=7, stride=1, padding=3, bias=False) # 1 / 2 [112, 112]
|
|
self.bn1 = nn.BatchNorm2d(c1)
|
|
self.act1 = nn.ReLU(inplace=True)
|
|
self.maxpool = nn.MaxPool2d(kernel_size=3, stride=1, padding=1) # 1 / 4 [56, 56]
|
|
|
|
# LVC
|
|
self.lvc = LVCBlock(c1, c2, num_codes=64) # c1值暂时未定
|
|
# LightMLPBlock
|
|
self.l_MLP = LightMLPBlock(c1, c2, ksize=1, stride=1, act="silu", act_layer=nn.GELU, mlp_ratio=4., drop=0.,
|
|
use_layer_scale=True, layer_scale_init_value=1e-5, drop_path=0., norm_layer=GroupNorm)
|
|
self.cnv1 = nn.Conv2d(ch, c2, kernel_size=1, stride=1, padding=0)
|
|
|
|
def forward(self, x):
|
|
x1 = self.maxpool(self.act1(self.bn1(self.conv1(x))))
|
|
# LVCBlock
|
|
x_lvc = self.lvc(x1)
|
|
# LightMLPBlock
|
|
x_lmlp = self.l_MLP(x1)
|
|
# concat
|
|
x = torch.cat((x_lvc, x_lmlp), dim=1)
|
|
x = self.cnv1(x)
|
|
return x
|
|
# by AI&CV EVCBlock ####### end #########################
|
|
|
|
# ------------------------------------ODConv start -------------------------------------
|
|
class ODConv2d_3rd(nn.Conv2d):
|
|
|
|
def __init__(self, in_channels, out_channels, kernel_size,
|
|
stride=1, padding=0, dilation=1, groups=1, bias=True,
|
|
K=4, r=1 / 16, save_parameters=False,
|
|
padding_mode='zeros', device=None, dtype=None) -> None:
|
|
factory_kwargs = {'device': device, 'dtype': dtype}
|
|
self.K = K
|
|
self.r = r
|
|
self.save_parameters = save_parameters
|
|
|
|
super().__init__(in_channels, out_channels, kernel_size, stride,
|
|
padding, dilation, groups, bias, padding_mode)
|
|
|
|
del self.weight
|
|
self.weight = nn.Parameter(torch.empty((
|
|
K,
|
|
out_channels,
|
|
in_channels // groups,
|
|
*self.kernel_size,
|
|
), **factory_kwargs))
|
|
|
|
if bias:
|
|
del self.bias
|
|
self.bias = nn.Parameter(torch.empty(K, out_channels, **factory_kwargs))
|
|
|
|
hidden_dim = max(int(in_channels * r), 16) # 设置下限为16
|
|
self.gap = nn.AdaptiveAvgPool2d(1)
|
|
self.reduction = nn.Linear(in_channels, hidden_dim)
|
|
self.fc = nn.Conv2d(in_channels, hidden_dim, 1, bias=False)
|
|
self.bn = nn.BatchNorm2d(hidden_dim)
|
|
self.act = nn.ReLU(inplace=True)
|
|
# self.act = nn.SiLU(inplace=True)
|
|
|
|
self.fc_f = nn.Linear(hidden_dim, out_channels)
|
|
if not save_parameters or self.kernel_size[0] * self.kernel_size[1] > 1:
|
|
self.fc_s = nn.Linear(hidden_dim, self.kernel_size[0] * self.kernel_size[1])
|
|
if not save_parameters or in_channels // groups > 1:
|
|
self.fc_c = nn.Linear(hidden_dim, in_channels // groups)
|
|
if not save_parameters or K > 1:
|
|
self.fc_w = nn.Linear(hidden_dim, K)
|
|
|
|
self.reset_parameters()
|
|
|
|
def reset_parameters(self) -> None:
|
|
fan_out = self.kernel_size[0] * self.kernel_size[1] * self.out_channels // self.groups
|
|
for i in range(self.K):
|
|
self.weight.data[i].normal_(0, math.sqrt(2.0 / fan_out))
|
|
if self.bias is not None:
|
|
self.bias.data.zero_()
|
|
|
|
def extra_repr(self):
|
|
return super().extra_repr() + f', K={self.K}, r={self.r:.4}'
|
|
|
|
def get_weight_bias(self, context):
|
|
B, C, H, W = context.shape
|
|
|
|
if C != self.in_channels:
|
|
raise ValueError(
|
|
f"Expected context{[B, C, H, W]} to have {self.in_channels} channels, but got {C} channels instead")
|
|
|
|
# x = self.gap(context).squeeze(-1).squeeze(-1) # B, c_in
|
|
# x = self.reduction(x) # B, hidden_dim
|
|
x = self.gap(context)
|
|
x = self.fc(x)
|
|
if x.size(0) > 1:
|
|
x = self.bn(x)
|
|
x = x.squeeze(-1).squeeze(-1)
|
|
x = self.act(x)
|
|
|
|
attn_f = self.fc_f(x).sigmoid() # B, c_out
|
|
attn = attn_f.view(B, 1, -1, 1, 1, 1) # B, 1, c_out, 1, 1, 1
|
|
if hasattr(self, 'fc_s'):
|
|
attn_s = self.fc_s(x).sigmoid() # B, k * k
|
|
attn = attn * attn_s.view(B, 1, 1, 1, *self.kernel_size) # B, 1, c_out, 1, k, k
|
|
if hasattr(self, 'fc_c'):
|
|
attn_c = self.fc_c(x).sigmoid() # B, c_in // groups
|
|
attn = attn * attn_c.view(B, 1, 1, -1, 1, 1) # B, 1, c_out, c_in // groups, k, k
|
|
if hasattr(self, 'fc_w'):
|
|
attn_w = self.fc_w(x).softmax(-1) # B, n
|
|
attn = attn * attn_w.view(B, -1, 1, 1, 1, 1) # B, n, c_out, c_in // groups, k, k
|
|
|
|
weight = (attn * self.weight).sum(1) # B, c_out, c_in // groups, k, k
|
|
weight = weight.view(-1, self.in_channels // self.groups, *self.kernel_size) # B * c_out, c_in // groups, k, k
|
|
|
|
bias = None
|
|
if self.bias is not None:
|
|
if hasattr(self, 'fc_w'):
|
|
bias = attn_w @ self.bias
|
|
else:
|
|
bias = self.bias.tile(B, 1)
|
|
bias = bias.view(-1) # B * c_out
|
|
|
|
return weight, bias
|
|
|
|
def forward(self, input, context=None):
|
|
B, C, H, W = input.shape
|
|
|
|
if C != self.in_channels:
|
|
raise ValueError(
|
|
f"Expected input{[B, C, H, W]} to have {self.in_channels} channels, but got {C} channels instead")
|
|
|
|
weight, bias = self.get_weight_bias(context or input)
|
|
|
|
output = nn.functional.conv2d(
|
|
input.view(1, B * C, H, W), weight, bias,
|
|
self.stride, self.padding, self.dilation, B * self.groups) # 1, B * c_out, h_out, w_out
|
|
output = output.view(B, self.out_channels, *output.shape[2:])
|
|
|
|
return output
|
|
|
|
def debug(self, input, context=None):
|
|
B, C, H, W = input.shape
|
|
|
|
if C != self.in_channels:
|
|
raise ValueError(
|
|
f"Expected input{[B, C, H, W]} to have {self.in_channels} channels, but got {C} channels instead")
|
|
|
|
output_size = [
|
|
((H, W)[i] + 2 * self.padding[i] - self.dilation[i] * (self.kernel_size[i] - 1) - 1) // self.stride[i] + 1
|
|
for i in range(2)
|
|
]
|
|
|
|
weight, bias = self.get_weight_bias(context or input)
|
|
|
|
weight = weight.view(B, self.groups, self.out_channels // self.groups,
|
|
-1) # B, groups, c_out // groups, c_in // groups * k * k
|
|
|
|
unfold = nn.functional.unfold(
|
|
input, self.kernel_size, self.dilation, self.padding, self.stride) # B, c_in * k * k, H_out * W_out
|
|
unfold = unfold.view(B, self.groups, -1,
|
|
output_size[0] * output_size[1]) # B, groups, c_in // groups * k * k, H_out * W_out
|
|
|
|
output = weight @ unfold # B, groups, c_out // groups, H_out * W_out
|
|
output = output.view(B, self.out_channels, *output_size) # B, c_out, H_out * W_out
|
|
|
|
if bias is not None:
|
|
output = output + bias.view(B, self.out_channels, 1, 1)
|
|
|
|
return output
|
|
|
|
|
|
class ODConv_3rd(nn.Module):
|
|
# Standard convolution
|
|
def __init__(self, c1, c2, k=1, s=1, kerNums=1, g=1, p=None,
|
|
act=True): # ch_in, ch_out, kernel, stride, padding, groups
|
|
super().__init__()
|
|
self.conv = ODConv2d_3rd(c1, c2, k, s, autopad(k, p), groups=g, K=kerNums)
|
|
self.bn = nn.BatchNorm2d(c2)
|
|
self.act = nn.SiLU() if act is True else (act if isinstance(act, nn.Module) else nn.Identity())
|
|
|
|
def forward(self, x):
|
|
return self.act(self.bn(self.conv(x)))
|
|
|
|
def forward_fuse(self, x):
|
|
return self.act(self.conv(x))
|
|
|
|
|
|
# ------------------------------------ODConv end -------------------------------------
|
|
# ------------------------------------Convnext start -------------------------------------
|
|
# ConvNextBlock
|
|
class ConvNextBlock(nn.Module):
|
|
|
|
def __init__(self, inputdim, dim, drop_path=0., layer_scale_init_value=1e-6,
|
|
kersize=7): # demo: [64, 64, 1] 1 denotes the number of repeats
|
|
super().__init__()
|
|
# 匹配yolov5配置文件加入outdim输出通道
|
|
# self.flag = True if dim == outdim else False
|
|
|
|
self.dwconv = nn.Conv2d(dim, dim, kernel_size=kersize, padding=kersize // 2, groups=dim) # depthwise conv
|
|
self.norm = LayerNorm_s(dim, eps=1e-6)
|
|
self.pwconv1 = nn.Linear(dim, 4 * dim)
|
|
self.act = nn.GELU()
|
|
self.pwconv2 = nn.Linear(4 * dim, dim)
|
|
self.gamma = nn.Parameter(layer_scale_init_value * torch.ones((dim)),
|
|
requires_grad=True) if layer_scale_init_value > 0 else None
|
|
self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
|
|
|
|
def forward(self, x):
|
|
# if self.flag == False:
|
|
# raise ValueError(
|
|
# f"Expected input out to have {dim} channels, but got {outdim} channels instead")
|
|
|
|
input = x
|
|
x = self.dwconv(x)
|
|
x = x.permute(0, 2, 3, 1) # (N, C, H, W) -> (N, H, W, C)
|
|
x = self.norm(x)
|
|
x = self.pwconv1(x)
|
|
x = self.act(x)
|
|
x = self.pwconv2(x)
|
|
if self.gamma is not None:
|
|
x = self.gamma * x
|
|
x = x.permute(0, 3, 1, 2) # (N, H, W, C) -> (N, C, H, W)
|
|
|
|
x = input + self.drop_path(x)
|
|
return x
|
|
|
|
|
|
class LayerNorm_s(nn.Module):
|
|
|
|
def __init__(self, normalized_shape, eps=1e-6, data_format="channels_last"):
|
|
super().__init__()
|
|
self.weight = nn.Parameter(torch.ones(normalized_shape))
|
|
self.bias = nn.Parameter(torch.zeros(normalized_shape))
|
|
self.eps = eps
|
|
self.data_format = data_format
|
|
if self.data_format not in ["channels_last", "channels_first"]:
|
|
raise NotImplementedError
|
|
self.normalized_shape = (normalized_shape,)
|
|
|
|
def forward(self, x):
|
|
if self.data_format == "channels_last":
|
|
return F.layer_norm(x, self.normalized_shape, self.weight, self.bias, self.eps)
|
|
elif self.data_format == "channels_first":
|
|
u = x.mean(1, keepdim=True)
|
|
s = (x - u).pow(2).mean(1, keepdim=True)
|
|
x = (x - u) / torch.sqrt(s + self.eps)
|
|
x = self.weight[:, None, None] * x + self.bias[:, None, None]
|
|
return x
|
|
|
|
|
|
class DropPath(nn.Module):
|
|
"""Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).
|
|
"""
|
|
|
|
def __init__(self, drop_prob=None):
|
|
super(DropPath, self).__init__()
|
|
self.drop_prob = drop_prob
|
|
|
|
def forward(self, x):
|
|
return drop_path_f(x, self.drop_prob, self.training)
|
|
|
|
|
|
def drop_path_f(x, drop_prob: float = 0., training: bool = False):
|
|
"""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 = keep_prob + torch.rand(shape, dtype=x.dtype, device=x.device)
|
|
random_tensor.floor_() # binarize
|
|
output = x.div(keep_prob) * random_tensor
|
|
return output
|
|
|
|
|
|
# ------------------------------------Convnext end -------------------------------------
|
|
|
|
class DetectMultiBackend(nn.Module):
|
|
# YOLOv5 MultiBackend class for python inference on various backends
|
|
def __init__(self, weights='yolov5s.pt', device=torch.device('cpu'), dnn=False, data=None, fp16=False, fuse=True):
|
|
# Usage:
|
|
# PyTorch: weights = *.pt
|
|
# TorchScript: *.torchscript
|
|
# ONNX Runtime: *.onnx
|
|
# ONNX OpenCV DNN: *.onnx --dnn
|
|
# OpenVINO: *_openvino_model
|
|
# CoreML: *.mlmodel
|
|
# TensorRT: *.engine
|
|
# TensorFlow SavedModel: *_saved_model
|
|
# TensorFlow GraphDef: *.pb
|
|
# TensorFlow Lite: *.tflite
|
|
# TensorFlow Edge TPU: *_edgetpu.tflite
|
|
# PaddlePaddle: *_paddle_model
|
|
from .experimental import attempt_download, attempt_load # scoped to avoid circular import
|
|
|
|
super().__init__()
|
|
w = str(weights[0] if isinstance(weights, list) else weights)
|
|
pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, paddle, triton = self._model_type(w)
|
|
fp16 &= pt or jit or onnx or engine # FP16
|
|
nhwc = coreml or saved_model or pb or tflite or edgetpu # BHWC formats (vs torch BCWH)
|
|
stride = 32 # default stride
|
|
cuda = torch.cuda.is_available() and device.type != 'cpu' # use CUDA
|
|
if not (pt or triton):
|
|
w = attempt_download(w) # download if not local
|
|
|
|
if pt: # PyTorch
|
|
model = attempt_load(weights if isinstance(weights, list) else w, device=device, inplace=True, fuse=fuse)
|
|
stride = max(int(model.stride.max()), 32) # model stride
|
|
names = model.module.names if hasattr(model, 'module') else model.names # get class names
|
|
model.half() if fp16 else model.float()
|
|
self.model = model # explicitly assign for to(), cpu(), cuda(), half()
|
|
elif jit: # TorchScript
|
|
LOGGER.info(f'Loading {w} for TorchScript inference...')
|
|
extra_files = {'config.txt': ''} # model metadata
|
|
model = torch.jit.load(w, _extra_files=extra_files, map_location=device)
|
|
model.half() if fp16 else model.float()
|
|
if extra_files['config.txt']: # load metadata dict
|
|
d = json.loads(extra_files['config.txt'],
|
|
object_hook=lambda d: {int(k) if k.isdigit() else k: v
|
|
for k, v in d.items()})
|
|
stride, names = int(d['stride']), d['names']
|
|
elif dnn: # ONNX OpenCV DNN
|
|
LOGGER.info(f'Loading {w} for ONNX OpenCV DNN inference...')
|
|
check_requirements('opencv-python>=4.5.4')
|
|
net = cv2.dnn.readNetFromONNX(w)
|
|
elif onnx: # ONNX Runtime
|
|
LOGGER.info(f'Loading {w} for ONNX Runtime inference...')
|
|
check_requirements(('onnx', 'onnxruntime-gpu' if cuda else 'onnxruntime'))
|
|
import onnxruntime
|
|
providers = ['CUDAExecutionProvider', 'CPUExecutionProvider'] if cuda else ['CPUExecutionProvider']
|
|
session = onnxruntime.InferenceSession(w, providers=providers)
|
|
output_names = [x.name for x in session.get_outputs()]
|
|
meta = session.get_modelmeta().custom_metadata_map # metadata
|
|
if 'stride' in meta:
|
|
stride, names = int(meta['stride']), eval(meta['names'])
|
|
elif xml: # OpenVINO
|
|
LOGGER.info(f'Loading {w} for OpenVINO inference...')
|
|
check_requirements('openvino') # requires openvino-dev: https://pypi.org/project/openvino-dev/
|
|
from openvino.runtime import Core, Layout, get_batch
|
|
ie = Core()
|
|
if not Path(w).is_file(): # if not *.xml
|
|
w = next(Path(w).glob('*.xml')) # get *.xml file from *_openvino_model dir
|
|
network = ie.read_model(model=w, weights=Path(w).with_suffix('.bin'))
|
|
if network.get_parameters()[0].get_layout().empty:
|
|
network.get_parameters()[0].set_layout(Layout('NCHW'))
|
|
batch_dim = get_batch(network)
|
|
if batch_dim.is_static:
|
|
batch_size = batch_dim.get_length()
|
|
executable_network = ie.compile_model(network, device_name='CPU') # device_name="MYRIAD" for Intel NCS2
|
|
stride, names = self._load_metadata(Path(w).with_suffix('.yaml')) # load metadata
|
|
elif engine: # TensorRT
|
|
LOGGER.info(f'Loading {w} for TensorRT inference...')
|
|
import tensorrt as trt # https://developer.nvidia.com/nvidia-tensorrt-download
|
|
check_version(trt.__version__, '7.0.0', hard=True) # require tensorrt>=7.0.0
|
|
if device.type == 'cpu':
|
|
device = torch.device('cuda:0')
|
|
Binding = namedtuple('Binding', ('name', 'dtype', 'shape', 'data', 'ptr'))
|
|
logger = trt.Logger(trt.Logger.INFO)
|
|
with open(w, 'rb') as f, trt.Runtime(logger) as runtime:
|
|
model = runtime.deserialize_cuda_engine(f.read())
|
|
context = model.create_execution_context()
|
|
bindings = OrderedDict()
|
|
output_names = []
|
|
fp16 = False # default updated below
|
|
dynamic = False
|
|
for i in range(model.num_bindings):
|
|
name = model.get_binding_name(i)
|
|
dtype = trt.nptype(model.get_binding_dtype(i))
|
|
if model.binding_is_input(i):
|
|
if -1 in tuple(model.get_binding_shape(i)): # dynamic
|
|
dynamic = True
|
|
context.set_binding_shape(i, tuple(model.get_profile_shape(0, i)[2]))
|
|
if dtype == np.float16:
|
|
fp16 = True
|
|
else: # output
|
|
output_names.append(name)
|
|
shape = tuple(context.get_binding_shape(i))
|
|
im = torch.from_numpy(np.empty(shape, dtype=dtype)).to(device)
|
|
bindings[name] = Binding(name, dtype, shape, im, int(im.data_ptr()))
|
|
binding_addrs = OrderedDict((n, d.ptr) for n, d in bindings.items())
|
|
batch_size = bindings['images'].shape[0] # if dynamic, this is instead max batch size
|
|
elif coreml: # CoreML
|
|
LOGGER.info(f'Loading {w} for CoreML inference...')
|
|
import coremltools as ct
|
|
model = ct.models.MLModel(w)
|
|
elif saved_model: # TF SavedModel
|
|
LOGGER.info(f'Loading {w} for TensorFlow SavedModel inference...')
|
|
import tensorflow as tf
|
|
keras = False # assume TF1 saved_model
|
|
model = tf.keras.models.load_model(w) if keras else tf.saved_model.load(w)
|
|
elif pb: # GraphDef https://www.tensorflow.org/guide/migrate#a_graphpb_or_graphpbtxt
|
|
LOGGER.info(f'Loading {w} for TensorFlow GraphDef inference...')
|
|
import tensorflow as tf
|
|
|
|
def wrap_frozen_graph(gd, inputs, outputs):
|
|
x = tf.compat.v1.wrap_function(lambda: tf.compat.v1.import_graph_def(gd, name=''), []) # wrapped
|
|
ge = x.graph.as_graph_element
|
|
return x.prune(tf.nest.map_structure(ge, inputs), tf.nest.map_structure(ge, outputs))
|
|
|
|
def gd_outputs(gd):
|
|
name_list, input_list = [], []
|
|
for node in gd.node: # tensorflow.core.framework.node_def_pb2.NodeDef
|
|
name_list.append(node.name)
|
|
input_list.extend(node.input)
|
|
return sorted(f'{x}:0' for x in list(set(name_list) - set(input_list)) if not x.startswith('NoOp'))
|
|
|
|
gd = tf.Graph().as_graph_def() # TF GraphDef
|
|
with open(w, 'rb') as f:
|
|
gd.ParseFromString(f.read())
|
|
frozen_func = wrap_frozen_graph(gd, inputs='x:0', outputs=gd_outputs(gd))
|
|
elif tflite or edgetpu: # https://www.tensorflow.org/lite/guide/python#install_tensorflow_lite_for_python
|
|
try: # https://coral.ai/docs/edgetpu/tflite-python/#update-existing-tf-lite-code-for-the-edge-tpu
|
|
from tflite_runtime.interpreter import Interpreter, load_delegate
|
|
except ImportError:
|
|
import tensorflow as tf
|
|
Interpreter, load_delegate = tf.lite.Interpreter, tf.lite.experimental.load_delegate,
|
|
if edgetpu: # TF Edge TPU https://coral.ai/software/#edgetpu-runtime
|
|
LOGGER.info(f'Loading {w} for TensorFlow Lite Edge TPU inference...')
|
|
delegate = {
|
|
'Linux': 'libedgetpu.so.1',
|
|
'Darwin': 'libedgetpu.1.dylib',
|
|
'Windows': 'edgetpu.dll'}[platform.system()]
|
|
interpreter = Interpreter(model_path=w, experimental_delegates=[load_delegate(delegate)])
|
|
else: # TFLite
|
|
LOGGER.info(f'Loading {w} for TensorFlow Lite inference...')
|
|
interpreter = Interpreter(model_path=w) # load TFLite model
|
|
interpreter.allocate_tensors() # allocate
|
|
input_details = interpreter.get_input_details() # inputs
|
|
output_details = interpreter.get_output_details() # outputs
|
|
# load metadata
|
|
with contextlib.suppress(zipfile.BadZipFile):
|
|
with zipfile.ZipFile(w, 'r') as model:
|
|
meta_file = model.namelist()[0]
|
|
meta = ast.literal_eval(model.read(meta_file).decode('utf-8'))
|
|
stride, names = int(meta['stride']), meta['names']
|
|
elif tfjs: # TF.js
|
|
raise NotImplementedError('ERROR: YOLOv5 TF.js inference is not supported')
|
|
elif paddle: # PaddlePaddle
|
|
LOGGER.info(f'Loading {w} for PaddlePaddle inference...')
|
|
check_requirements('paddlepaddle-gpu' if cuda else 'paddlepaddle')
|
|
import paddle.inference as pdi
|
|
if not Path(w).is_file(): # if not *.pdmodel
|
|
w = next(Path(w).rglob('*.pdmodel')) # get *.pdmodel file from *_paddle_model dir
|
|
weights = Path(w).with_suffix('.pdiparams')
|
|
config = pdi.Config(str(w), str(weights))
|
|
if cuda:
|
|
config.enable_use_gpu(memory_pool_init_size_mb=2048, device_id=0)
|
|
predictor = pdi.create_predictor(config)
|
|
input_handle = predictor.get_input_handle(predictor.get_input_names()[0])
|
|
output_names = predictor.get_output_names()
|
|
elif triton: # NVIDIA Triton Inference Server
|
|
LOGGER.info(f'Using {w} as Triton Inference Server...')
|
|
check_requirements('tritonclient[all]')
|
|
from thirdTool.yolo5x_qr.utils.triton import TritonRemoteModel
|
|
model = TritonRemoteModel(url=w)
|
|
nhwc = model.runtime.startswith('tensorflow')
|
|
else:
|
|
raise NotImplementedError(f'ERROR: {w} is not a supported format')
|
|
|
|
# class names
|
|
if 'names' not in locals():
|
|
names = yaml_load(data)['names'] if data else {i: f'class{i}' for i in range(999)}
|
|
if names[0] == 'n01440764' and len(names) == 1000: # ImageNet
|
|
names = yaml_load(ROOT / 'data/ImageNet.yaml')['names'] # human-readable names
|
|
|
|
self.__dict__.update(locals()) # assign all variables to self
|
|
|
|
def forward(self, im, augment=False, visualize=False):
|
|
# YOLOv5 MultiBackend inference
|
|
b, ch, h, w = im.shape # batch, channel, height, width
|
|
if self.fp16 and im.dtype != torch.float16:
|
|
im = im.half() # to FP16
|
|
if self.nhwc:
|
|
im = im.permute(0, 2, 3, 1) # torch BCHW to numpy BHWC shape(1,320,192,3)
|
|
|
|
if self.pt: # PyTorch
|
|
y = self.model(im, augment=augment, visualize=visualize) if augment or visualize else self.model(im)
|
|
elif self.jit: # TorchScript
|
|
y = self.model(im)
|
|
elif self.dnn: # ONNX OpenCV DNN
|
|
im = im.cpu().numpy() # torch to numpy
|
|
self.net.setInput(im)
|
|
y = self.net.forward()
|
|
elif self.onnx: # ONNX Runtime
|
|
im = im.cpu().numpy() # torch to numpy
|
|
y = self.session.run(self.output_names, {self.session.get_inputs()[0].name: im})
|
|
elif self.xml: # OpenVINO
|
|
im = im.cpu().numpy() # FP32
|
|
y = list(self.executable_network([im]).values())
|
|
elif self.engine: # TensorRT
|
|
if self.dynamic and im.shape != self.bindings['images'].shape:
|
|
i = self.model.get_binding_index('images')
|
|
self.context.set_binding_shape(i, im.shape) # reshape if dynamic
|
|
self.bindings['images'] = self.bindings['images']._replace(shape=im.shape)
|
|
for name in self.output_names:
|
|
i = self.model.get_binding_index(name)
|
|
self.bindings[name].data.resize_(tuple(self.context.get_binding_shape(i)))
|
|
s = self.bindings['images'].shape
|
|
assert im.shape == s, f"input size {im.shape} {'>' if self.dynamic else 'not equal to'} max model size {s}"
|
|
self.binding_addrs['images'] = int(im.data_ptr())
|
|
self.context.execute_v2(list(self.binding_addrs.values()))
|
|
y = [self.bindings[x].data for x in sorted(self.output_names)]
|
|
elif self.coreml: # CoreML
|
|
im = im.cpu().numpy()
|
|
im = Image.fromarray((im[0] * 255).astype('uint8'))
|
|
# im = im.resize((192, 320), Image.ANTIALIAS)
|
|
y = self.model.predict({'image': im}) # coordinates are xywh normalized
|
|
if 'confidence' in y:
|
|
box = xywh2xyxy(y['coordinates'] * [[w, h, w, h]]) # xyxy pixels
|
|
conf, cls = y['confidence'].max(1), y['confidence'].argmax(1).astype(np.float)
|
|
y = np.concatenate((box, conf.reshape(-1, 1), cls.reshape(-1, 1)), 1)
|
|
else:
|
|
y = list(reversed(y.values())) # reversed for segmentation models (pred, proto)
|
|
elif self.paddle: # PaddlePaddle
|
|
im = im.cpu().numpy().astype(np.float32)
|
|
self.input_handle.copy_from_cpu(im)
|
|
self.predictor.run()
|
|
y = [self.predictor.get_output_handle(x).copy_to_cpu() for x in self.output_names]
|
|
elif self.triton: # NVIDIA Triton Inference Server
|
|
y = self.model(im)
|
|
else: # TensorFlow (SavedModel, GraphDef, Lite, Edge TPU)
|
|
im = im.cpu().numpy()
|
|
if self.saved_model: # SavedModel
|
|
y = self.model(im, training=False) if self.keras else self.model(im)
|
|
elif self.pb: # GraphDef
|
|
y = self.frozen_func(x=self.tf.constant(im))
|
|
else: # Lite or Edge TPU
|
|
input = self.input_details[0]
|
|
int8 = input['dtype'] == np.uint8 # is TFLite quantized uint8 model
|
|
if int8:
|
|
scale, zero_point = input['quantization']
|
|
im = (im / scale + zero_point).astype(np.uint8) # de-scale
|
|
self.interpreter.set_tensor(input['index'], im)
|
|
self.interpreter.invoke()
|
|
y = []
|
|
for output in self.output_details:
|
|
x = self.interpreter.get_tensor(output['index'])
|
|
if int8:
|
|
scale, zero_point = output['quantization']
|
|
x = (x.astype(np.float32) - zero_point) * scale # re-scale
|
|
y.append(x)
|
|
y = [x if isinstance(x, np.ndarray) else x.numpy() for x in y]
|
|
y[0][..., :4] *= [w, h, w, h] # xywh normalized to pixels
|
|
|
|
if isinstance(y, (list, tuple)):
|
|
return self.from_numpy(y[0]) if len(y) == 1 else [self.from_numpy(x) for x in y]
|
|
else:
|
|
return self.from_numpy(y)
|
|
|
|
def from_numpy(self, x):
|
|
return torch.from_numpy(x).to(self.device) if isinstance(x, np.ndarray) else x
|
|
|
|
def warmup(self, imgsz=(1, 3, 640, 640)):
|
|
# Warmup model by running inference once
|
|
warmup_types = self.pt, self.jit, self.onnx, self.engine, self.saved_model, self.pb, self.triton
|
|
if any(warmup_types) and (self.device.type != 'cpu' or self.triton):
|
|
im = torch.empty(*imgsz, dtype=torch.half if self.fp16 else torch.float, device=self.device) # input
|
|
for _ in range(2 if self.jit else 1): #
|
|
self.forward(im) # warmup
|
|
|
|
@staticmethod
|
|
def _model_type(p='path/to/model.pt'):
|
|
# Return model type from model path, i.e. path='path/to/model.onnx' -> type=onnx
|
|
# types = [pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, paddle]
|
|
# from export import export_formats
|
|
from thirdTool.yolo5x_qr.utils.downloads import is_url
|
|
sf = list(export_formats().Suffix) # export suffixes
|
|
if not is_url(p, check=False):
|
|
check_suffix(p, sf) # checks
|
|
url = urlparse(p) # if url may be Triton inference server
|
|
types = [s in Path(p).name for s in sf]
|
|
types[8] &= not types[9] # tflite &= not edgetpu
|
|
triton = not any(types) and all([any(s in url.scheme for s in ['http', 'grpc']), url.netloc])
|
|
return types + [triton]
|
|
|
|
@staticmethod
|
|
def _load_metadata(f=Path('path/to/meta.yaml')):
|
|
# Load metadata from meta.yaml if it exists
|
|
if f.exists():
|
|
d = yaml_load(f)
|
|
return d['stride'], d['names'] # assign stride, names
|
|
return None, None
|
|
|
|
|
|
class AutoShape(nn.Module):
|
|
# YOLOv5 input-robust model wrapper for passing cv2/np/PIL/torch inputs. Includes preprocessing, inference and NMS
|
|
conf = 0.25 # NMS confidence threshold
|
|
iou = 0.45 # NMS IoU threshold
|
|
agnostic = False # NMS class-agnostic
|
|
multi_label = False # NMS multiple labels per box
|
|
classes = None # (optional list) filter by class, i.e. = [0, 15, 16] for COCO persons, cats and dogs
|
|
max_det = 1000 # maximum number of detections per image
|
|
amp = False # Automatic Mixed Precision (AMP) inference
|
|
|
|
def __init__(self, model, verbose=True):
|
|
super().__init__()
|
|
if verbose:
|
|
LOGGER.info('Adding AutoShape... ')
|
|
copy_attr(self, model, include=('yaml', 'nc', 'hyp', 'names', 'stride', 'abc'), exclude=()) # copy attributes
|
|
self.dmb = isinstance(model, DetectMultiBackend) # DetectMultiBackend() instance
|
|
self.pt = not self.dmb or model.pt # PyTorch model
|
|
self.model = model.eval()
|
|
if self.pt:
|
|
m = self.model.model.model[-1] if self.dmb else self.model.model[-1] # Detect()
|
|
m.inplace = False # Detect.inplace=False for safe multithread inference
|
|
m.export = True # do not output loss values
|
|
|
|
def _apply(self, fn):
|
|
# Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffers
|
|
self = super()._apply(fn)
|
|
if self.pt:
|
|
m = self.model.model.model[-1] if self.dmb else self.model.model[-1] # Detect()
|
|
m.stride = fn(m.stride)
|
|
m.grid = list(map(fn, m.grid))
|
|
if isinstance(m.anchor_grid, list):
|
|
m.anchor_grid = list(map(fn, m.anchor_grid))
|
|
return self
|
|
|
|
@smart_inference_mode()
|
|
def forward(self, ims, size=640, augment=False, profile=False):
|
|
# Inference from various sources. For size(height=640, width=1280), RGB images example inputs are:
|
|
# file: ims = 'data/images/zidane.jpg' # str or PosixPath
|
|
# URI: = 'https://ultralytics.com/images/zidane.jpg'
|
|
# OpenCV: = cv2.imread('image.jpg')[:,:,::-1] # HWC BGR to RGB x(640,1280,3)
|
|
# PIL: = Image.open('image.jpg') or ImageGrab.grab() # HWC x(640,1280,3)
|
|
# numpy: = np.zeros((640,1280,3)) # HWC
|
|
# torch: = torch.zeros(16,3,320,640) # BCHW (scaled to size=640, 0-1 values)
|
|
# multiple: = [Image.open('image1.jpg'), Image.open('image2.jpg'), ...] # list of images
|
|
|
|
dt = (Profile(), Profile(), Profile())
|
|
with dt[0]:
|
|
if isinstance(size, int): # expand
|
|
size = (size, size)
|
|
p = next(self.model.parameters()) if self.pt else torch.empty(1, device=self.model.device) # param
|
|
autocast = self.amp and (p.device.type != 'cpu') # Automatic Mixed Precision (AMP) inference
|
|
if isinstance(ims, torch.Tensor): # torch
|
|
with amp.autocast(autocast):
|
|
return self.model(ims.to(p.device).type_as(p), augment=augment) # inference
|
|
|
|
# Pre-process
|
|
n, ims = (len(ims), list(ims)) if isinstance(ims, (list, tuple)) else (1, [ims]) # number, list of images
|
|
shape0, shape1, files = [], [], [] # image and inference shapes, filenames
|
|
for i, im in enumerate(ims):
|
|
f = f'image{i}' # filename
|
|
if isinstance(im, (str, Path)): # filename or uri
|
|
im, f = Image.open(requests.get(im, stream=True).raw if str(im).startswith('http') else im), im
|
|
im = np.asarray(exif_transpose(im))
|
|
elif isinstance(im, Image.Image): # PIL Image
|
|
im, f = np.asarray(exif_transpose(im)), getattr(im, 'filename', f) or f
|
|
files.append(Path(f).with_suffix('.jpg').name)
|
|
if im.shape[0] < 5: # image in CHW
|
|
im = im.transpose((1, 2, 0)) # reverse dataloader .transpose(2, 0, 1)
|
|
im = im[..., :3] if im.ndim == 3 else cv2.cvtColor(im, cv2.COLOR_GRAY2BGR) # enforce 3ch input
|
|
s = im.shape[:2] # HWC
|
|
shape0.append(s) # image shape
|
|
g = max(size) / max(s) # gain
|
|
shape1.append([int(y * g) for y in s])
|
|
ims[i] = im if im.data.contiguous else np.ascontiguousarray(im) # update
|
|
shape1 = [make_divisible(x, self.stride) for x in np.array(shape1).max(0)] # inf shape
|
|
x = [letterbox(im, shape1, auto=False)[0] for im in ims] # pad
|
|
x = np.ascontiguousarray(np.array(x).transpose((0, 3, 1, 2))) # stack and BHWC to BCHW
|
|
x = torch.from_numpy(x).to(p.device).type_as(p) / 255 # uint8 to fp16/32
|
|
|
|
with amp.autocast(autocast):
|
|
# Inference
|
|
with dt[1]:
|
|
y = self.model(x, augment=augment) # forward
|
|
|
|
# Post-process
|
|
with dt[2]:
|
|
y = non_max_suppression(y if self.dmb else y[0],
|
|
self.conf,
|
|
self.iou,
|
|
self.classes,
|
|
self.agnostic,
|
|
self.multi_label,
|
|
max_det=self.max_det) # NMS
|
|
for i in range(n):
|
|
scale_boxes(shape1, y[i][:, :4], shape0[i])
|
|
|
|
return Detections(ims, y, files, dt, self.names, x.shape)
|
|
|
|
|
|
class Detections:
|
|
# YOLOv5 detections class for inference results
|
|
def __init__(self, ims, pred, files, times=(0, 0, 0), names=None, shape=None):
|
|
super().__init__()
|
|
d = pred[0].device # device
|
|
gn = [torch.tensor([*(im.shape[i] for i in [1, 0, 1, 0]), 1, 1], device=d) for im in ims] # normalizations
|
|
self.ims = ims # list of images as numpy arrays
|
|
self.pred = pred # list of tensors pred[0] = (xyxy, conf, cls)
|
|
self.names = names # class names
|
|
self.files = files # image filenames
|
|
self.times = times # profiling times
|
|
self.xyxy = pred # xyxy pixels
|
|
self.xywh = [xyxy2xywh(x) for x in pred] # xywh pixels
|
|
self.xyxyn = [x / g for x, g in zip(self.xyxy, gn)] # xyxy normalized
|
|
self.xywhn = [x / g for x, g in zip(self.xywh, gn)] # xywh normalized
|
|
self.n = len(self.pred) # number of images (batch size)
|
|
self.t = tuple(x.t / self.n * 1E3 for x in times) # timestamps (ms)
|
|
self.s = tuple(shape) # inference BCHW shape
|
|
|
|
def _run(self, pprint=False, show=False, save=False, crop=False, render=False, labels=True, save_dir=Path('')):
|
|
s, crops = '', []
|
|
for i, (im, pred) in enumerate(zip(self.ims, self.pred)):
|
|
s += f'\nimage {i + 1}/{len(self.pred)}: {im.shape[0]}x{im.shape[1]} ' # string
|
|
if pred.shape[0]:
|
|
for c in pred[:, -1].unique():
|
|
n = (pred[:, -1] == c).sum() # detections per class
|
|
s += f"{n} {self.names[int(c)]}{'s' * (n > 1)}, " # add to string
|
|
s = s.rstrip(', ')
|
|
if show or save or render or crop:
|
|
annotator = Annotator(im, example=str(self.names))
|
|
for *box, conf, cls in reversed(pred): # xyxy, confidence, class
|
|
label = f'{self.names[int(cls)]} {conf:.2f}'
|
|
if crop:
|
|
file = save_dir / 'crops' / self.names[int(cls)] / self.files[i] if save else None
|
|
crops.append({
|
|
'box': box,
|
|
'conf': conf,
|
|
'cls': cls,
|
|
'label': label,
|
|
'im': save_one_box(box, im, file=file, save=save)})
|
|
else: # all others
|
|
annotator.box_label(box, label if labels else '', color=colors(cls))
|
|
im = annotator.im
|
|
else:
|
|
s += '(no detections)'
|
|
|
|
im = Image.fromarray(im.astype(np.uint8)) if isinstance(im, np.ndarray) else im # from np
|
|
if show:
|
|
if is_jupyter():
|
|
from IPython.display import display
|
|
display(im)
|
|
else:
|
|
im.show(self.files[i])
|
|
if save:
|
|
f = self.files[i]
|
|
im.save(save_dir / f) # save
|
|
if i == self.n - 1:
|
|
LOGGER.info(f"Saved {self.n} image{'s' * (self.n > 1)} to {colorstr('bold', save_dir)}")
|
|
if render:
|
|
self.ims[i] = np.asarray(im)
|
|
if pprint:
|
|
s = s.lstrip('\n')
|
|
return f'{s}\nSpeed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {self.s}' % self.t
|
|
if crop:
|
|
if save:
|
|
LOGGER.info(f'Saved results to {save_dir}\n')
|
|
return crops
|
|
|
|
@TryExcept('Showing images is not supported in this environment')
|
|
def show(self, labels=True):
|
|
self._run(show=True, labels=labels) # show results
|
|
|
|
def save(self, labels=True, save_dir='runs/detect/exp', exist_ok=False):
|
|
save_dir = increment_path(save_dir, exist_ok, mkdir=True) # increment save_dir
|
|
self._run(save=True, labels=labels, save_dir=save_dir) # save results
|
|
|
|
def crop(self, save=True, save_dir='runs/detect/exp', exist_ok=False):
|
|
save_dir = increment_path(save_dir, exist_ok, mkdir=True) if save else None
|
|
return self._run(crop=True, save=save, save_dir=save_dir) # crop results
|
|
|
|
def render(self, labels=True):
|
|
self._run(render=True, labels=labels) # render results
|
|
return self.ims
|
|
|
|
def pandas(self):
|
|
# return detections as pandas DataFrames, i.e. print(results.pandas().xyxy[0])
|
|
new = copy(self) # return copy
|
|
ca = 'xmin', 'ymin', 'xmax', 'ymax', 'confidence', 'class', 'name' # xyxy columns
|
|
cb = 'xcenter', 'ycenter', 'width', 'height', 'confidence', 'class', 'name' # xywh columns
|
|
for k, c in zip(['xyxy', 'xyxyn', 'xywh', 'xywhn'], [ca, ca, cb, cb]):
|
|
a = [[x[:5] + [int(x[5]), self.names[int(x[5])]] for x in x.tolist()] for x in getattr(self, k)] # update
|
|
setattr(new, k, [pd.DataFrame(x, columns=c) for x in a])
|
|
return new
|
|
|
|
def tolist(self):
|
|
# return a list of Detections objects, i.e. 'for result in results.tolist():'
|
|
r = range(self.n) # iterable
|
|
x = [Detections([self.ims[i]], [self.pred[i]], [self.files[i]], self.times, self.names, self.s) for i in r]
|
|
# for d in x:
|
|
# for k in ['ims', 'pred', 'xyxy', 'xyxyn', 'xywh', 'xywhn']:
|
|
# setattr(d, k, getattr(d, k)[0]) # pop out of list
|
|
return x
|
|
|
|
def print(self):
|
|
LOGGER.info(self.__str__())
|
|
|
|
def __len__(self): # override len(results)
|
|
return self.n
|
|
|
|
def __str__(self): # override print(results)
|
|
return self._run(pprint=True) # print results
|
|
|
|
def __repr__(self):
|
|
return f'YOLOv5 {self.__class__} instance\n' + self.__str__()
|
|
|
|
|
|
class Proto(nn.Module):
|
|
# YOLOv5 mask Proto module for segmentation models
|
|
def __init__(self, c1, c_=256, c2=32): # ch_in, number of protos, number of masks
|
|
super().__init__()
|
|
self.cv1 = Conv(c1, c_, k=3)
|
|
self.upsample = nn.Upsample(scale_factor=2, mode='nearest')
|
|
self.cv2 = Conv(c_, c_, k=3)
|
|
self.cv3 = Conv(c_, c2)
|
|
|
|
def forward(self, x):
|
|
return self.cv3(self.cv2(self.upsample(self.cv1(x))))
|
|
|
|
|
|
class Classify(nn.Module):
|
|
# YOLOv5 classification head, i.e. x(b,c1,20,20) to x(b,c2)
|
|
def __init__(self,
|
|
c1,
|
|
c2,
|
|
k=1,
|
|
s=1,
|
|
p=None,
|
|
g=1,
|
|
dropout_p=0.0): # ch_in, ch_out, kernel, stride, padding, groups, dropout probability
|
|
super().__init__()
|
|
c_ = 1280 # efficientnet_b0 size
|
|
self.conv = Conv(c1, c_, k, s, autopad(k, p), g)
|
|
self.pool = nn.AdaptiveAvgPool2d(1) # to x(b,c_,1,1)
|
|
self.drop = nn.Dropout(p=dropout_p, inplace=True)
|
|
self.linear = nn.Linear(c_, c2) # to x(b,c2)
|
|
|
|
def forward(self, x):
|
|
if isinstance(x, list):
|
|
x = torch.cat(x, 1)
|
|
return self.linear(self.drop(self.pool(self.conv(x)).flatten(1)))
|
|
|
|
|
|
from torch.autograd import Function
|
|
|
|
# by AI&CV SOCA ####### start #########################
|
|
# SOCA moudle 单幅图像超分辨率
|
|
class Covpool(Function):
|
|
@staticmethod
|
|
def forward(ctx, input):
|
|
x = input
|
|
batchSize = x.data.shape[0]
|
|
dim = x.data.shape[1]
|
|
h = x.data.shape[2]
|
|
w = x.data.shape[3]
|
|
M = h * w
|
|
x = x.reshape(batchSize, dim, M)
|
|
I_hat = (-1. / M / M) * torch.ones(M, M, device=x.device) + (1. / M) * torch.eye(M, M, device=x.device)
|
|
I_hat = I_hat.view(1, M, M).repeat(batchSize, 1, 1).type(x.dtype)
|
|
y = x.bmm(I_hat).bmm(x.transpose(1, 2))
|
|
ctx.save_for_backward(input, I_hat)
|
|
return y
|
|
|
|
@staticmethod
|
|
def backward(ctx, grad_output):
|
|
input, I_hat = ctx.saved_tensors
|
|
x = input
|
|
batchSize = x.data.shape[0]
|
|
dim = x.data.shape[1]
|
|
h = x.data.shape[2]
|
|
w = x.data.shape[3]
|
|
M = h * w
|
|
x = x.reshape(batchSize, dim, M)
|
|
grad_input = grad_output + grad_output.transpose(1, 2)
|
|
grad_input = grad_input.bmm(x).bmm(I_hat)
|
|
grad_input = grad_input.reshape(batchSize, dim, h, w)
|
|
return grad_input
|
|
|
|
|
|
class Sqrtm(Function):
|
|
@staticmethod
|
|
def forward(ctx, input, iterN):
|
|
x = input
|
|
batchSize = x.data.shape[0]
|
|
dim = x.data.shape[1]
|
|
dtype = x.dtype
|
|
I3 = 3.0 * torch.eye(dim, dim, device=x.device).view(1, dim, dim).repeat(batchSize, 1, 1).type(dtype)
|
|
normA = (1.0 / 3.0) * x.mul(I3).sum(dim=1).sum(dim=1)
|
|
A = x.div(normA.view(batchSize, 1, 1).expand_as(x))
|
|
Y = torch.zeros(batchSize, iterN, dim, dim, requires_grad=False, device=x.device)
|
|
Z = torch.eye(dim, dim, device=x.device).view(1, dim, dim).repeat(batchSize, iterN, 1, 1)
|
|
if iterN < 2:
|
|
ZY = 0.5 * (I3 - A)
|
|
Y[:, 0, :, :] = A.bmm(ZY)
|
|
else:
|
|
ZY = 0.5 * (I3 - A)
|
|
Y[:, 0, :, :] = A.bmm(ZY)
|
|
Z[:, 0, :, :] = ZY
|
|
for i in range(1, iterN - 1):
|
|
ZY = 0.5 * (I3 - Z[:, i - 1, :, :].bmm(Y[:, i - 1, :, :]))
|
|
Y[:, i, :, :] = Y[:, i - 1, :, :].bmm(ZY)
|
|
Z[:, i, :, :] = ZY.bmm(Z[:, i - 1, :, :])
|
|
ZY = 0.5 * Y[:, iterN - 2, :, :].bmm(I3 - Z[:, iterN - 2, :, :].bmm(Y[:, iterN - 2, :, :]))
|
|
y = ZY * torch.sqrt(normA).view(batchSize, 1, 1).expand_as(x)
|
|
ctx.save_for_backward(input, A, ZY, normA, Y, Z)
|
|
ctx.iterN = iterN
|
|
return y
|
|
|
|
@staticmethod
|
|
def backward(ctx, grad_output):
|
|
input, A, ZY, normA, Y, Z = ctx.saved_tensors
|
|
iterN = ctx.iterN
|
|
x = input
|
|
batchSize = x.data.shape[0]
|
|
dim = x.data.shape[1]
|
|
dtype = x.dtype
|
|
der_postCom = grad_output * torch.sqrt(normA).view(batchSize, 1, 1).expand_as(x)
|
|
der_postComAux = (grad_output * ZY).sum(dim=1).sum(dim=1).div(2 * torch.sqrt(normA))
|
|
I3 = 3.0 * torch.eye(dim, dim, device=x.device).view(1, dim, dim).repeat(batchSize, 1, 1).type(dtype)
|
|
if iterN < 2:
|
|
der_NSiter = 0.5 * (der_postCom.bmm(I3 - A) - A.bmm(der_sacleTrace))
|
|
else:
|
|
dldY = 0.5 * (der_postCom.bmm(I3 - Y[:, iterN - 2, :, :].bmm(Z[:, iterN - 2, :, :])) -
|
|
Z[:, iterN - 2, :, :].bmm(Y[:, iterN - 2, :, :]).bmm(der_postCom))
|
|
dldZ = -0.5 * Y[:, iterN - 2, :, :].bmm(der_postCom).bmm(Y[:, iterN - 2, :, :])
|
|
for i in range(iterN - 3, -1, -1):
|
|
YZ = I3 - Y[:, i, :, :].bmm(Z[:, i, :, :])
|
|
ZY = Z[:, i, :, :].bmm(Y[:, i, :, :])
|
|
dldY_ = 0.5 * (dldY.bmm(YZ) -
|
|
Z[:, i, :, :].bmm(dldZ).bmm(Z[:, i, :, :]) -
|
|
ZY.bmm(dldY))
|
|
dldZ_ = 0.5 * (YZ.bmm(dldZ) -
|
|
Y[:, i, :, :].bmm(dldY).bmm(Y[:, i, :, :]) -
|
|
dldZ.bmm(ZY))
|
|
dldY = dldY_
|
|
dldZ = dldZ_
|
|
der_NSiter = 0.5 * (dldY.bmm(I3 - A) - dldZ - A.bmm(dldY))
|
|
grad_input = der_NSiter.div(normA.view(batchSize, 1, 1).expand_as(x))
|
|
grad_aux = der_NSiter.mul(x).sum(dim=1).sum(dim=1)
|
|
for i in range(batchSize):
|
|
grad_input[i, :, :] += (der_postComAux[i] \
|
|
- grad_aux[i] / (normA[i] * normA[i])) \
|
|
* torch.ones(dim, device=x.device).diag()
|
|
return grad_input, None
|
|
|
|
|
|
def CovpoolLayer(var):
|
|
return Covpool.apply(var)
|
|
|
|
|
|
def SqrtmLayer(var, iterN):
|
|
return Sqrtm.apply(var, iterN)
|
|
|
|
|
|
class SOCA(nn.Module):
|
|
# second-order Channel attention
|
|
def __init__(self, channel, reduction=8):
|
|
super(SOCA, self).__init__()
|
|
self.max_pool = nn.MaxPool2d(kernel_size=2)
|
|
|
|
self.conv_du = nn.Sequential(
|
|
nn.Conv2d(channel, channel // reduction, 1, padding=0, bias=True),
|
|
nn.ReLU(inplace=True),
|
|
nn.Conv2d(channel // reduction, channel, 1, padding=0, bias=True),
|
|
nn.Sigmoid()
|
|
)
|
|
|
|
def forward(self, x):
|
|
batch_size, C, h, w = x.shape # x: NxCxHxW
|
|
N = int(h * w)
|
|
min_h = min(h, w)
|
|
h1 = 1000
|
|
w1 = 1000
|
|
if h < h1 and w < w1:
|
|
x_sub = x
|
|
elif h < h1 and w > w1:
|
|
W = (w - w1) // 2
|
|
x_sub = x[:, :, :, W:(W + w1)]
|
|
elif w < w1 and h > h1:
|
|
H = (h - h1) // 2
|
|
x_sub = x[:, :, H:H + h1, :]
|
|
else:
|
|
H = (h - h1) // 2
|
|
W = (w - w1) // 2
|
|
x_sub = x[:, :, H:(H + h1), W:(W + w1)]
|
|
cov_mat = CovpoolLayer(x_sub) # Global Covariance pooling layer
|
|
cov_mat_sqrt = SqrtmLayer(cov_mat,
|
|
5) # Matrix square root layer( including pre-norm,Newton-Schulz iter. and post-com. with 5 iteration)
|
|
cov_mat_sum = torch.mean(cov_mat_sqrt, 1)
|
|
cov_mat_sum = cov_mat_sum.view(batch_size, C, 1, 1)
|
|
y_cov = self.conv_du(cov_mat_sum)
|
|
return y_cov * x
|
|
|
|
|
|
# by AI&CV SOCA ####### end ######################### |