|
"""YOLOv5-specific modules |
|
|
|
Usage: |
|
$ python path/to/models/yolo.py --cfg yolov5s.yaml |
|
""" |
|
|
|
import argparse |
|
import sys |
|
from copy import deepcopy |
|
from pathlib import Path |
|
|
|
FILE = Path(__file__).absolute() |
|
sys.path.append(FILE.parents[1].as_posix()) |
|
|
|
from models.common import * |
|
from models.experimental import * |
|
from utils.autoanchor import check_anchor_order |
|
from utils.general import make_divisible, check_file, set_logging |
|
from utils.plots import feature_visualization |
|
from utils.torch_utils import time_sync, fuse_conv_and_bn, model_info, scale_img, initialize_weights, \ |
|
select_device, copy_attr |
|
|
|
try: |
|
import thop |
|
except ImportError: |
|
thop = None |
|
|
|
LOGGER = logging.getLogger(__name__) |
|
|
|
|
|
class Detect(nn.Module): |
|
stride = None |
|
onnx_dynamic = False |
|
|
|
def __init__(self, nc=80, anchors=(), ch=(), inplace=True): |
|
super().__init__() |
|
self.nc = nc |
|
self.no = nc + 5 |
|
self.nl = len(anchors) |
|
self.na = len(anchors[0]) // 2 |
|
self.grid = [torch.zeros(1)] * self.nl |
|
a = torch.tensor(anchors).float().view(self.nl, -1, 2) |
|
self.register_buffer('anchors', a) |
|
self.register_buffer('anchor_grid', a.clone().view(self.nl, 1, -1, 1, 1, 2)) |
|
self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch) |
|
self.inplace = inplace |
|
|
|
def forward(self, x): |
|
|
|
z = [] |
|
for i in range(self.nl): |
|
x[i] = self.m[i](x[i]) |
|
bs, _, ny, nx = x[i].shape |
|
x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous() |
|
|
|
if not self.training: |
|
if self.grid[i].shape[2:4] != x[i].shape[2:4] or self.onnx_dynamic: |
|
self.grid[i] = self._make_grid(nx, ny).to(x[i].device) |
|
|
|
y = x[i].sigmoid() |
|
if self.inplace: |
|
y[..., 0:2] = (y[..., 0:2] * 2. - 0.5 + self.grid[i]) * self.stride[i] |
|
y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] |
|
else: |
|
xy = (y[..., 0:2] * 2. - 0.5 + self.grid[i]) * self.stride[i] |
|
wh = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i].view(1, self.na, 1, 1, 2) |
|
y = torch.cat((xy, wh, y[..., 4:]), -1) |
|
z.append(y.view(bs, -1, self.no)) |
|
|
|
return x if self.training else (torch.cat(z, 1), x) |
|
|
|
@staticmethod |
|
def _make_grid(nx=20, ny=20): |
|
yv, xv = torch.meshgrid([torch.arange(ny), torch.arange(nx)]) |
|
return torch.stack((xv, yv), 2).view((1, 1, ny, nx, 2)).float() |
|
|
|
|
|
class Model(nn.Module): |
|
def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, anchors=None): |
|
super().__init__() |
|
if isinstance(cfg, dict): |
|
self.yaml = cfg |
|
else: |
|
import yaml |
|
self.yaml_file = Path(cfg).name |
|
with open(cfg) as f: |
|
self.yaml = yaml.safe_load(f) |
|
|
|
|
|
ch = self.yaml['ch'] = self.yaml.get('ch', ch) |
|
if nc and nc != self.yaml['nc']: |
|
LOGGER.info(f"Overriding model.yaml nc={self.yaml['nc']} with nc={nc}") |
|
self.yaml['nc'] = nc |
|
if anchors: |
|
LOGGER.info(f'Overriding model.yaml anchors with anchors={anchors}') |
|
self.yaml['anchors'] = round(anchors) |
|
self.model, self.save = parse_model(deepcopy(self.yaml), ch=[ch]) |
|
self.names = [str(i) for i in range(self.yaml['nc'])] |
|
self.inplace = self.yaml.get('inplace', True) |
|
|
|
|
|
|
|
m = self.model[-1] |
|
if isinstance(m, Detect): |
|
s = 256 |
|
m.inplace = self.inplace |
|
m.stride = torch.tensor([s / x.shape[-2] for x in self.forward(torch.zeros(1, ch, s, s))]) |
|
m.anchors /= m.stride.view(-1, 1, 1) |
|
check_anchor_order(m) |
|
self.stride = m.stride |
|
self._initialize_biases() |
|
|
|
|
|
|
|
initialize_weights(self) |
|
self.info() |
|
LOGGER.info('') |
|
|
|
def forward(self, x, augment=False, profile=False, visualize=False): |
|
if augment: |
|
return self.forward_augment(x) |
|
return self.forward_once(x, profile, visualize) |
|
|
|
def forward_augment(self, x): |
|
img_size = x.shape[-2:] |
|
s = [1, 0.83, 0.67] |
|
f = [None, 3, None] |
|
y = [] |
|
for si, fi in zip(s, f): |
|
xi = scale_img(x.flip(fi) if fi else x, si, gs=int(self.stride.max())) |
|
yi = self.forward_once(xi)[0] |
|
|
|
yi = self._descale_pred(yi, fi, si, img_size) |
|
y.append(yi) |
|
return torch.cat(y, 1), None |
|
|
|
def forward_once(self, x, profile=False, visualize=False): |
|
y, dt = [], [] |
|
for m in self.model: |
|
if m.f != -1: |
|
x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f] |
|
|
|
if profile: |
|
o = thop.profile(m, inputs=(x,), verbose=False)[0] / 1E9 * 2 if thop else 0 |
|
t = time_sync() |
|
for _ in range(10): |
|
_ = m(x) |
|
dt.append((time_sync() - t) * 100) |
|
if m == self.model[0]: |
|
LOGGER.info(f"{'time (ms)':>10s} {'GFLOPs':>10s} {'params':>10s} {'module'}") |
|
LOGGER.info(f'{dt[-1]:10.2f} {o:10.2f} {m.np:10.0f} {m.type}') |
|
|
|
x = m(x) |
|
y.append(x if m.i in self.save else None) |
|
|
|
if visualize: |
|
feature_visualization(x, m.type, m.i, save_dir=visualize) |
|
|
|
if profile: |
|
LOGGER.info('%.1fms total' % sum(dt)) |
|
return x |
|
|
|
def _descale_pred(self, p, flips, scale, img_size): |
|
|
|
if self.inplace: |
|
p[..., :4] /= scale |
|
if flips == 2: |
|
p[..., 1] = img_size[0] - p[..., 1] |
|
elif flips == 3: |
|
p[..., 0] = img_size[1] - p[..., 0] |
|
else: |
|
x, y, wh = p[..., 0:1] / scale, p[..., 1:2] / scale, p[..., 2:4] / scale |
|
if flips == 2: |
|
y = img_size[0] - y |
|
elif flips == 3: |
|
x = img_size[1] - x |
|
p = torch.cat((x, y, wh, p[..., 4:]), -1) |
|
return p |
|
|
|
def _initialize_biases(self, cf=None): |
|
|
|
|
|
m = self.model[-1] |
|
for mi, s in zip(m.m, m.stride): |
|
b = mi.bias.view(m.na, -1) |
|
b.data[:, 4] += math.log(8 / (640 / s) ** 2) |
|
b.data[:, 5:] += math.log(0.6 / (m.nc - 0.99)) if cf is None else torch.log(cf / cf.sum()) |
|
mi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True) |
|
|
|
def _print_biases(self): |
|
m = self.model[-1] |
|
for mi in m.m: |
|
b = mi.bias.detach().view(m.na, -1).T |
|
LOGGER.info( |
|
('%6g Conv2d.bias:' + '%10.3g' * 6) % (mi.weight.shape[1], *b[:5].mean(1).tolist(), b[5:].mean())) |
|
|
|
|
|
|
|
|
|
|
|
|
|
def fuse(self): |
|
LOGGER.info('Fusing layers... ') |
|
for m in self.model.modules(): |
|
if isinstance(m, (Conv, DWConvClass)) and hasattr(m, 'bn'): |
|
m.conv = fuse_conv_and_bn(m.conv, m.bn) |
|
delattr(m, 'bn') |
|
m.forward = m.forward_fuse |
|
self.info() |
|
return self |
|
|
|
def autoshape(self): |
|
LOGGER.info('Adding AutoShape... ') |
|
m = AutoShape(self) |
|
copy_attr(m, self, include=('yaml', 'nc', 'hyp', 'names', 'stride'), exclude=()) |
|
return m |
|
|
|
def info(self, verbose=False, img_size=640): |
|
model_info(self, verbose, img_size) |
|
|
|
|
|
def parse_model(d, ch): |
|
LOGGER.info('\n%3s%18s%3s%10s %-40s%-30s' % ('', 'from', 'n', 'params', 'module', 'arguments')) |
|
anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple'] |
|
na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors |
|
no = na * (nc + 5) |
|
|
|
layers, save, c2 = [], [], ch[-1] |
|
for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']): |
|
m = eval(m) if isinstance(m, str) else m |
|
for j, a in enumerate(args): |
|
try: |
|
args[j] = eval(a) if isinstance(a, str) else a |
|
except: |
|
pass |
|
|
|
n = max(round(n * gd), 1) if n > 1 else n |
|
if m in [Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, |
|
C3, C3TR, C3SPP]: |
|
c1, c2 = ch[f], args[0] |
|
if c2 != no: |
|
c2 = make_divisible(c2 * gw, 8) |
|
|
|
args = [c1, c2, *args[1:]] |
|
if m in [BottleneckCSP, C3, C3TR]: |
|
args.insert(2, n) |
|
n = 1 |
|
elif m is nn.BatchNorm2d: |
|
args = [ch[f]] |
|
elif m is Concat: |
|
c2 = sum([ch[x] for x in f]) |
|
elif m is Detect: |
|
args.append([ch[x] for x in f]) |
|
if isinstance(args[1], int): |
|
args[1] = [list(range(args[1] * 2))] * len(f) |
|
elif m is Contract: |
|
c2 = ch[f] * args[0] ** 2 |
|
elif m is Expand: |
|
c2 = ch[f] // args[0] ** 2 |
|
else: |
|
c2 = ch[f] |
|
|
|
m_ = nn.Sequential(*[m(*args) for _ in range(n)]) if n > 1 else m(*args) |
|
t = str(m)[8:-2].replace('__main__.', '') |
|
np = sum([x.numel() for x in m_.parameters()]) |
|
m_.i, m_.f, m_.type, m_.np = i, f, t, np |
|
LOGGER.info('%3s%18s%3s%10.0f %-40s%-30s' % (i, f, n, np, t, args)) |
|
save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1) |
|
layers.append(m_) |
|
if i == 0: |
|
ch = [] |
|
ch.append(c2) |
|
return nn.Sequential(*layers), sorted(save) |
|
|
|
|
|
if __name__ == '__main__': |
|
parser = argparse.ArgumentParser() |
|
parser.add_argument('--cfg', type=str, default='yolov5s.yaml', help='model.yaml') |
|
parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') |
|
opt = parser.parse_args() |
|
opt.cfg = check_file(opt.cfg) |
|
set_logging() |
|
device = select_device(opt.device) |
|
|
|
|
|
model = Model(opt.cfg).to(device) |
|
model.train() |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|