yolov5 multi-task 检测人脸landmark
yolov5源码地址：https://github.com/ultralytics/yolov5
文末有完整代码
克隆下最新的yolov5，在根目录下创建landmark文件夹
文件夹中创建三个文件（predict，test，train_landmark），类似yolov5的分类和语义分割任务
没有提及修改的都是直接添加就行
Landmark文件夹
landmark/train_landmark.py
import argparse
import logging
import math
import os
import sys
import random
import time
from pathlib import Path
from threading import Thread
from warnings import warn
import numpy as np
import torch.distributed as dist
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
import torch.optim.lr_scheduler as lr_scheduler
import torch.utils.data
import yaml
from torch.cuda import amp
from torch.nn.parallel import DistributedDataParallel as DDP
from torch.utils.tensorboard import SummaryWriter
from tqdm import tqdm
FILE = Path(__file__).resolve()
ROOT = FILE.parents[1] # YOLOv5 root directory
if str(ROOT) not in sys.path:
sys.path.append(str(ROOT)) # add ROOT to PATH
ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative
import landmark.test as test # import test.py to get mAP after each epoch
from models.experimental import attempt_load
from models.yolo import Model
from utils.callbacks import Callbacks
from utils.autoanchor import check_anchors
# from utils.facedataloaders import create_dataloader
from utils.landmark.dataloaders import create_dataloader
from utils.general import labels_to_class_weights, increment_path, labels_to_image_weights, init_seeds, \
fitness, strip_optimizer, get_latest_run, check_dataset, check_file, check_git_status, check_img_size, \
print_mutation, set_logging, intersect_dicts, check_suffix, check_yaml,colorstr,LOGGER
from utils.downloads import attempt_download,is_url
from utils.landmark.loss import compute_loss
from utils.landmark.plots import plot_labelsf, plot_resultsf, plot_evolution
from utils.plots import plot_images
from utils.torch_utils import ModelEMA, select_device, torch_distributed_zero_first
logger = logging.getLogger(__name__)
try:
import wandb
except ImportError:
wandb = None
# logger.info("Install Weights & Biases for experiment logging via 'pip install wandb' (recommended)")
def train(hyp, opt, device, tb_writer=None, wandb=None):
logger.info(f'Hyperparameters {hyp}')
save_dir, epochs, batch_size, total_batch_size, weights, rank = \
Path(opt.save_dir), opt.epochs, opt.batch_size, opt.total_batch_size, opt.weights, opt.global_rank
# Directories
wdir = save_dir / 'weights'
wdir.mkdir(parents=True, exist_ok=True) # make dir
last = wdir / 'last.pt'
best = wdir / 'best.pt'
results_file = save_dir / 'results.txt'
# Save run settings
with open(save_dir / 'hyp.yaml', 'w') as f:
yaml.dump(hyp, f, sort_keys=False)
with open(save_dir / 'opt.yaml', 'w') as f:
yaml.dump(vars(opt), f, sort_keys=False)
# Configure
plots = not opt.evolve # create plots
cuda = device.type != 'cpu'
init_seeds(2 + rank)
with open(opt.data) as f:
data_dict = yaml.load(f, Loader=yaml.FullLoader) # data dict
with torch_distributed_zero_first(rank):
check_dataset(data_dict) # check
train_path = data_dict['train']
test_path = data_dict['val']
nc = 1 if opt.single_cls else int(data_dict['nc']) # number of classes
names = ['item'] if opt.single_cls and len(data_dict['names']) != 1 else data_dict['names'] # class names
assert len(names) == nc, '%g names found for nc=%g dataset in %s' % (len(names), nc, opt.data) # check
# Model
check_suffix(weights, '.pt') # check weights
pretrained = weights.endswith('.pt')
if pretrained:
with torch_distributed_zero_first(rank):
attempt_download(weights) # download if not found locally
ckpt = torch.load(weights, map_location=device) # load checkpoint
if hyp.get('anchors'):
ckpt['model'].yaml['anchors'] = round(hyp['anchors']) # force autoanchor
model = Model(opt.cfg or ckpt['model'].yaml, ch=3, nc=nc).to(device) # create
exclude = ['anchor'] if opt.cfg or hyp.get('anchors') else [] # exclude keys
state_dict = ckpt['model'].float().state_dict() # to FP32
state_dict = intersect_dicts(state_dict, model.state_dict(), exclude=exclude) # intersect
model.load_state_dict(state_dict, strict=False) # load
logger.info('Transferred %g/%g items from %s' % (len(state_dict), len(model.state_dict()), weights)) # report
else:
model = Model(opt.cfg, ch=3, nc=nc).to(device) # create
# Freeze
freeze = [] # parameter names to freeze (full or partial)
for k, v in model.named_parameters():
v.requires_grad = True # train all layers
if any(x in k for x in freeze):
print('freezing %s' % k)
v.requires_grad = False
# Optimizer
nbs = 64 # nominal batch size
accumulate = max(round(nbs / total_batch_size), 1) # accumulate loss before optimizing
hyp['weight_decay'] *= total_batch_size * accumulate / nbs # scale weight_decay
pg0, pg1, pg2 = [], [], [] # optimizer parameter groups
for k, v in model.named_modules():
if hasattr(v, 'bias') and isinstance(v.bias, nn.Parameter):
pg2.append(v.bias) # biases
if isinstance(v, nn.BatchNorm2d):
pg0.append(v.weight) # no decay
elif hasattr(v, 'weight') and isinstance(v.weight, nn.Parameter):
pg1.append(v.weight) # apply decay
if opt.adam:
optimizer = optim.Adam(pg0, lr=hyp['lr0'], betas=(hyp['momentum'], 0.999)) # adjust beta1 to momentum
else:
optimizer = optim.SGD(pg0, lr=hyp['lr0'], momentum=hyp['momentum'], nesterov=True)
optimizer.add_param_group({'params': pg1, 'weight_decay': hyp['weight_decay']}) # add pg1 with weight_decay
optimizer.add_param_group({'params': pg2}) # add pg2 (biases)
logger.info('Optimizer groups: %g .bias, %g conv.weight, %g other' % (len(pg2), len(pg1), len(pg0)))
del pg0, pg1, pg2
# Scheduler https://arxiv.org/pdf/1812.01187.pdf
# https://pytorch.org/docs/stable/_modules/torch/optim/lr_scheduler.html#OneCycleLR
lf = lambda x: ((1 + math.cos(x * math.pi / epochs)) / 2) * (1 - hyp['lrf']) + hyp['lrf'] # cosine
scheduler = lr_scheduler.LambdaLR(optimizer, lr_lambda=lf)
# plot_lr_scheduler(optimizer, scheduler, epochs)
# Logging
if wandb and wandb.run is None:
opt.hyp = hyp # add hyperparameters
wandb_run = wandb.init(config=opt, resume="allow",
project='YOLOv5' if opt.project == 'runs/train' else Path(opt.project).stem,
name=save_dir.stem,
id=ckpt.get('wandb_id') if 'ckpt' in locals() else None)
loggers = {'wandb': wandb} # loggers dict
# Resume
start_epoch, best_fitness = 0, 0.0
if pretrained:
# Optimizer
if ckpt['optimizer'] is not None:
optimizer.load_state_dict(ckpt['optimizer'])
best_fitness = ckpt['best_fitness']
# Results
if ckpt.get('training_results') is not None:
with open(results_file, 'w') as file:
file.write(ckpt['training_results']) # write results.txt
# Epochs
# start_epoch = ckpt['epoch'] + 1
if opt.resume:
assert start_epoch > 0, '%s training to %g epochs is finished, nothing to resume.' % (weights, epochs)
if epochs < start_epoch:
logger.info('%s has been trained for %g epochs. Fine-tuning for %g additional epochs.' %
(weights, ckpt['epoch'], epochs))
epochs += ckpt['epoch'] # finetune additional epochs
del ckpt, state_dict
# Image sizes
gs = int(max(model.stride)) # grid size (max stride)
imgsz, imgsz_test = [check_img_size(x, gs) for x in opt.img_size] # verify imgsz are gs-multiples
# DP mode
if cuda and rank == -1 and torch.cuda.device_count() > 1:
model = torch.nn.DataParallel(model)
# SyncBatchNorm
if opt.sync_bn and cuda and rank != -1:
model = torch.nn.SyncBatchNorm.convert_sync_batchnorm(model).to(device)
logger.info('Using SyncBatchNorm()')
# EMA
ema = ModelEMA(model) if rank in [-1, 0] else None
# DDP mode
if cuda and rank != -1:
model = DDP(model, device_ids=[opt.local_rank], output_device=opt.local_rank)
# Trainloader
dataloader, dataset = create_dataloader(train_path, imgsz, batch_size, gs, opt,
hyp=hyp, augment=True, cache=opt.cache_images, rect=opt.rect, rank=rank,
world_size=opt.world_size, workers=opt.workers,
image_weights=opt.image_weights,prefix=colorstr('train: '))
mlc = np.concatenate(dataset.labels, 0)[:, 0].max() # max label class
nb = len(dataloader) # number of batches
assert mlc < nc, 'Label class %g exceeds nc=%g in %s. Possible class labels are 0-%g' % (mlc, nc, opt.data, nc - 1)
# Process 0
if rank in [-1, 0]:
ema.updates = start_epoch * nb // accumulate # set EMA updates
testloader = create_dataloader(test_path, imgsz_test, total_batch_size, gs, opt, # testloader
hyp=hyp, cache=opt.cache_images and not opt.notest, rect=True,
rank=-1, world_size=opt.world_size, workers=opt.workers, pad=0.5,prefix=colorstr('val: '))[0]
if not opt.resume:
labels = np.concatenate(dataset.labels, 0)
c = torch.tensor(labels[:, 0]) # classes
# cf = torch.bincount(c.long(), minlength=nc) + 1. # frequency
# model._initialize_biases(cf.to(device))
if plots:
plot_labelsf(labels, save_dir, loggers)
if tb_writer:
tb_writer.add_histogram('classes', c, 0)
# Anchors
if not opt.noautoanchor:
check_anchors(dataset, model=model, thr=hyp['anchor_t'], imgsz=imgsz)
# Model parameters
hyp['cls'] *= nc / 80. # scale coco-tuned hyp['cls'] to current dataset
model.nc = nc # attach number of classes to model
model.hyp = hyp # attach hyperparameters to model
model.gr = 1.0 # iou loss ratio (obj_loss = 1.0 or iou)
model.class_weights = labels_to_class_weights(dataset.labels, nc).to(device) * nc # attach class weights
model.names = names
# Start training
t0 = time.time()
nw = max(round(hyp['warmup_epochs'] * nb), 1000) # number of warmup iterations, max(3 epochs, 1k iterations)
# nw = min(nw, (epochs - start_epoch) / 2 * nb) # limit warmup to < 1/2 of training
maps = np.zeros(nc) # mAP per class
results = (0, 0, 0, 0, 0, 0, 0) # P, R, [email protected], [email protected], val_loss(box, obj, cls)
scheduler.last_epoch = start_epoch - 1 # do not move
scaler = amp.GradScaler(enabled=cuda)
logger.info('Image sizes %g train, %g test\n'
'Using %g dataloader workers\nLogging results to %s\n'
'Starting training for %g epochs...' % (imgsz, imgsz_test, dataloader.num_workers, save_dir, epochs))
for epoch in range(start_epoch, epochs): # epoch ------------------------------------------------------------------
model.train()
# Update image weights (optional)
if opt.image_weights:
# Generate indices
if rank in [-1, 0]:
cw = model.class_weights.cpu().numpy() * (1 - maps) ** 2 / nc # class weights
iw = labels_to_image_weights(dataset.labels, nc=nc, class_weights=cw) # image weights
dataset.indices = random.choices(range(dataset.n), weights=iw, k=dataset.n) # rand weighted idx
# Broadcast if DDP
if rank != -1:
indices = (torch.tensor(dataset.indices) if rank == 0 else torch.zeros(dataset.n)).int()
dist.broadcast(indices, 0)
if rank != 0:
dataset.indices = indices.cpu().numpy()
# Update mosaic border
# b = int(random.uniform(0.25 * imgsz, 0.75 * imgsz + gs) // gs * gs)
# dataset.mosaic_border = [b - imgsz, -b] # height, width borders
mloss = torch.zeros(5, device=device) # mean losses
if rank != -1:
dataloader.sampler.set_epoch(epoch)
pbar = enumerate(dataloader)
logger.info(
('\n' + '%10s' * 9) % ('Epoch', 'gpu_mem', 'box', 'obj', 'cls', 'landmark', 'total', 'targets', 'img_size'))
if rank in [-1, 0]:
pbar = tqdm(pbar, total=nb) # progress bar
optimizer.zero_grad()
for i, (imgs, targets, paths, _) in pbar: # batch -------------------------------------------------------------
ni = i + nb * epoch # number integrated batches (since train start)
imgs = imgs.to(device, non_blocking=True).float() / 255.0 # uint8 to float32, 0-255 to 0.0-1.0
# Warmup
if ni <= nw:
xi = [0, nw] # x interp
# model.gr = np.interp(ni, xi, [0.0, 1.0]) # iou loss ratio (obj_loss = 1.0 or iou)
accumulate = max(1, np.interp(ni, xi, [1, nbs / total_batch_size]).round())
for j, x in enumerate(optimizer.param_groups):
# bias lr falls from 0.1 to lr0, all other lrs rise from 0.0 to lr0
x['lr'] = np.interp(ni, xi, [hyp['warmup_bias_lr'] if j == 2 else 0.0, x['initial_lr'] * lf(epoch)])
if 'momentum' in x:
x['momentum'] = np.interp(ni, xi, [hyp['warmup_momentum'], hyp['momentum']])
# Multi-scale
if opt.multi_scale:
sz = random.randrange(imgsz * 0.5, imgsz * 1.5 + gs) // gs * gs # size
sf = sz / max(imgs.shape[2:]) # scale factor
if sf != 1:
ns = [math.ceil(x * sf / gs) * gs for x in imgs.shape[2:]] # new shape (stretched to gs-multiple)
imgs = F.interpolate(imgs, size=ns, mode='bilinear', align_corners=False)
# Forward
with amp.autocast(enabled=cuda):
pred = model(imgs) # forward
loss, loss_items = compute_loss(pred, targets.to(device), model) # loss scaled by batch_size
if rank != -1:
loss *= opt.world_size # gradient averaged between devices in DDP mode
# Backward
scaler.scale(loss).backward()
# Optimize
if ni % accumulate == 0:
scaler.step(optimizer) # optimizer.step
scaler.update()
optimizer.zero_grad()
if ema:
ema.update(model)
# Print
if rank in [-1, 0]:
mloss = (mloss * i + loss_items) / (i + 1) # update mean losses
mem = '%.3gG' % (torch.cuda.memory_reserved() / 1E9 if torch.cuda.is_available() else 0) # (GB)
s = ('%10s' * 2 + '%10.4g' * 7) % (
'%g/%g' % (epoch, epochs - 1), mem, *mloss, targets.shape[0], imgs.shape[-1])
pbar.set_description(s)
# Plot
if plots and ni < 3:
f = save_dir / f'train_batch{ni}.jpg' # filename
Thread(target=plot_images, args=(imgs, targets, paths, f), daemon=True).start()
# if tb_writer:
# tb_writer.add_image(f, result, dataformats='HWC', global_step=epoch)
# tb_writer.add_graph(model, imgs) # add model to tensorboard
elif plots and ni == 3 and wandb:
wandb.log({"Mosaics": [wandb.Image(str(x), caption=x.name) for x in save_dir.glob('train*.jpg')]})
# end batch ------------------------------------------------------------------------------------------------
# end epoch ----------------------------------------------------------------------------------------------------
# Scheduler
lr = [x['lr'] for x in optimizer.param_groups] # for tensorboard
scheduler.step()
# DDP process 0 or single-GPU
if rank in [-1, 0] and epoch >= 0:
# mAP
if ema:
ema.update_attr(model, include=['yaml', 'nc', 'hyp', 'gr', 'names', 'stride', 'class_weights'])
final_epoch = epoch + 1 == epochs
if not opt.notest or final_epoch: # Calculate mAP
results, maps, times = test.test(opt.data,
batch_size=total_batch_size,
imgsz=imgsz_test,
model=ema.ema,
single_cls=opt.single_cls,
dataloader=testloader,
save_dir=save_dir,
plots=False,
log_imgs=opt.log_imgs if wandb else 0)
# Write
with open(results_file, 'a') as f:
f.write(s + '%10.4g' * 7 % results + '\n') # P, R, [email protected], [email protected], val_loss(box, obj, cls)
if len(opt.name) and opt.bucket:
os.system('gsutil cp %s gs://%s/results/results%s.txt' % (results_file, opt.bucket, opt.name))
# Log
tags = ['train/box_loss', 'train/obj_loss', 'train/cls_loss', # train loss
'metrics/precision', 'metrics/recall', 'metrics/mAP_0.5', 'metrics/mAP_0.5:0.95',
'val/box_loss', 'val/obj_loss', 'val/cls_loss', # val loss
'x/lr0', 'x/lr1', 'x/lr2'] # params
for x, tag in zip(list(mloss[:-1]) + list(results) + lr, tags):
if tb_writer:
tb_writer.add_scalar(tag, x, epoch) # tensorboard
if wandb:
wandb.log({tag: x}) # W&B
# Update best mAP
fi = fitness(np.array(results).reshape(1, -1)) # weighted combination of [P, R, [email protected], [email protected]]
if fi > best_fitness:
best_fitness = fi
# Save model
save = (not opt.nosave) or (final_epoch and not opt.evolve)
if save:
with open(results_file, 'r') as f: # create checkpoint
ckpt = {'epoch': epoch,
'best_fitness': best_fitness,
'training_results': f.read(),
'model': ema.ema,
'optimizer': None if final_epoch else optimizer.state_dict(),
'wandb_id': wandb_run.id if wandb else None}
# Save last, best and delete
torch.save(ckpt, last)
if best_fitness == fi:
torch.save(ckpt, best)
del ckpt
# end epoch ----------------------------------------------------------------------------------------------------
# end training
if rank in [-1, 0]:
# Strip optimizers
final = best if best.exists() else last # final model
for f in [last, best]:
if f.exists():
strip_optimizer(f) # strip optimizers
if opt.bucket:
os.system(f'gsutil cp {final} gs://{opt.bucket}/weights') # upload
# Plots
if plots:
plot_resultsf(save_dir=save_dir) # save as results.png
if wandb:
files = ['results.png', 'precision_recall_curve.png', 'confusion_matrix.png']
wandb.log({"Results": [wandb.Image(str(save_dir / f), caption=f) for f in files
if (save_dir / f).exists()]})
if opt.log_artifacts:
wandb.log_artifact(artifact_or_path=str(final), type='model', name=save_dir.stem)
# Test best.pt
logger.info('%g epochs completed in %.3f hours.\n' % (epoch - start_epoch + 1, (time.time() - t0) / 3600))
for f in last, best:
if f.exists():
strip_optimizer(f) # strip optimizers
if f is best:
LOGGER.info(f'\nValidating {f}...')
results, _, _ = test.test(
opt.data,
batch_size=total_batch_size,
imgsz=imgsz_test,
conf_thres=0.001,
iou_thres=0.65,
model=attempt_load(final, device),
single_cls=opt.single_cls,
dataloader=testloader,
save_dir=save_dir,
save_json=False,
plots=True) # val best model with plots
else:
dist.destroy_process_group()
wandb.run.finish() if wandb and wandb.run else None
torch.cuda.empty_cache()
return results
def parse_opt(known=False):
parser = argparse.ArgumentParser()
parser.add_argument('--weights', type=str, default=ROOT / 'yolov5n-lmk.pt', help='initial weights path')
parser.add_argument('--cfg', type=str, default=ROOT / 'models/landmark/yolov5n-landmark.yaml',help='model.yaml path')
parser.add_argument('--data', type=str, default=ROOT / 'data/widerface.yaml', help='data.yaml path')
parser.add_argument('--hyp', type=str, default=ROOT / 'data/hyps/hyp.scratch.yaml', help='hyperparameters path')
parser.add_argument('--epochs', type=int, default=300)
parser.add_argument('--batch-size', type=int, default=16, help='total batch size for all GPUs')
parser.add_argument('--img-size', nargs='+', type=int, default=[800, 800], help='[train, test] image sizes')
parser.add_argument('--rect', action='store_true', help='rectangular training')
parser.add_argument('--resume', nargs='?', const=True, default=False, help='resume most recent training')
parser.add_argument('--nosave', action='store_true', help='only save final checkpoint')
parser.add_argument('--notest', action='store_true', help='only test final epoch')
parser.add_argument('--noautoanchor', action='store_true', help='disable autoanchor check')
parser.add_argument('--evolve', type=int, nargs='?', const=300, help='evolve hyperparameters for x generations')
parser.add_argument('--bucket', type=str, default='', help='gsutil bucket')
parser.add_argument('--cache-images', action='store_true', help='cache images for faster training')
parser.add_argument('--image-weights', action='store_true', help='use weighted image selection for training')
parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
parser.add_argument('--multi-scale', action='store_true', default=False, help='vary img-size +/- 50%%')
parser.add_argument('--single-cls', action='store_true', help='train multi-class data as single-class')
parser.add_argument('--adam', action='store_true', help='use torch.optim.Adam() optimizer')
parser.add_argument('--sync-bn', action='store_true', help='use SyncBatchNorm, only available in DDP mode')
parser.add_argument('--local_rank', type=int, default=-1, help='DDP parameter, do not modify')
parser.add_argument('--log-imgs', type=int, default=16, help='number of images for W&B logging, max 100')
parser.add_argument('--log-artifacts', action='store_true', help='log artifacts, i.e. final trained model')
parser.add_argument('--workers', type=int, default=4, help='maximum number of dataloader workers')
parser.add_argument('--project', default=ROOT / 'runs/train-lmk', help='save to project/name')
parser.add_argument('--name', default='exp', help='save to project/name')
parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment')
return parser.parse_known_args()[0] if known else parser.parse_args()
def main(opt, callbacks=Callbacks()):
# Set DDP variables
opt.total_batch_size = opt.batch_size
opt.world_size = int(os.environ['WORLD_SIZE']) if 'WORLD_SIZE' in os.environ else 1
opt.global_rank = int(os.environ['RANK']) if 'RANK' in os.environ else -1
set_logging(str(opt.global_rank))
if opt.global_rank in [-1, 0]:
check_git_status()
# Resume
if opt.resume and not opt.evolve: # resume from specified or most recent last.pt
last = Path(check_file(opt.resume) if isinstance(opt.resume, str) else get_latest_run())
opt_yaml = last.parent.parent / 'opt.yaml' # train options yaml
opt_data = opt.data # original dataset
if opt_yaml.is_file():
with open(opt_yaml, errors='ignore') as f:
d = yaml.safe_load(f)
else:
d = torch.load(last, map_location='cpu')['opt']
opt = argparse.Namespace(**d) # replace
opt.cfg, opt.weights, opt.resume = '', str(last), True # reinstate
if is_url(opt_data):
opt.data = check_file(opt_data) # avoid HUB resume auth timeout
else:
opt.data, opt.cfg, opt.hyp, opt.weights, opt.project = \
check_file(opt.data), check_yaml(opt.cfg), check_yaml(opt.hyp), str(opt.weights), str(opt.project) # checks
assert len(opt.cfg) or len(opt.weights), 'either --cfg or --weights must be specified'
if opt.evolve:
if opt.project == str(ROOT / 'runs/train'): # if default project name, rename to runs/evolve
opt.project = str(ROOT / 'runs/evolve')
opt.exist_ok, opt.resume = opt.resume, False # pass resume to exist_ok and disable resume
if opt.name == 'cfg':
opt.name = Path(opt.cfg).stem # use model.yaml as name
opt.save_dir = str(increment_path(Path(opt.project) / opt.name, exist_ok=opt.exist_ok))
# DDP mode
device = select_device(opt.device, batch_size=opt.batch_size)
if opt.local_rank != -1:
assert torch.cuda.device_count() > opt.local_rank
torch.cuda.set_device(opt.local_rank)
device = torch.device('cuda', opt.local_rank)
dist.init_process_group(backend='nccl', init_method='env://') # distributed backend
assert opt.batch_size % opt.world_size == 0, '--batch-size must be multiple of CUDA device count'
opt.batch_size = opt.total_batch_size // opt.world_size
# Hyperparameters
with open(opt.hyp) as f:
hyp = yaml.load(f, Loader=yaml.FullLoader) # load hyps
if 'box' not in hyp:
warn('Compatibility: %s missing "box" which was renamed from "giou" in %s' %
(opt.hyp, 'https://github.com/ultralytics/yolov5/pull/1120'))
hyp['box'] = hyp.pop('giou')
# Train
logger.info(opt)
if not opt.evolve:
tb_writer = None # init loggers
if opt.global_rank in [-1, 0]:
logger.info(f'Start Tensorboard with "tensorboard --logdir {opt.project}", view at http://localhost:6006/')
tb_writer = SummaryWriter(opt.save_dir) # Tensorboard
train(hyp, opt, device, tb_writer, wandb)
# Evolve hyperparameters (optional)
else:
# Hyperparameter evolution metadata (mutation scale 0-1, lower_limit, upper_limit)
meta = {'lr0': (1, 1e-5, 1e-1), # initial learning rate (SGD=1E-2, Adam=1E-3)
'lrf': (1, 0.01, 1.0), # final OneCycleLR learning rate (lr0 * lrf)
'momentum': (0.3, 0.6, 0.98), # SGD momentum/Adam beta1
'weight_decay': (1, 0.0, 0.001), # optimizer weight decay
'warmup_epochs': (1, 0.0, 5.0), # warmup epochs (fractions ok)
'warmup_momentum': (1, 0.0, 0.95), # warmup initial momentum
'warmup_bias_lr': (1, 0.0, 0.2), # warmup initial bias lr
'box': (1, 0.02, 0.2), # box loss gain
'cls': (1, 0.2, 4.0), # cls loss gain
'cls_pw': (1, 0.5, 2.0), # cls BCELoss positive_weight
'obj': (1, 0.2, 4.0), # obj loss gain (scale with pixels)
'obj_pw': (1, 0.5, 2.0), # obj BCELoss positive_weight
'iou_t': (0, 0.1, 0.7), # IoU training threshold
'anchor_t': (1, 2.0, 8.0), # anchor-multiple threshold
'anchors': (2, 2.0, 10.0), # anchors per output grid (0 to ignore)
'fl_gamma': (0, 0.0, 2.0), # focal loss gamma (efficientDet default gamma=1.5)
'hsv_h': (1, 0.0, 0.1), # image HSV-Hue augmentation (fraction)
'hsv_s': (1, 0.0, 0.9), # image HSV-Saturation augmentation (fraction)
'hsv_v': (1, 0.0, 0.9), # image HSV-Value augmentation (fraction)
'degrees': (1, 0.0, 45.0), # image rotation (+/- deg)
'translate': (1, 0.0, 0.9), # image translation (+/- fraction)
'scale': (1, 0.0, 0.9), # image scale (+/- gain)
'shear': (1, 0.0, 10.0), # image shear (+/- deg)
'perspective': (0, 0.0, 0.001), # image perspective (+/- fraction), range 0-0.001
'flipud': (1, 0.0, 1.0), # image flip up-down (probability)
'fliplr': (0, 0.0, 1.0), # image flip left-right (probability)
'mosaic': (1, 0.0, 1.0), # image mixup (probability)
'mixup': (1, 0.0, 1.0)} # image mixup (probability)
assert opt.local_rank == -1, 'DDP mode not implemented for --evolve'
opt.notest, opt.nosave = True, True # only test/save final epoch
# ei = [isinstance(x, (int, float)) for x in hyp.values()] # evolvable indices
yaml_file = Path(opt.save_dir) / 'hyp_evolved.yaml' # save best result here
if opt.bucket:
os.system('gsutil cp gs://%s/evolve.txt .' % opt.bucket) # download evolve.txt if exists
for _ in range(opt.evolve): # generations to evolve
if Path('evolve.txt').exists(): # if evolve.txt exists: select best hyps and mutate
# Select parent(s)
parent = 'single' # parent selection method: 'single' or 'weighted'
x = np.loadtxt('evolve.txt', ndmin=2)
n = min(5, len(x)) # number of previous results to consider
x = x[np.argsort(-fitness(x))][:n] # top n mutations
w = fitness(x) - fitness(x).min() + 1E-6 # weights
if parent == 'single' or len(x) == 1:
# x = x[random.randint(0, n - 1)] # random selection
x = x[random.choices(range(n), weights=w)[0]] # weighted selection
elif parent == 'weighted':
x = (x * w.reshape(n, 1)).sum(0) / w.sum() # weighted combination
# Mutate
mp, s = 0.8, 0.2 # mutation probability, sigma
npr = np.random
npr.seed(int(time.time()))
g = np.array([x[0] for x in meta.values()]) # gains 0-1
ng = len(meta)
v = np.ones(ng)
while all(v == 1): # mutate until a change occurs (prevent duplicates)
v = (g * (npr.random(ng) < mp) * npr.randn(ng) * npr.random() * s + 1).clip(0.3, 3.0)
for i, k in enumerate(hyp.keys()): # plt.hist(v.ravel(), 300)
hyp[k] = float(x[i + 7] * v[i]) # mutate
# Constrain to limits
for k, v in meta.items():
hyp[k] = max(hyp[k], v[1]) # lower limit
hyp[k] = min(hyp[k], v[2]) # upper limit
hyp[k] = round(hyp[k], 5) # significant digits
# Train mutation
results = train(hyp.copy(), opt, device, wandb=wandb)
# Write mutation results
# keys = ('metrics/precision', 'metrics/recall', 'metrics/mAP_0.5', 'metrics/mAP_0.5:0.95', 'val/box_loss',
# 'val/obj_loss', 'val/cls_loss')
print_mutation(hyp.copy(), results, yaml_file, opt.bucket)
# print_mutation(keys, results, hyp.copy(), save_dir, opt.bucket)
# Plot results
plot_evolution(yaml_file)
print(f'Hyperparameter evolution complete. Best results saved as: {yaml_file}\n'
f'Command to train a new model with these hyperparameters: $ python train.py --hyp {yaml_file}')
def run(**kwargs):
# Usage: import train; train.run(data='coco128.yaml', imgsz=320, weights='yolov5m.pt')
opt = parse_opt(True)
for k, v in kwargs.items():
setattr(opt, k, v)
main(opt)
return opt
if __name__ == "__main__":
opt = parse_opt()
main(opt)
landmark/test.py
import argparse
import json
import os
import sys
from pathlib import Path
from threading import Thread
import numpy as np
import torch
import yaml
from tqdm import tqdm
FILE = Path(__file__).resolve()
ROOT = FILE.parents[1] # YOLOv5 root directory
if str(ROOT) not in sys.path:
sys.path.append(str(ROOT)) # add ROOT to PATH
ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative
from models.experimental import attempt_load
from utils.landmark.dataloaders import create_dataloader
from utils.general import coco80_to_coco91_class, check_dataset, check_file, check_img_size, box_iou, \
scale_boxes, xyxy2xywh, xywh2xyxy, set_logging, increment_path
from utils.landmark.general import non_max_suppression_face
from utils.landmark.loss import compute_loss
from utils.metrics import ConfusionMatrix
from utils.landmark.metrics import ap_per_classf
from utils.landmark.plots import plot_imagesf, output_to_targetf, plot_study_txt
from utils.torch_utils import select_device, time_sync
def test(data,
weights=None,
batch_size=32,
imgsz=640,
conf_thres=0.001,
iou_thres=0.6, # for NMS
save_json=False,
single_cls=False,
augment=False,
verbose=False,
model=None,
dataloader=None,
save_dir=Path(''), # for saving images
save_txt=False, # for auto-labelling
save_hybrid=False, # for hybrid auto-labelling
save_conf=False, # save auto-label confidences
plots=True,
log_imgs=0): # number of logged images
# Initialize/load model and set device
training = model is not None
if training: # called by train.py
device = next(model.parameters()).device # get model device
else: # called directly
set_logging()
device = select_device(opt.device, batch_size=batch_size)
# Directories
save_dir = Path(increment_path(Path(opt.project) / opt.name, exist_ok=opt.exist_ok)) # increment run
(save_dir / 'labels' if save_txt else save_dir).mkdir(parents=True, exist_ok=True) # make dir
# Load model
model = attempt_load(weights) # load FP32 model
imgsz = check_img_size(imgsz, s=model.stride.max()) # check img_size
# Multi-GPU disabled, incompatible with .half() https://github.com/ultralytics/yolov5/issues/99
# if device.type != 'cpu' and torch.cuda.device_count() > 1:
# model = nn.DataParallel(model)
# Half
half = device.type != 'cpu' # half precision only supported on CUDA
if half:
model.half()
# Configure
model.eval()
is_coco = data.endswith('coco.yaml') # is COCO dataset
with open(data) as f:
data = yaml.load(f, Loader=yaml.FullLoader) # model dict
check_dataset(data) # check
nc = 1 if single_cls else int(data['nc']) # number of classes
iouv = torch.linspace(0.5, 0.95, 10).to(device) # iou vector for [email protected]:0.95
niou = iouv.numel()
# Logging
log_imgs, wandb = min(log_imgs, 100), None # ceil
try:
import wandb # Weights & Biases
except ImportError:
log_imgs = 0
# Dataloader
if not training:
img = torch.zeros((1, 3, imgsz, imgsz), device=device) # init img
_ = model(img.half() if half else img) if device.type != 'cpu' else None # run once
path = data['test'] if opt.task == 'test' else data['val'] # path to val/test images
dataloader = create_dataloader(path, imgsz, batch_size, model.stride.max(), opt, pad=0.5, rect=True)[0]
seen = 0
confusion_matrix = ConfusionMatrix(nc=nc)
names = {k: v for k, v in enumerate(model.names if hasattr(model, 'names') else model.module.names)}
coco91class = coco80_to_coco91_class()
s = ('%20s' + '%12s' * 6) % ('Class', 'Images', 'Targets', 'P', 'R', '[email protected]', '[email protected]:.95')
p, r, f1, mp, mr, map50, map, t0, t1 = 0., 0., 0., 0., 0., 0., 0., 0., 0.
loss = torch.zeros(3, device=device)
jdict, stats, ap, ap_class, wandb_images = [], [], [], [], []
for batch_i, (img, targets, paths, shapes) in enumerate(tqdm(dataloader, desc=s)):
img = img.to(device, non_blocking=True)
img = img.half() if half else img.float() # uint8 to fp16/32
img /= 255.0 # 0 - 255 to 0.0 - 1.0
targets = targets.to(device)
nb, _, height, width = img.shape # batch size, channels, height, width
with torch.no_grad():
# Run model
t = time_sync()
inf_out, train_out = model(img, augment=augment) # inference and training outputs
t0 += time_sync() - t
# Compute loss
if training:
loss += compute_loss([x.float() for x in train_out], targets, model)[1][:3] # box, obj, cls
# Run NMS
targets[:, 2:6] *= torch.Tensor([width, height, width, height]).to(device) # to pixels
lb = [targets[targets[:, 0] == i, 1:] for i in range(nb)] if save_hybrid else [] # for autolabelling
t = time_sync()
#output = non_max_suppression(inf_out, conf_thres=conf_thres, iou_thres=iou_thres, labels=lb)
output = non_max_suppression_face(inf_out, conf_thres=conf_thres, iou_thres=iou_thres, labels=lb)
t1 += time_sync() - t
# Statistics per image
for si, pred in enumerate(output):
pred = torch.cat((pred[:, :5], pred[:, 15:]), 1) # throw landmark in thresh
labels = targets[targets[:, 0] == si, 1:]
nl = len(labels)
tcls = labels[:, 0].tolist() if nl else [] # target class
path = Path(paths[si])
seen += 1
if len(pred) == 0:
if nl:
stats.append((torch.zeros(0, niou, dtype=torch.bool), torch.Tensor(), torch.Tensor(), tcls))
if plots:
confusion_matrix.process_batch(detections=None, labels=labels[:, 0])
continue
# Predictions
predn = pred.clone()
scale_boxes(img[si].shape[1:], predn[:, :4], shapes[si][0], shapes[si][1]) # native-space pred
# Append to text file
if save_txt:
gn = torch.tensor(shapes[si][0])[[1, 0, 1, 0]] # normalization gain whwh
for *xyxy, conf, cls in predn.tolist():
xywh = (xyxy2xywh(torch.tensor(xyxy).view(1, 4)) / gn).view(-1).tolist() # normalized xywh
line = (cls, *xywh, conf) if save_conf else (cls, *xywh) # label format
with open(save_dir / 'labels' / (path.stem + '.txt'), 'a') as f:
f.write(('%g ' * len(line)).rstrip() % line + '\n')
# W&B logging
if plots and len(wandb_images) < log_imgs:
box_data = [{"position": {"minX": xyxy[0], "minY": xyxy[1], "maxX": xyxy[2], "maxY": xyxy[3]},
"class_id": int(cls),
"box_caption": "%s %.3f" % (names[cls], conf),
"scores": {"class_score": conf},
"domain": "pixel"} for *xyxy, conf, cls in pred.tolist()]
boxes = {"predictions": {"box_data": box_data, "class_labels": names}} # inference-space
wandb_images.append(wandb.Image(img[si], boxes=boxes, caption=path.name))
# Append to pycocotools JSON dictionary
if save_json:
# [{"image_id": 42, "category_id": 18, "bbox": [258.15, 41.29, 348.26, 243.78], "score": 0.236}, ...
image_id = int(path.stem) if path.stem.isnumeric() else path.stem
box = xyxy2xywh(predn[:, :4]) # xywh
box[:, :2] -= box[:, 2:] / 2 # xy center to top-left corner
for p, b in zip(pred.tolist(), box.tolist()):
jdict.append({'image_id': image_id,
'category_id': coco91class[int(p[15])] if is_coco else int(p[15]),
'bbox': [round(x, 3) for x in b],
'score': round(p[4], 5)})
# Assign all predictions as incorrect
correct = torch.zeros(pred.shape[0], niou, dtype=torch.bool, device=device)
if nl:
detected = [] # target indices
tcls_tensor = labels[:, 0]
# target boxes
tbox = xywh2xyxy(labels[:, 1:5])
scale_boxes(img[si].shape[1:], tbox, shapes[si][0], shapes[si][1]) # native-space labels
if plots:
confusion_matrix.process_batch(pred, torch.cat((labels[:, 0:1], tbox), 1))
# Per target class
for cls in torch.unique(tcls_tensor):
ti = (cls == tcls_tensor).nonzero(as_tuple=False).view(-1) # prediction indices
pi = (cls == pred[:, 5]).nonzero(as_tuple=False).view(-1) # target indices
# Search for detections
if pi.shape[0]:
# Prediction to target ious
ious, i = box_iou(predn[pi, :4], tbox[ti]).max(1) # best ious, indices
# Append detections
detected_set = set()
for j in (ious > iouv[0]).nonzero(as_tuple=False):
d = ti[i[j]] # detected target
if d.item() not in detected_set:
detected_set.add(d.item())
detected.append(d)
correct[pi[j]] = ious[j] > iouv # iou_thres is 1xn
if len(detected) == nl: # all targets already located in image
break
# Append statistics (correct, conf, pcls, tcls)
stats.append((correct.cpu(), pred[:, 4].cpu(), pred[:, 5].cpu(), tcls))
# Plot images
if plots and batch_i < 3:
plot_imagesf(img, targets, paths, save_dir / f'val_batch{batch_i}_labels.jpg', names) # labels
plot_imagesf(img, output_to_targetf(output), paths, save_dir / f'val_batch{batch_i}_pred.jpg', names) # pred
# Compute statistics
stats = [np.concatenate(x, 0) for x in zip(*stats)] # to numpy
if len(stats) and stats[0].any():
p, r, ap, f1, ap_class = ap_per_classf(*stats, plot=plots, save_dir=save_dir, names=names)
p, r, ap50, ap = p[:, 0], r[:, 0], ap[:, 0], ap.mean(1) # [P, R, [email protected], [email protected]:0.95]
mp, mr, map50, map = p.mean(), r.mean(), ap50.mean(), ap.mean()
nt = np.bincount(stats[3].astype(np.int64), minlength=nc) # number of targets per class
else:
nt = torch.zeros(1)
# Print results
pf = '%20s' + '%12.3g' * 6 # print format
print(pf % ('all', seen, nt.sum(), mp, mr, map50, map))
# Print results per class
if verbose and nc > 1 and len(stats):
for i, c in enumerate(ap_class):
print(pf % (names[c], seen, nt[c], p[i], r[i], ap50[i], ap[i]))
# Print speeds
t = tuple(x / seen * 1E3 for x in (t0, t1, t0 + t1)) + (imgsz, imgsz, batch_size) # tuple
if not training:
print('Speed: %.1f/%.1f/%.1f ms inference/NMS/total per %gx%g image at batch-size %g' % t)
# Plots
if plots:
confusion_matrix.plot(save_dir=save_dir, names=list(names.values()))
if wandb and wandb.run:
wandb.log({"Images": wandb_images})
wandb.log({"Validation": [wandb.Image(str(f), caption=f.name) for f in sorted(save_dir.glob('test*.jpg'))]})
# Save JSON
if save_json and len(jdict):
w = Path(weights[0] if isinstance(weights, list) else weights).stem if weights is not None else '' # weights
anno_json = '../coco/annotations/instances_val2017.json' # annotations json
pred_json = str(save_dir / f"{w}_predictions.json") # predictions json
print('\nEvaluating pycocotools mAP... saving %s...' % pred_json)
with open(pred_json, 'w') as f:
json.dump(jdict, f)
try: # https://github.com/cocodataset/cocoapi/blob/master/PythonAPI/pycocoEvalDemo.ipynb
from pycocotools.coco import COCO
from pycocotools.cocoeval import COCOeval
anno = COCO(anno_json) # init annotations api
pred = anno.loadRes(pred_json) # init predictions api
eval = COCOeval(anno, pred, 'bbox')
if is_coco:
eval.params.imgIds = [int(Path(x).stem) for x in dataloader.dataset.img_files] # image IDs to evaluate
eval.evaluate()
eval.accumulate()
eval.summarize()
map, map50 = eval.stats[:2] # update results ([email protected]:0.95, [email protected])
except Exception as e:
print(f'pycocotools unable to run: {e}')
# Return results
if not training:
s = f"\n{len(list(save_dir.glob('labels/*.txt')))} labels saved to {save_dir / 'labels'}" if save_txt else ''
print(f"Results saved to {save_dir}{s}")
model.float() # for training
maps = np.zeros(nc) + map
for i, c in enumerate(ap_class):
maps[c] = ap[i]
return (mp, mr, map50, map, *(loss.cpu() / len(dataloader)).tolist()), maps, t
if __name__ == '__main__':
parser = argparse.ArgumentParser(prog='test.py')
parser.add_argument('--weights', nargs='+', type=str, default=ROOT / 'yolov5n-lmk.pt', help='model.pt path(s)')
parser.add_argument('--data', type=str, default=ROOT / 'data/widerface.yaml', help='*.data path')
parser.add_argument('--batch-size', type=int, default=32, help='size of each image batch')
parser.add_argument('--img-size', type=int, default=640, help='inference size (pixels)')
parser.add_argument('--conf-thres', type=float, default=0.001, help='object confidence threshold')
parser.add_argument('--iou-thres', type=float, default=0.6, help='IOU threshold for NMS')
parser.add_argument('--task', default='val', help="'val', 'test', 'study'")
parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
parser.add_argument('--single-cls', action='store_true', help='treat as single-class dataset')
parser.add_argument('--augment', action='store_true', help='augmented inference')
parser.add_argument('--verbose', action='store_true', help='report mAP by class')
parser.add_argument('--save-txt', action='store_true', help='save results to *.txt')
parser.add_argument('--save-hybrid', action='store_true', help='save label+prediction hybrid results to *.txt')
parser.add_argument('--save-conf', action='store_true', help='save confidences in --save-txt labels')
parser.add_argument('--save-json', action='store_true', help='save a cocoapi-compatible JSON results file')
parser.add_argument('--project', default='runs/test', help='save to project/name')
parser.add_argument('--name', default='exp', help='save to project/name')
parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment')
opt = parser.parse_args()
opt.save_json |= opt.data.endswith('coco.yaml')
opt.data = check_file(opt.data) # check file
print(opt)
if opt.task in ['val', 'test']: # run normally
test(opt.data,
opt.weights,
opt.batch_size,
opt.img_size,
opt.conf_thres,
opt.iou_thres,
opt.save_json,
opt.single_cls,
opt.augment,
opt.verbose,
save_txt=opt.save_txt | opt.save_hybrid,
save_hybrid=opt.save_hybrid,
save_conf=opt.save_conf,
)
elif opt.task == 'study': # run over a range of settings and save/plot
for weights in ['yolov5s.pt', 'yolov5m.pt', 'yolov5l.pt', 'yolov5x.pt']:
f = 'study_%s_%s.txt' % (Path(opt.data).stem, Path(weights).stem) # filename to save to
x = list(range(320, 800, 64)) # x axis
y = [] # y axis
for i in x: # img-size
print('\nRunning %s point %s...' % (f, i))
r, _, t = test(opt.data, weights, opt.batch_size, i, opt.conf_thres, opt.iou_thres, opt.save_json,
plots=False)
y.append(r + t) # results and times
np.savetxt(f, y, fmt='%10.4g') # save
os.system('zip -r study.zip study_*.txt')
plot_study_txt(f, x) # plot
landmark/predict.py
import argparse
import os
import platform
import sys
from pathlib import Path
import torch
FILE = Path(__file__).resolve()
ROOT = FILE.parents[1] # YOLOv5 root directory
if str(ROOT) not in sys.path:
sys.path.append(str(ROOT)) # add ROOT to PATH
ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative
from models.common import DetectMultiBackend
from utils.dataloaders import IMG_FORMATS, VID_FORMATS, LoadImages, LoadScreenshots, LoadStreams
from utils.general import (LOGGER, Profile, check_file, check_img_size, check_imshow, check_requirements, colorstr, cv2,
increment_path, print_args, scale_boxes,strip_optimizer)
from utils.landmark.general import non_max_suppression_face,scale_coords_landmarks
from utils.plots import Annotator, colors, save_one_box
from utils.torch_utils import select_device, smart_inference_mode
@smart_inference_mode()
def run(
weights=ROOT / 'yolov5n-lmk.pt', # model.pt path(s)
source=ROOT / 'data/images', # file/dir/URL/glob/screen/0(webcam)
data=ROOT / 'data/coco128.yaml', # dataset.yaml path
imgsz=(640, 640), # inference size (height, width)
conf_thres=0.25, # confidence threshold
iou_thres=0.45, # NMS IOU threshold
max_det=1000, # maximum detections per image
device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu
view_img=False, # show results
save_txt=False, # save results to *.txt
save_conf=False, # save confidences in --save-txt labels
save_crop=False, # save cropped prediction boxes
nosave=False, # do not save images/videos
classes=None, # filter by class: --class 0, or --class 0 2 3
agnostic_nms=False, # class-agnostic NMS
augment=False, # augmented inference
visualize=False, # visualize features
update=False, # update all models
project=ROOT / 'runs/predict-lmk', # save results to project/name
name='exp', # save results to project/name
exist_ok=False, # existing project/name ok, do not increment
line_thickness=3, # bounding box thickness (pixels)
hide_labels=False, # hide labels
hide_conf=False, # hide confidences
half=False, # use FP16 half-precision inference
dnn=False, # use OpenCV DNN for ONNX inference
vid_stride=1, # video frame-rate stride
):
source = str(source)
save_img = not nosave and not source.endswith('.txt') # save inference images
is_file = Path(source).suffix[1:] in (IMG_FORMATS + VID_FORMATS)
is_url = source.lower().startswith(('rtsp://', 'rtmp://', 'http://', 'https://'))
webcam = source.isnumeric() or source.endswith('.txt') or (is_url and not is_file)
screenshot = source.lower().startswith('screen')
if is_url and is_file:
source = check_file(source) # download
# Directories
save_dir = increment_path(Path(project) / name, exist_ok=exist_ok) # increment run
(save_dir / 'labels' if save_txt else save_dir).mkdir(parents=True, exist_ok=True) # make dir
# Load model
device = select_device(device)
model = DetectMultiBackend(weights, device=device, dnn=dnn, data=data, fp16=half)
stride, names, pt = model.stride, model.names, model.pt
imgsz = check_img_size(imgsz, s=stride) # check image size
# Dataloader
bs = 1 # batch_size
if webcam:
view_img = check_imshow()
dataset = LoadStreams(source, img_size=imgsz, stride=stride, auto=pt, vid_stride=vid_stride)
bs = len(dataset)
elif screenshot:
dataset = LoadScreenshots(source, img_size=imgsz, stride=stride, auto=pt)
else:
dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt, vid_stride=vid_stride)
vid_path, vid_writer = [None] * bs, [None] * bs
# Run inference
model.warmup(imgsz=(1 if pt else bs, 3, *imgsz)) # warmup
seen, windows, dt = 0, [], (Profile(), Profile(), Profile())
for path, im, im0s, vid_cap, s in dataset:
with dt[0]:
im = torch.from_numpy(im).to(model.device)
im = im.half() if model.fp16 else im.float() # uint8 to fp16/32
im /= 255 # 0 - 255 to 0.0 - 1.0
if len(im.shape) == 3:
im = im[None] # expand for batch dim
# Inference
with dt[1]:
visualize = increment_path(save_dir / Path(path).stem, mkdir=True) if visualize else False
pred, proto = model(im, augment=augment, visualize=visualize)[:2]
# NMS
with dt[2]:
pred = non_max_suppression_face(pred, conf_thres, iou_thres)
# Second-stage classifier (optional)
# pred = utils.general.apply_classifier(pred, classifier_model, im, im0s)
# Process predictions
for i, det in enumerate(pred): # per image
seen += 1
if webcam: # batch_size >= 1
p, im0, frame = path[i], im0s[i].copy(), dataset.count
s += f'{i}: '
else:
p, im0, frame = path, im0s.copy(), getattr(dataset, 'frame', 0)
p = Path(p) # to Path
save_path = str(save_dir / p.name) # im.jpg
txt_path = str(save_dir / 'labels' / p.stem) + ('' if dataset.mode == 'image' else f'_{frame}') # im.txt
s += '%gx%g ' % im.shape[2:] # print string
imc = im0.copy() if save_crop else im0 # for save_crop
annotator = Annotator(im0, line_width=line_thickness, example=str(names))
if len(det):
# Rescale boxes from img_size to im0 size
det[:, :4] = scale_boxes(im.shape[2:], det[:, :4], im0.shape).round()
# Print results
for c in det[:, 5].unique():
n = (det[:, 5] == c).sum() # detections per class
# s += f"{n} {names[int(c)]}{'s' * (n > 1)}, " # add to string
det[:, 5:15] = scale_coords_landmarks(im.shape[2:], det[:, 5:15], im0.shape).round()
# Write results
for j in range(det.size()[0]):
xyxy = det[j, :4].view(-1).tolist()
conf = det[j, 4].cpu().detach().numpy()
landmarks = det[j, 5:15].view(-1).tolist()
cls = det[j, 15].cpu().detach().numpy()
if save_txt:
line = (cls, *xyxy, conf) if save_conf else (cls, *xyxy)
with open(f'{txt_path}.txt', 'a') as f:
f.write(('%g ' * len(line)).rstrip() % line + '\n')
if save_img or save_crop or view_img: # Add bbox to image
c = int(cls) # integer class
label = None if hide_labels else (names[c] if hide_conf else f'{names[c]} {conf:.2f}')
annotator.box_label(xyxy, label, color=colors(c, True))
annotator.landmark(landmarks)
if save_crop:
save_one_box(xyxy, imc, file=save_dir / 'crops' / names[c] / f'{p.stem}.jpg', BGR=True)
# Stream results
im0 = annotator.result()
if view_img:
if platform.system() == 'Linux' and p not in windows:
windows.append(p)
cv2.namedWindow(str(p), cv2.WINDOW_NORMAL | cv2.WINDOW_KEEPRATIO) # allow window resize (Linux)
cv2.resizeWindow(str(p), im0.shape[1], im0.shape[0])
cv2.imshow(str(p), im0)
cv2.waitKey(1) # 1 millisecond
# Save results (image with detections)
if save_img:
if dataset.mode == 'image':
cv2.imwrite(save_path, im0)
else: # 'video' or 'stream'
if vid_path[i] != save_path: # new video
vid_path[i] = save_path
if isinstance(vid_writer[i], cv2.VideoWriter):
vid_writer[i].release() # release previous video writer
if vid_cap: # video
fps = vid_cap.get(cv2.CAP_PROP_FPS)
w = int(vid_cap.get(cv2.CAP_PROP_FRAME_WIDTH))
h = int(vid_cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
else: # stream
fps, w, h = 30, im0.shape[1], im0.shape[0]
save_path = str(Path(save_path).with_suffix('.mp4')) # force *.mp4 suffix on results videos
vid_writer[i] = cv2.VideoWriter(save_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (w, h))
vid_writer[i].write(im0)
# Print time (inference-only)
LOGGER.info(f"{s}{'' if len(det) else '(no detections), '}{dt[1].dt * 1E3:.1f}ms")
# Print results
t = tuple(x.t / seen * 1E3 for x in dt) # speeds per image
LOGGER.info(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {(1, 3, *imgsz)}' % t)
if save_txt or save_img:
s = f"\n{len(list(save_dir.glob('labels/*.txt')))} labels saved to {save_dir / 'labels'}" if save_txt else ''
LOGGER.info(f"Results saved to {colorstr('bold', save_dir)}{s}")
if update:
strip_optimizer(weights[0]) # update model (to fix SourceChangeWarning)
def parse_opt():
parser = argparse.ArgumentParser()
parser.add_argument('--weights', nargs='+', type=str, default=ROOT / 'yolov5n-lmk.pt', help='model path(s)')
parser.add_argument('--source', type=str, default=ROOT / 'data/images', help='file/dir/URL/glob/screen/0(webcam)')
parser.add_argument('--data', type=str, default=ROOT / 'data/widerface.yaml', help='(optional) dataset.yaml path')
parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[640], help='inference size h,w')
parser.add_argument('--conf-thres', type=float, default=0.25, help='confidence threshold')
parser.add_argument('--iou-thres', type=float, default=0.45, help='NMS IoU threshold')
parser.add_argument('--max-det', type=int, default=1000, help='maximum detections per image')
parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
parser.add_argument('--view-img', action='store_true', help='show results')
parser.add_argument('--save-txt', action='store_true', help='save results to *.txt')
parser.add_argument('--save-conf', action='store_true', help='save confidences in --save-txt labels')
parser.add_argument('--save-crop', action='store_true', help='save cropped prediction boxes')
parser.add_argument('--nosave', action='store_true', help='do not save images/videos')
parser.add_argument('--classes', nargs='+', type=int, help='filter by class: --classes 0, or --classes 0 2 3')
parser.add_argument('--agnostic-nms', action='store_true', help='class-agnostic NMS')
parser.add_argument('--augment', action='store_true', help='augmented inference')
parser.add_argument('--visualize', action='store_true', help='visualize features')
parser.add_argument('--update', action='store_true', help='update all models')
parser.add_argument('--project', default=ROOT / 'runs/predict-lmk', help='save results to project/name')
parser.add_argument('--name', default='exp', help='save results to project/name')
parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment')
parser.add_argument('--line-thickness', default=3, type=int, help='bounding box thickness (pixels)')
parser.add_argument('--hide-labels', default=False, action='store_true', help='hide labels')
parser.add_argument('--hide-conf', default=False, action='store_true', help='hide confidences')
parser.add_argument('--half', action='store_true', help='use FP16 half-precision inference')
parser.add_argument('--dnn', action='store_true', help='use OpenCV DNN for ONNX inference')
parser.add_argument('--vid-stride', type=int, default=1, help='video frame-rate stride')
opt = parser.parse_args()
opt.imgsz *= 2 if len(opt.imgsz) == 1 else 1 # expand
print_args(vars(opt))
return opt
def main(opt):
check_requirements(exclude=('tensorboard', 'thop'))
run(**vars(opt))
if __name__ == "__main__":
opt = parse_opt()
main(opt)
landmark/export.py
# YOLOv5 by Ultralytics, GPL-3.0 license
"""
Export a YOLOv5 PyTorch model to other formats. TensorFlow exports authored by https://github.com/zldrobit
Format | `export.py --include` | Model
--- | --- | ---
PyTorch | - | yolov5s.pt
TorchScript | `torchscript` | yolov5s.torchscript
ONNX | `onnx` | yolov5s.onnx
OpenVINO | `openvino` | yolov5s_openvino_model/
TensorRT | `engine` | yolov5s.engine
CoreML | `coreml` | yolov5s.mlmodel
TensorFlow SavedModel | `saved_model` | yolov5s_saved_model/
TensorFlow GraphDef | `pb` | yolov5s.pb
TensorFlow Lite | `tflite` | yolov5s.tflite
TensorFlow Edge TPU | `edgetpu` | yolov5s_edgetpu.tflite
TensorFlow.js | `tfjs` | yolov5s_web_model/
PaddlePaddle | `paddle` | yolov5s_paddle_model/
Requirements:
$ pip install -r requirements.txt coremltools onnx onnx-simplifier onnxruntime openvino-dev tensorflow-cpu # CPU
$ pip install -r requirements.txt coremltools onnx onnx-simplifier onnxruntime-gpu openvino-dev tensorflow # GPU
Usage:
$ python export.py --weights yolov5s.pt --include torchscript onnx openvino engine coreml tflite ...
Inference:
$ python detect.py --weights yolov5s.pt # PyTorch
yolov5s.torchscript # TorchScript
yolov5s.onnx # ONNX Runtime or OpenCV DNN with --dnn
yolov5s.xml # OpenVINO
yolov5s.engine # TensorRT
yolov5s.mlmodel # CoreML (macOS-only)
yolov5s_saved_model # TensorFlow SavedModel
yolov5s.pb # TensorFlow GraphDef
yolov5s.tflite # TensorFlow Lite
yolov5s_edgetpu.tflite # TensorFlow Edge TPU
yolov5s_paddle_model # PaddlePaddle
TensorFlow.js:
$ cd .. && git clone https://github.com/zldrobit/tfjs-yolov5-example.git && cd tfjs-yolov5-example
$ npm install
$ ln -s ../../yolov5/yolov5s_web_model public/yolov5s_web_model
$ npm start
"""
import argparse
import json
import os
import platform
import re
import subprocess
import sys
import time
import warnings
from pathlib import Path
import pandas as pd
import torch
import torch.nn as nn
from torch.utils.mobile_optimizer import optimize_for_mobile
FILE = Path(__file__).resolve()
ROOT = FILE.parents[1] # YOLOv5 root directory
if str(ROOT) not in sys.path:
sys.path.append(str(ROOT)) # add ROOT to PATH
if platform.system() != 'Windows':
ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative
from models.experimental import attempt_loadf
from models.yolo import ClassificationModel, Detect, DetectionModel, SegmentationModel, LmkDetect
from utils.dataloaders import LoadImages
from utils.general import (LOGGER, Profile, check_dataset, check_img_size, check_requirements, check_version,
check_yaml, colorstr, file_size, get_default_args, print_args, url2file, yaml_save)
from utils.torch_utils import select_device, smart_inference_mode
MACOS = platform.system() == 'Darwin' # macOS environment
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'])
def try_export(inner_func):
# YOLOv5 export decorator, i..e @try_export
inner_args = get_default_args(inner_func)
def outer_func(*args, **kwargs):
prefix = inner_args['prefix']
try:
with Profile() as dt:
f, model = inner_func(*args, **kwargs)
LOGGER.info(f'{prefix} export success ✅ {dt.t:.1f}s, saved as {f} ({file_size(f):.1f} MB)')
return f, model
except Exception as e:
LOGGER.info(f'{prefix} export failure ❌ {dt.t:.1f}s: {e}')
return None, None
return outer_func
@try_export
def export_torchscript(model, im, file, optimize, prefix=colorstr('TorchScript:')):
# YOLOv5 TorchScript model export
LOGGER.info(f'\n{prefix} starting export with torch {torch.__version__}...')
f = file.with_suffix('.torchscript')
ts = torch.jit.trace(model, im, strict=False)
d = {"shape": im.shape, "stride": int(max(model.stride)), "names": model.names}
extra_files = {'config.txt': json.dumps(d)} # torch._C.ExtraFilesMap()
if optimize: # https://pytorch.org/tutorials/recipes/mobile_interpreter.html
optimize_for_mobile(ts)._save_for_lite_interpreter(str(f), _extra_files=extra_files)
else:
ts.save(str(f), _extra_files=extra_files)
return f, None
@try_export
def export_onnx(model, im, file, opset, dynamic, simplify, prefix=colorstr('ONNX:')):
# YOLOv5 ONNX export
check_requirements('onnx')
import onnx
LOGGER.info(f'\n{prefix} starting export with onnx {onnx.__version__}...')
f = file.with_suffix('.onnx')
output_names = ['output0', 'output1'] if isinstance(model, SegmentationModel) else ['output0']
if dynamic:
dynamic = {'images': {0: 'batch', 2: 'height', 3: 'width'}} # shape(1,3,640,640)
if isinstance(model, SegmentationModel):
dynamic['output0'] = {0: 'batch', 1: 'anchors'} # shape(1,25200,85)
dynamic['output1'] = {0: 'batch', 2: 'mask_height', 3: 'mask_width'} # shape(1,32,160,160)
elif isinstance(model, DetectionModel):
dynamic['output0'] = {0: 'batch', 1: 'anchors'} # shape(1,25200,85)
torch.onnx.export(
model.cpu() if dynamic else model, # --dynamic only compatible with cpu
im.cpu() if dynamic else im,
f,
verbose=False,
opset_version=opset,
do_constant_folding=True,
input_names=['images'],
output_names=output_names,
dynamic_axes=dynamic or None)
# Checks
model_onnx = onnx.load(f) # load onnx model
onnx.checker.check_model(model_onnx) # check onnx model
# Metadata
d = {'stride': int(max(model.stride)), 'names': model.names}
for k, v in d.items():
meta = model_onnx.metadata_props.add()
meta.key, meta.value = k, str(v)
onnx.save(model_onnx, f)
# Simplify
if simplify:
try:
cuda = torch.cuda.is_available()
check_requirements(('onnxruntime-gpu' if cuda else 'onnxruntime', 'onnx-simplifier>=0.4.1'))
import onnxsim
LOGGER.info(f'{prefix} simplifying with onnx-simplifier {onnxsim.__version__}...')
model_onnx, check = onnxsim.simplify(model_onnx)
assert check, 'assert check failed'
onnx.save(model_onnx, f)
except Exception as e:
LOGGER.info(f'{prefix} simplifier failure: {e}')
return f, model_onnx
@try_export
def export_openvino(file, metadata, half, prefix=colorstr('OpenVINO:')):
# YOLOv5 OpenVINO export
check_requirements('openvino-dev') # requires openvino-dev: https://pypi.org/project/openvino-dev/
import openvino.inference_engine as ie
LOGGER.info(f'\n{prefix} starting export with openvino {ie.__version__}...')
f = str(file).replace('.pt', f'_openvino_model{os.sep}')
cmd = f"mo --input_model {file.with_suffix('.onnx')} --output_dir {f} --data_type {'FP16' if half else 'FP32'}"
subprocess.run(cmd.split(), check=True, env=os.environ) # export
yaml_save(Path(f) / file.with_suffix('.yaml').name, metadata) # add metadata.yaml
return f, None
@try_export
def export_paddle(model, im, file, metadata, prefix=colorstr('PaddlePaddle:')):
# YOLOv5 Paddle export
check_requirements(('paddlepaddle', 'x2paddle'))
import x2paddle
from x2paddle.convert import pytorch2paddle
LOGGER.info(f'\n{prefix} starting export with X2Paddle {x2paddle.__version__}...')
f = str(file).replace('.pt', f'_paddle_model{os.sep}')
pytorch2paddle(module=model, save_dir=f, jit_type='trace', input_examples=[im]) # export
yaml_save(Path(f) / file.with_suffix('.yaml').name, metadata) # add metadata.yaml
return f, None
@try_export
def export_coreml(model, im, file, int8, half, prefix=colorstr('CoreML:')):
# YOLOv5 CoreML export
check_requirements('coremltools')
import coremltools as ct
LOGGER.info(f'\n{prefix} starting export with coremltools {ct.__version__}...')
f = file.with_suffix('.mlmodel')
ts = torch.jit.trace(model, im, strict=False) # TorchScript model
ct_model = ct.convert(ts, inputs=[ct.ImageType('image', shape=im.shape, scale=1 / 255, bias=[0, 0, 0])])
bits, mode = (8, 'kmeans_lut') if int8 else (16, 'linear') if half else (32, None)
if bits < 32:
if MACOS: # quantization only supported on macOS
with warnings.catch_warnings():
warnings.filterwarnings("ignore", category=DeprecationWarning) # suppress numpy==1.20 float warning
ct_model = ct.models.neural_network.quantization_utils.quantize_weights(ct_model, bits, mode)
else:
print(f'{prefix} quantization only supported on macOS, skipping...')
ct_model.save(f)
return f, ct_model
@try_export
def export_engine(model, im, file, half, dynamic, simplify, workspace=4, verbose=False, prefix=colorstr('TensorRT:')):
# YOLOv5 TensorRT export https://developer.nvidia.com/tensorrt
assert im.device.type != 'cpu', 'export running on CPU but must be on GPU, i.e. `python export.py --device 0`'
try:
import tensorrt as trt
except Exception:
if platform.system() == 'Linux':
check_requirements('nvidia-tensorrt', cmds='-U --index-url https://pypi.ngc.nvidia.com')
import tensorrt as trt
if trt.__version__[0] == '7': # TensorRT 7 handling https://github.com/ultralytics/yolov5/issues/6012
grid = model.model[-1].anchor_grid
model.model[-1].anchor_grid = [a[..., :1, :1, :] for a in grid]
export_onnx(model, im, file, 12, dynamic, simplify) # opset 12
model.model[-1].anchor_grid = grid
else: # TensorRT >= 8
check_version(trt.__version__, '8.0.0', hard=True) # require tensorrt>=8.0.0
export_onnx(model, im, file, 12, dynamic, simplify) # opset 12
onnx = file.with_suffix('.onnx')
LOGGER.info(f'\n{prefix} starting export with TensorRT {trt.__version__}...')
assert onnx.exists(), f'failed to export ONNX file: {onnx}'
f = file.with_suffix('.engine') # TensorRT engine file
logger = trt.Logger(trt.Logger.INFO)
if verbose:
logger.min_severity = trt.Logger.Severity.VERBOSE
builder = trt.Builder(logger)
config = builder.create_builder_config()
config.max_workspace_size = workspace * 1 << 30
# config.set_memory_pool_limit(trt.MemoryPoolType.WORKSPACE, workspace << 30) # fix TRT 8.4 deprecation notice
flag = (1 << int(trt.NetworkDefinitionCreationFlag.EXPLICIT_BATCH))
network = builder.create_network(flag)
parser = trt.OnnxParser(network, logger)
if not parser.parse_from_file(str(onnx)):
raise RuntimeError(f'failed to load ONNX file: {onnx}')
inputs = [network.get_input(i) for i in range(network.num_inputs)]
outputs = [network.get_output(i) for i in range(network.num_outputs)]
for inp in inputs:
LOGGER.info(f'{prefix} input "{inp.name}" with shape{inp.shape} {inp.dtype}')
for out in outputs:
LOGGER.info(f'{prefix} output "{out.name}" with shape{out.shape} {out.dtype}')
if dynamic:
if im.shape[0] <= 1:
LOGGER.warning(f"{prefix} WARNING ⚠️ --dynamic model requires maximum --batch-size argument")
profile = builder.create_optimization_profile()
for inp in inputs:
profile.set_shape(inp.name, (1, *im.shape[1:]), (max(1, im.shape[0] // 2), *im.shape[1:]), im.shape)
config.add_optimization_profile(profile)
LOGGER.info(f'{prefix} building FP{16 if builder.platform_has_fast_fp16 and half else 32} engine as {f}')
if builder.platform_has_fast_fp16 and half:
config.set_flag(trt.BuilderFlag.FP16)
with builder.build_engine(network, config) as engine, open(f, 'wb') as t:
t.write(engine.serialize())
return f, None
@try_export
def export_saved_model(model,
im,
file,
dynamic,
tf_nms=False,
agnostic_nms=False,
topk_per_class=100,
topk_all=100,
iou_thres=0.45,
conf_thres=0.25,
keras=False,
prefix=colorstr('TensorFlow SavedModel:')):
# YOLOv5 TensorFlow SavedModel export
try:
import tensorflow as tf
except Exception:
check_requirements(f"tensorflow{'' if torch.cuda.is_available() else '-macos' if MACOS else '-cpu'}")
import tensorflow as tf
from tensorflow.python.framework.convert_to_constants import convert_variables_to_constants_v2
from models.tf import TFModel
LOGGER.info(f'\n{prefix} starting export with tensorflow {tf.__version__}...')
f = str(file).replace('.pt', '_saved_model')
batch_size, ch, *imgsz = list(im.shape) # BCHW
tf_model = TFModel(cfg=model.yaml, model=model, nc=model.nc, imgsz=imgsz)
im = tf.zeros((batch_size, *imgsz, ch)) # BHWC order for TensorFlow
_ = tf_model.predict(im, tf_nms, agnostic_nms, topk_per_class, topk_all, iou_thres, conf_thres)
inputs = tf.keras.Input(shape=(*imgsz, ch), batch_size=None if dynamic else batch_size)
outputs = tf_model.predict(inputs, tf_nms, agnostic_nms, topk_per_class, topk_all, iou_thres, conf_thres)
keras_model = tf.keras.Model(inputs=inputs, outputs=outputs)
keras_model.trainable = False
keras_model.summary()
if keras:
keras_model.save(f, save_format='tf')
else:
spec = tf.TensorSpec(keras_model.inputs[0].shape, keras_model.inputs[0].dtype)
m = tf.function(lambda x: keras_model(x)) # full model
m = m.get_concrete_function(spec)
frozen_func = convert_variables_to_constants_v2(m)
tfm = tf.Module()
tfm.__call__ = tf.function(lambda x: frozen_func(x)[:4] if tf_nms else frozen_func(x), [spec])
tfm.__call__(im)
tf.saved_model.save(tfm,
f,
options=tf.saved_model.SaveOptions(experimental_custom_gradients=False) if check_version(
tf.__version__, '2.6') else tf.saved_model.SaveOptions())
return f, keras_model
@try_export
def export_pb(keras_model, file, prefix=colorstr('TensorFlow GraphDef:')):
# YOLOv5 TensorFlow GraphDef *.pb export https://github.com/leimao/Frozen_Graph_TensorFlow
import tensorflow as tf
from tensorflow.python.framework.convert_to_constants import convert_variables_to_constants_v2
LOGGER.info(f'\n{prefix} starting export with tensorflow {tf.__version__}...')
f = file.with_suffix('.pb')
m = tf.function(lambda x: keras_model(x)) # full model
m = m.get_concrete_function(tf.TensorSpec(keras_model.inputs[0].shape, keras_model.inputs[0].dtype))
frozen_func = convert_variables_to_constants_v2(m)
frozen_func.graph.as_graph_def()
tf.io.write_graph(graph_or_graph_def=frozen_func.graph, logdir=str(f.parent), name=f.name, as_text=False)
return f, None
@try_export
def export_tflite(keras_model, im, file, int8, data, nms, agnostic_nms, prefix=colorstr('TensorFlow Lite:')):
# YOLOv5 TensorFlow Lite export
import tensorflow as tf
LOGGER.info(f'\n{prefix} starting export with tensorflow {tf.__version__}...')
batch_size, ch, *imgsz = list(im.shape) # BCHW
f = str(file).replace('.pt', '-fp16.tflite')
converter = tf.lite.TFLiteConverter.from_keras_model(keras_model)
converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS]
converter.target_spec.supported_types = [tf.float16]
converter.optimizations = [tf.lite.Optimize.DEFAULT]
if int8:
from models.tf import representative_dataset_gen
dataset = LoadImages(check_dataset(check_yaml(data))['train'], img_size=imgsz, auto=False)
converter.representative_dataset = lambda: representative_dataset_gen(dataset, ncalib=100)
converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8]
converter.target_spec.supported_types = []
converter.inference_input_type = tf.uint8 # or tf.int8
converter.inference_output_type = tf.uint8 # or tf.int8
converter.experimental_new_quantizer = True
f = str(file).replace('.pt', '-int8.tflite')
if nms or agnostic_nms:
converter.target_spec.supported_ops.append(tf.lite.OpsSet.SELECT_TF_OPS)
tflite_model = converter.convert()
open(f, "wb").write(tflite_model)
return f, None
@try_export
def export_edgetpu(file, prefix=colorstr('Edge TPU:')):
# YOLOv5 Edge TPU export https://coral.ai/docs/edgetpu/models-intro/
cmd = 'edgetpu_compiler --version'
help_url = 'https://coral.ai/docs/edgetpu/compiler/'
assert platform.system() == 'Linux', f'export only supported on Linux. See {help_url}'
if subprocess.run(f'{cmd} >/dev/null', shell=True).returncode != 0:
LOGGER.info(f'\n{prefix} export requires Edge TPU compiler. Attempting install from {help_url}')
sudo = subprocess.run('sudo --version >/dev/null', shell=True).returncode == 0 # sudo installed on system
for c in (
'curl https://packages.cloud.google.com/apt/doc/apt-key.gpg | sudo apt-key add -',
'echo "deb https://packages.cloud.google.com/apt coral-edgetpu-stable main" | sudo tee /etc/apt/sources.list.d/coral-edgetpu.list',
'sudo apt-get update', 'sudo apt-get install edgetpu-compiler'):
subprocess.run(c if sudo else c.replace('sudo ', ''), shell=True, check=True)
ver = subprocess.run(cmd, shell=True, capture_output=True, check=True).stdout.decode().split()[-1]
LOGGER.info(f'\n{prefix} starting export with Edge TPU compiler {ver}...')
f = str(file).replace('.pt', '-int8_edgetpu.tflite') # Edge TPU model
f_tfl = str(file).replace('.pt', '-int8.tflite') # TFLite model
cmd = f"edgetpu_compiler -s -d -k 10 --out_dir {file.parent} {f_tfl}"
subprocess.run(cmd.split(), check=True)
return f, None
@try_export
def export_tfjs(file, prefix=colorstr('TensorFlow.js:')):
# YOLOv5 TensorFlow.js export
check_requirements('tensorflowjs')
import tensorflowjs as tfjs
LOGGER.info(f'\n{prefix} starting export with tensorflowjs {tfjs.__version__}...')
f = str(file).replace('.pt', '_web_model') # js dir
f_pb = file.with_suffix('.pb') # *.pb path
f_json = f'{f}/model.json' # *.json path
cmd = f'tensorflowjs_converter --input_format=tf_frozen_model ' \
f'--output_node_names=Identity,Identity_1,Identity_2,Identity_3 {f_pb} {f}'
subprocess.run(cmd.split())
json = Path(f_json).read_text()
with open(f_json, 'w') as j: # sort JSON Identity_* in ascending order
subst = re.sub(
r'{"outputs": {"Identity.?.?": {"name": "Identity.?.?"}, '
r'"Identity.?.?": {"name": "Identity.?.?"}, '
r'"Identity.?.?": {"name": "Identity.?.?"}, '
r'"Identity.?.?": {"name": "Identity.?.?"}}}', r'{"outputs": {"Identity": {"name": "Identity"}, '
r'"Identity_1": {"name": "Identity_1"}, '
r'"Identity_2": {"name": "Identity_2"}, '
r'"Identity_3": {"name": "Identity_3"}}}', json)
j.write(subst)
return f, None
@smart_inference_mode()
def run(
data=ROOT / 'data/coco128.yaml', # 'dataset.yaml path'
weights=ROOT / 'yolov5s.pt', # weights path
imgsz=(640, 640), # image (height, width)
batch_size=1, # batch size
device='cpu', # cuda device, i.e. 0 or 0,1,2,3 or cpu
include=('torchscript', 'onnx'), # include formats
half=False, # FP16 half-precision export
inplace=False, # set YOLOv5 Detect() inplace=True
keras=False, # use Keras
optimize=False, # TorchScript: optimize for mobile
int8=False, # CoreML/TF INT8 quantization
dynamic=False, # ONNX/TF/TensorRT: dynamic axes
simplify=False, # ONNX: simplify model
opset=12, # ONNX: opset version
verbose=False, # TensorRT: verbose log
workspace=4, # TensorRT: workspace size (GB)
nms=False, # TF: add NMS to model
agnostic_nms=False, # TF: add agnostic NMS to model
topk_per_class=100, # TF.js NMS: topk per class to keep
topk_all=100, # TF.js NMS: topk for all classes to keep
iou_thres=0.45, # TF.js NMS: IoU threshold
conf_thres=0.25, # TF.js NMS: confidence threshold
):
t = time.time()
include = [x.lower() for x in include] # to lowercase
fmts = tuple(export_formats()['Argument'][1:]) # --include arguments
flags = [x in include for x in fmts]
assert sum(flags) == len(include), f'ERROR: Invalid --include {include}, valid --include arguments are {fmts}'
jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, paddle = flags # export booleans
file = Path(url2file(weights) if str(weights).startswith(('http:/', 'https:/')) else weights) # PyTorch weights
# Load PyTorch model
device = select_device(device)
if half:
assert device.type != 'cpu' or coreml, '--half only compatible with GPU export, i.e. use --device 0'
assert not dynamic, '--half not compatible with --dynamic, i.e. use either --half or --dynamic but not both'
model = attempt_loadf(weights, device=device, inplace=True, fuse=True) # load FP32 model
# Checks
imgsz *= 2 if len(imgsz) == 1 else 1 # expand
if optimize:
assert device.type == 'cpu', '--optimize not compatible with cuda devices, i.e. use --device cpu'
# Input
gs = int(max(model.stride)) # grid size (max stride)
imgsz = [check_img_size(x, gs) for x in imgsz] # verify img_size are gs-multiples
im = torch.zeros(batch_size, 3, *imgsz).to(device) # image size(1,3,320,192) BCHW iDetection
# Update model
model.eval()
for k, m in model.named_modules():
m._non_persistent_buffers_set = set() # pytorch 1.6.0 compatibility
if isinstance(m, (Detect, LmkDetect)):
m.inplace = inplace
m.dynamic = dynamic
m.export = True
for _ in range(2):
y = model(im) # dry runs
if half and not coreml:
im, model = im.half(), model.half() # to FP16
shape = tuple((y[0] if isinstance(y, tuple) else y).shape) # model output shape
metadata = {'stride': int(max(model.stride)), 'names': model.names} # model metadata
LOGGER.info(f"\n{colorstr('PyTorch:')} starting from {file} with output shape {shape} ({file_size(file):.1f} MB)")
# Exports
f = [''] * len(fmts) # exported filenames
warnings.filterwarnings(action='ignore', category=torch.jit.TracerWarning) # suppress TracerWarning
if jit: # TorchScript
f[0], _ = export_torchscript(model, im, file, optimize)
if engine: # TensorRT required before ONNX
f[1], _ = export_engine(model, im, file, half, dynamic, simplify, workspace, verbose)
if onnx or xml: # OpenVINO requires ONNX
f[2], _ = export_onnx(model, im, file, opset, dynamic, simplify)
if xml: # OpenVINO
f[3], _ = export_openvino(file, metadata, half)
if coreml: # CoreML
f[4], _ = export_coreml(model, im, file, int8, half)
if any((saved_model, pb, tflite, edgetpu, tfjs)): # TensorFlow formats
assert not tflite or not tfjs, 'TFLite and TF.js models must be exported separately, please pass only one type.'
assert not isinstance(model, ClassificationModel), 'ClassificationModel export to TF formats not yet supported.'
f[5], s_model = export_saved_model(model.cpu(),
im,
file,
dynamic,
tf_nms=nms or agnostic_nms or tfjs,
agnostic_nms=agnostic_nms or tfjs,
topk_per_class=topk_per_class,
topk_all=topk_all,
iou_thres=iou_thres,
conf_thres=conf_thres,
keras=keras)
if pb or tfjs: # pb prerequisite to tfjs
f[6], _ = export_pb(s_model, file)
if tflite or edgetpu:
f[7], _ = export_tflite(s_model, im, file, int8 or edgetpu, data=data, nms=nms, agnostic_nms=agnostic_nms)
if edgetpu:
f[8], _ = export_edgetpu(file)
if tfjs:
f[9], _ = export_tfjs(file)
if paddle: # PaddlePaddle
f[10], _ = export_paddle(model, im, file, metadata)
# Finish
f = [str(x) for x in f if x] # filter out '' and None
if any(f):
cls, det, seg = (isinstance(model, x) for x in (ClassificationModel, DetectionModel, SegmentationModel)) # type
dir = Path('segment' if seg else 'classify' if cls else '')
h = '--half' if half else '' # --half FP16 inference arg
s = "# WARNING ⚠️ ClassificationModel not yet supported for PyTorch Hub AutoShape inference" if cls else \
"# WARNING ⚠️ SegmentationModel not yet supported for PyTorch Hub AutoShape inference" if seg else ''
LOGGER.info(f'\nExport complete ({time.time() - t:.1f}s)'
f"\nResults saved to {colorstr('bold', file.parent.resolve())}"
f"\nDetect: python {dir / ('detect.py' if det else 'predict.py')} --weights {f[-1]} {h}"
f"\nValidate: python {dir / 'val.py'} --weights {f[-1]} {h}"
f"\nPyTorch Hub: model = torch.hub.load('ultralytics/yolov5', 'custom', '{f[-1]}') {s}"
f"\nVisualize: https://netron.app")
return f # return list of exported files/dirs
def parse_opt():
parser = argparse.ArgumentParser()
parser.add_argument('--data', type=str, default=ROOT / 'data/widerface.yaml', help='dataset.yaml path')
parser.add_argument('--weights', nargs='+', type=str, default=ROOT / 'yolov5n-lmk.pt', help='model.pt path(s)')
parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[640, 640], help='image (h, w)')
parser.add_argument('--batch-size', type=int, default=1, help='batch size')
parser.add_argument('--device', default='cpu', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
parser.add_argument('--half', action='store_true', help='FP16 half-precision export')
parser.add_argument('--inplace', action='store_true', help='set YOLOv5 Detect() inplace=True')
parser.add_argument('--keras', action='store_true', help='TF: use Keras')
parser.add_argument('--optimize', action='store_true', help='TorchScript: optimize for mobile')
parser.add_argument('--int8', action='store_true', help='CoreML/TF INT8 quantization')
parser.add_argument('--dynamic', action='store_true', help='ONNX/TF/TensorRT: dynamic axes')
parser.add_argument('--simplify', action='store_true', help='ONNX: simplify model')
parser.add_argument('--opset', type=int, default=11, help='ONNX: opset version')
parser.add_argument('--verbose', action='store_true', help='TensorRT: verbose log')
parser.add_argument('--workspace', type=int, default=4, help='TensorRT: workspace size (GB)')
parser.add_argument('--nms', action='store_true', help='TF: add NMS to model')
parser.add_argument('--agnostic-nms', action='store_true', help='TF: add agnostic NMS to model')
parser.add_argument('--topk-per-class', type=int, default=100, help='TF.js NMS: topk per class to keep')
parser.add_argument('--topk-all', type=int, default=100, help='TF.js NMS: topk for all classes to keep')
parser.add_argument('--iou-thres', type=float, default=0.45, help='TF.js NMS: IoU threshold')
parser.add_argument('--conf-thres', type=float, default=0.25, help='TF.js NMS: confidence threshold')
parser.add_argument(
'--include',
nargs='+',
default=['onnx'],
help='torchscript, onnx, openvino, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, paddle')
opt = parser.parse_args()
print_args(vars(opt))
return opt
def main(opt):
for opt.weights in (opt.weights if isinstance(opt.weights, list) else [opt.weights]):
run(**vars(opt))
if __name__ == "__main__":
opt = parse_opt()
main(opt)
utils文件夹
添加完以上三个文件后在utils文件夹中添加landmark文件夹
utils/landmark/general.py
# YOLOv5 by Ultralytics, GPL-3.0 license
"""
General utils
"""
import time
import torch
import torchvision
from ..metrics import box_iou
from ..general import xywh2xyxy
# for face landmark
def non_max_suppression_face(prediction, conf_thres=0.25, iou_thres=0.45, classes=None, agnostic=False, labels=()):
"""Performs Non-Maximum Suppression (NMS) on inference results
Returns:
detections with shape: nx16 (x1, y1, x2, y2, conf, keypoint*10, cls)
"""
nc = prediction.shape[2] - 15 # number of classes
xc = prediction[..., 4] > conf_thres # candidates
# Settings
min_wh, max_wh = 2, 4096 # (pixels) minimum and maximum box width and height
time_limit = 10.0 # seconds to quit after
redundant = True # require redundant detections
multi_label = nc > 1 # multiple labels per box (adds 0.5ms/img)
merge = False # use merge-NMS
t = time.time()
output = [torch.zeros((0, 16), device=prediction.device)] * prediction.shape[0]
for xi, x in enumerate(prediction): # image index, image inference
# Apply constraints
# x[((x[..., 2:4] < min_wh) | (x[..., 2:4] > max_wh)).any(1), 4] = 0 # width-height
x = x[xc[xi]] # confidence
# Cat apriori labels if autolabelling
if labels and len(labels[xi]):
l = labels[xi]
v = torch.zeros((len(l), nc + 15), device=x.device)
v[:, :4] = l[:, 1:5] # box
v[:, 4] = 1.0 # conf
v[range(len(l)), l[:, 0].long() + 15] = 1.0 # cls
x = torch.cat((x, v), 0)
# If none remain process next image
if not x.shape[0]:
continue
# Compute conf
x[:, 15:] *= x[:, 4:5] # conf = obj_conf * cls_conf
# Box (center x, center y, width, height) to (x1, y1, x2, y2)
box = xywh2xyxy(x[:, :4])
# Detections matrix nx6 (xyxy, conf, landmarks, cls)
if multi_label:
i, j = (x[:, 15:] > conf_thres).nonzero(as_tuple=False).T
x = torch.cat((box[i], x[i, j + 15, None], x[i, 5:15], j[:, None].float()), 1)
else: # best class only
conf, j = x[:, 15:].max(1, keepdim=True)
x = torch.cat((box, conf, x[:, 5:15], j.float()), 1)[conf.view(-1) > conf_thres]
# Filter by class
if classes is not None:
x = x[(x[:, 5:6] == torch.tensor(classes, device=x.device)).any(1)]
# If none remain process next image
n = x.shape[0] # number of boxes
if not n:
continue
# Batched NMS
c = x[:, 15:16] * (0 if agnostic else max_wh) # classes
boxes, scores = x[:, :4] + c, x[:, 4] # boxes (offset by class), scores
i = torchvision.ops.nms(boxes, scores, iou_thres) # NMS
# if i.shape[0] > max_det: # limit detections
# i = i[:max_det]
if merge and (1 < n < 3E3): # Merge NMS (boxes merged using weighted mean)
# update boxes as boxes(i,4) = weights(i,n) * boxes(n,4)
iou = box_iou(boxes[i], boxes) > iou_thres # iou matrix
weights = iou * scores[None] # box weights
x[i, :4] = torch.mm(weights, x[:, :4]).float() / weights.sum(1, keepdim=True) # merged boxes
if redundant:
i = i[iou.sum(1) > 1] # require redundancy
output[xi] = x[i]
if (time.time() - t) > time_limit:
break # time limit exceeded
return output
def scale_coords_landmarks(img1_shape, coords, img0_shape, ratio_pad=None):
# Rescale coords (xyxy) from img1_shape to img0_shape
if ratio_pad is None: # calculate from img0_shape
gain = min(img1_shape[0] / img0_shape[0], img1_shape[1] / img0_shape[1]) # gain = old / new
pad = (img1_shape[1] - img0_shape[1] * gain) / 2, (img1_shape[0] - img0_shape[0] * gain) / 2 # wh padding
else:
gain = ratio_pad[0][0]
pad = ratio_pad[1]
coords[:, [0, 2, 4, 6, 8]] -= pad[0] # x padding
coords[:, [1, 3, 5, 7, 9]] -= pad[1] # y padding
coords[:, :10] /= gain
# clip_coords(coords, img0_shape)
coords[:, 0].clamp_(0, img0_shape[1]) # x1
coords[:, 1].clamp_(0, img0_shape[0]) # y1
coords[:, 2].clamp_(0, img0_shape[1]) # x2
coords[:, 3].clamp_(0, img0_shape[0]) # y2
coords[:, 4].clamp_(0, img0_shape[1]) # x3
coords[:, 5].clamp_(0, img0_shape[0]) # y3
coords[:, 6].clamp_(0, img0_shape[1]) # x4
coords[:, 7].clamp_(0, img0_shape[0]) # y4
coords[:, 8].clamp_(0, img0_shape[1]) # x5
coords[:, 9].clamp_(0, img0_shape[0]) # y5
return coords
utils/landmark/loss.py
import numpy as np
import torch
import torch.nn as nn
from ..loss import FocalLoss, smooth_BCE
from ..landmark.metrics import bbox_iouf
from ..torch_utils import is_parallel
class WingLoss(nn.Module):
def __init__(self, w=10, e=2):
super(WingLoss, self).__init__()
# https://arxiv.org/pdf/1711.06753v4.pdf Figure 5
self.w = w
self.e = e
self.C = self.w - self.w * np.log(1 + self.w / self.e)
def forward(self, x, t, sigma=1):
weight = torch.ones_like(t)
weight[torch.where(t == -1)] = 0
diff = weight * (x - t)
abs_diff = diff.abs()
flag = (abs_diff.data < self.w).float()
y = flag * self.w * torch.log(1 + abs_diff / self.e) + (1 - flag) * (abs_diff - self.C)
return y.sum()
class LandmarksLoss(nn.Module):
# BCEwithLogitLoss() with reduced missing label effects.
def __init__(self, alpha=1.0):
super(LandmarksLoss, self).__init__()
self.loss_fcn = WingLoss() # nn.SmoothL1Loss(reduction='sum')
self.alpha = alpha
def forward(self, pred, truel, mask):
loss = self.loss_fcn(pred * mask, truel * mask)
return loss / (torch.sum(mask) + 10e-14)
def compute_loss(p, targets, model): # predictions, targets, model
device = targets.device
lcls, lbox, lobj, lmark = torch.zeros(1, device=device), \
torch.zeros(1, device=device), \
torch.zeros(1,device=device), \
torch.zeros(1, device=device)
tcls, tbox, indices, anchors, tlandmarks, lmks_mask = build_targets(p, targets, model) # targets
h = model.hyp # hyperparameters
# Define criteria
BCEcls = nn.BCEWithLogitsLoss(pos_weight=torch.tensor([h['cls_pw']], device=device)) # weight=model.class_weights)
BCEobj = nn.BCEWithLogitsLoss(pos_weight=torch.tensor([h['obj_pw']], device=device))
landmarks_loss = LandmarksLoss(1.0)
# Class label smoothing https://arxiv.org/pdf/1902.04103.pdf eqn 3
cp, cn = smooth_BCE(eps=0.0)
# Focal loss
g = h['fl_gamma'] # focal loss gamma
if g > 0:
BCEcls, BCEobj = FocalLoss(BCEcls, g), FocalLoss(BCEobj, g)
# Losses
nt = 0 # number of targets
no = len(p) # number of outputs
balance = [4.0, 1.0, 0.4] if no == 3 else [4.0, 1.0, 0.4, 0.1] # P3-5 or P3-6
for i, pi in enumerate(p): # layer index, layer predictions
b, a, gj, gi = indices[i] # image, anchor, gridy, gridx
tobj = torch.zeros_like(pi[..., 0], device=device) # target obj
n = b.shape[0] # number of targets
if n:
nt += n # cumulative targets
l_ = [b, a, gj, gi]
ps = pi[b, a, gj, gi] # prediction subset corresponding to targets
# Regression
pxy = ps[:, :2].sigmoid() * 2. - 0.5
pwh = (ps[:, 2:4].sigmoid() * 2) ** 2 * anchors[i]
pbox = torch.cat((pxy, pwh), 1) # predicted box
iou = bbox_iouf(pbox.T, tbox[i], x1y1x2y2=False, CIoU=True) # iou(prediction, target)
lbox += (1.0 - iou).mean() # iou loss
# Objectness
# 获取target所对应的obj,网格中存在gt目标的会被标记为iou与gt的交并比 (这个obj相当于前景和背景的关系)
tobj[b, a, gj, gi] = (1.0 - model.gr) + model.gr * iou.detach().clamp(0).type(tobj.dtype) # iou ratio
# Classification
if model.nc > 1: # cls loss (only if multiple classes)
t = torch.full_like(ps[:, 15:], cn, device=device) # targets
t[range(n), tcls[i]] = cp
lcls += BCEcls(ps[:, 15:], t) # BCE
# Append targets to text file
# with open('targets.txt', 'a') as file:
# [file.write('%11.5g ' * 4 % tuple(x) + '\n') for x in torch.cat((txy[i], twh[i]), 1)]
# landmarks loss
# plandmarks = ps[:,5:15].sigmoid() * 8. - 4.
plandmarks = ps[:, 5:15]
plandmarks[:, 0:2] = plandmarks[:, 0:2] * anchors[i]
plandmarks[:, 2:4] = plandmarks[:, 2:4] * anchors[i]
plandmarks[:, 4:6] = plandmarks[:, 4:6] * anchors[i]
plandmarks[:, 6:8] = plandmarks[:, 6:8] * anchors[i]
plandmarks[:, 8:10] = plandmarks[:, 8:10] * anchors[i]
lmark += landmarks_loss(plandmarks, tlandmarks[i], lmks_mask[i])
lobj += BCEobj(pi[..., 4], tobj) * balance[i] # obj loss
s = 3 / no # output count scaling
lbox *= h['box'] * s
lobj *= h['obj'] * s * (1.4 if no == 4 else 1.)
lcls *= h['cls'] * s
lmark *= h['landmark'] * s
bs = tobj.shape[0] # batch size
loss = lbox + lobj + lcls + lmark
return loss * bs, torch.cat((lbox, lobj, lcls, lmark, loss)).detach()
def build_targets(p, targets, model):
# Build targets for compute_loss(), input targets(image,class,x,y,w,h)
det = model.module.model[-1] if is_parallel(model) else model.model[-1] # Detect() module
na, nt = det.na, targets.shape[0] # number of anchors, targets
tcls, tbox, indices, anch, landmarks, lmks_mask = [], [], [], [], [], []
# gain = torch.ones(7, device=targets.device) # normalized to gridspace gain
gain = torch.ones(17, device=targets.device)
ai = torch.arange(na, device=targets.device).float().view(na, 1).repeat(1, nt) # same as .repeat_interleave(nt)
targets = torch.cat((targets.repeat(na, 1, 1), ai[:, :, None]), 2) # append anchor indices
g = 0.5 # bias
off = torch.tensor([[0, 0],
[1, 0], [0, 1], [-1, 0], [0, -1], # j,k,l,m
# [1, 1], [1, -1], [-1, 1], [-1, -1], # jk,jm,lk,lm
], device=targets.device).float() * g # offsets
for i in range(det.nl):
anchors = det.anchors[i]
gain[2:6] = torch.tensor(p[i].shape)[[3, 2, 3, 2]] # xyxy gain
# landmarks 10
gain[6:16] = torch.tensor(p[i].shape)[[3, 2, 3, 2, 3, 2, 3, 2, 3, 2]] # xyxy gain
# Match targets to anchors
t = targets * gain
if nt:
# Matches
r = t[:, :, 4:6] / anchors[:, None] # wh ratio
j = torch.max(r, 1. / r).max(2)[0] < model.hyp['anchor_t'] # compare
# j = wh_iou(anchors, t[:, 4:6]) > model.hyp['iou_t'] # iou(3,n)=wh_iou(anchors(3,2), gwh(n,2))
t = t[j] # filter
# Offsets
gxy = t[:, 2:4] # grid xy
gxi = gain[[2, 3]] - gxy # inverse
j, k = ((gxy % 1. < g) & (gxy > 1.)).T
l, m = ((gxi % 1. < g) & (gxi > 1.)).T
j = torch.stack((torch.ones_like(j), j, k, l, m))
t = t.repeat((5, 1, 1))[j]
offsets = (torch.zeros_like(gxy)[None] + off[:, None])[j]
else:
t = targets[0]
offsets = 0
# Define
b, c = t[:, :2].long().T # image, class
gxy = t[:, 2:4] # grid xy
gwh = t[:, 4:6] # grid wh
gij = (gxy - offsets).long()
gi, gj = gij.T # grid xy indices
# Append
a = t[:, 16].long() # anchor indices
indices.append((b, a, gj.clamp_(0, gain[3] - 1), gi.clamp_(0, gain[2] - 1))) # image, anchor, grid indices
tbox.append(torch.cat((gxy - gij, gwh), 1)) # box
anch.append(anchors[a]) # anchors
tcls.append(c) # class
# landmarks
lks = t[:, 6:16]
# lks_mask = lks > 0
# lks_mask = lks_mask.float()
lks_mask = torch.where(lks < 0, torch.full_like(lks, 0.), torch.full_like(lks, 1.0))
# 应该是关键点的坐标除以anch的宽高才对，便于模型学习。使用gwh会导致不同关键点的编码不同，没有统一的参考标准
lks[:, [0, 1]] = (lks[:, [0, 1]] - gij)
lks[:, [2, 3]] = (lks[:, [2, 3]] - gij)
lks[:, [4, 5]] = (lks[:, [4, 5]] - gij)
lks[:, [6, 7]] = (lks[:, [6, 7]] - gij)
lks[:, [8, 9]] = (lks[:, [8, 9]] - gij)
lks_mask_new = lks_mask
lmks_mask.append(lks_mask_new)
landmarks.append(lks)
# print('lks: ', lks.size())
return tcls, tbox, indices, anch, landmarks, lmks_mask
utils/landmark/metrics.py
# YOLOv5 by Ultralytics, GPL-3.0 license
"""
Model validation metrics
"""
import math
from pathlib import Path
import numpy as np
import torch
from ..metrics import plot_pr_curve, compute_ap
def ap_per_classf(tp, conf, pred_cls, target_cls, plot=False, save_dir='precision-recall_curve.png', names=(),
prefix=""):
""" Compute the average precision, given the recall and precision curves.
Source: https://github.com/rafaelpadilla/Object-Detection-Metrics.
# Arguments
tp: True positives (nparray, nx1 or nx10).
conf: Objectness value from 0-1 (nparray).
pred_cls: Predicted object classes (nparray).
target_cls: True object classes (nparray).
plot: Plot precision-recall curve at [email protected]
save_dir: Plot save directory
# Returns
The average precision as computed in py-faster-rcnn.
"""
# Sort by objectness
i = np.argsort(-conf)
tp, conf, pred_cls = tp[i], conf[i], pred_cls[i]
# Find unique classes
unique_classes = np.unique(target_cls)
# Create Precision-Recall curve and compute AP for each class
px, py = np.linspace(0, 1, 1000), [] # for plotting
pr_score = 0.1 # score to evaluate P and R https://github.com/ultralytics/yolov3/issues/898
s = [unique_classes.shape[0], tp.shape[1]] # number class, number iou thresholds (i.e. 10 for mAP0.5...0.95)
ap, p, r = np.zeros(s), np.zeros(s), np.zeros(s)
for ci, c in enumerate(unique_classes):
i = pred_cls == c
n_l = (target_cls == c).sum() # number of labels
n_p = i.sum() # number of predictions
if n_p == 0 or n_l == 0:
continue
else:
# Accumulate FPs and TPs
fpc = (1 - tp[i]).cumsum(0)
tpc = tp[i].cumsum(0)
# Recall
recall = tpc / (n_l + 1e-16) # recall curve
r[ci] = np.interp(-pr_score, -conf[i], recall[:, 0]) # r at pr_score, negative x, xp because xp decreases
# Precision
precision = tpc / (tpc + fpc) # precision curve
p[ci] = np.interp(-pr_score, -conf[i], precision[:, 0]) # p at pr_score
# AP from recall-precision curve
for j in range(tp.shape[1]):
ap[ci, j], mpre, mrec = compute_ap(recall[:, j], precision[:, j])
if plot and (j == 0):
py.append(np.interp(px, mrec, mpre)) # precision at [email protected]
# Compute F1 score (harmonic mean of precision and recall)
f1 = 2 * p * r / (p + r + 1e-16)
if plot:
# plot_pr_curve(px, py, ap, save_dir, names)
plot_pr_curve(px, py, ap, Path(save_dir) / f'{prefix}PR_curve.png', names)
return p, r, ap, f1, unique_classes.astype('int32')
def bbox_iouf(box1, box2, x1y1x2y2=True, GIoU=False, DIoU=False, CIoU=False, eps=1e-9):
# Returns the IoU of box1 to box2. box1 is 4, box2 is nx4
box2 = box2.T
# Get the coordinates of bounding boxes
if x1y1x2y2: # x1, y1, x2, y2 = box1
b1_x1, b1_y1, b1_x2, b1_y2 = box1[0], box1[1], box1[2], box1[3]
b2_x1, b2_y1, b2_x2, b2_y2 = box2[0], box2[1], box2[2], box2[3]
else: # transform from xywh to xyxy
b1_x1, b1_x2 = box1[0] - box1[2] / 2, box1[0] + box1[2] / 2
b1_y1, b1_y2 = box1[1] - box1[3] / 2, box1[1] + box1[3] / 2
b2_x1, b2_x2 = box2[0] - box2[2] / 2, box2[0] + box2[2] / 2
b2_y1, b2_y2 = box2[1] - box2[3] / 2, box2[1] + box2[3] / 2
# Intersection area
inter = (torch.min(b1_x2, b2_x2) - torch.max(b1_x1, b2_x1)).clamp(0) * \
(torch.min(b1_y2, b2_y2) - torch.max(b1_y1, b2_y1)).clamp(0)
# Union Area
w1, h1 = b1_x2 - b1_x1, b1_y2 - b1_y1 + eps
w2, h2 = b2_x2 - b2_x1, b2_y2 - b2_y1 + eps
union = w1 * h1 + w2 * h2 - inter + eps
iou = inter / union
if GIoU or DIoU or CIoU:
# convex (smallest enclosing box) width
cw = torch.max(b1_x2, b2_x2) - torch.min(b1_x1, b2_x1)
ch = torch.max(b1_y2, b2_y2) - torch.min(b1_y1, b2_y1) # convex height
if CIoU or DIoU: # Distance or Complete IoU https://arxiv.org/abs/1911.08287v1
c2 = cw ** 2 + ch ** 2 + eps # convex diagonal squared
rho2 = ((b2_x1 + b2_x2 - b1_x1 - b1_x2) ** 2 +
(b2_y1 + b2_y2 - b1_y1 - b1_y2) ** 2) / 4 # center distance squared
if DIoU:
return iou - rho2 / c2 # DIoU
elif CIoU: # https://github.com/Zzh-tju/DIoU-SSD-pytorch/blob/master/utils/box/box_utils.py#L47
v = (4 / math.pi ** 2) * \
torch.pow(torch.atan(w2 / h2) - torch.atan(w1 / h1), 2)
with torch.no_grad():
alpha = v / ((1 + eps) - iou + v)
return iou - (rho2 / c2 + v * alpha) # CIoU
else: # GIoU https://arxiv.org/pdf/1902.09630.pdf
c_area = cw * ch + eps # convex area
return iou - (c_area - union) / c_area # GIoU
else:
return iou # IoU
utils/landmark/plots.py
# YOLOv5 by Ultralytics, GPL-3.0 license
"""
Plotting utils
"""
import math
import os
import random
from pathlib import Path
import cv2
import matplotlib
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import seaborn as sn
import torch
import yaml
from PIL import Image
from ..metrics import fitness
from .. import threaded
from ..general import xywh2xyxy,xyxy2xywh
from ..plots import Annotator, colors,hist2d
# add for landmark
def plot_evolution(yaml_file='data/hyp.finetune.yaml'): # from utils.plots import *; plot_evolution()
# Plot hyperparameter evolution results in evolve.txt
with open(yaml_file) as f:
hyp = yaml.load(f, Loader=yaml.SafeLoader)
x = np.loadtxt('evolve.txt', ndmin=2)
f = fitness(x)
# weights = (f - f.min()) ** 2 # for weighted results
plt.figure(figsize=(10, 12), tight_layout=True)
matplotlib.rc('font', **{'size': 8})
for i, (k, v) in enumerate(hyp.items()):
y = x[:, i + 7]
# mu = (y * weights).sum() / weights.sum() # best weighted result
mu = y[f.argmax()] # best single result
plt.subplot(6, 5, i + 1)
plt.scatter(y, f, c=hist2d(y, f, 20), cmap='viridis', alpha=.8, edgecolors='none')
plt.plot(mu, f.max(), 'k+', markersize=15)
plt.title('%s = %.3g' % (k, mu), fontdict={'size': 9}) # limit to 40 characters
if i % 5 != 0:
plt.yticks([])
print('%15s: %.3g' % (k, mu))
plt.savefig('evolve.png', dpi=200)
print('\nPlot saved as evolve.png')
def plot_study_txt(path='study/', x=None): # from utils.plots import *; plot_study_txt()
# Plot study.txt generated by test.py
fig, ax = plt.subplots(2, 4, figsize=(10, 6), tight_layout=True)
ax = ax.ravel()
fig2, ax2 = plt.subplots(1, 1, figsize=(8, 4), tight_layout=True)
for f in [Path(path) / f'study_coco_{x}.txt' for x in ['yolov5s', 'yolov5m', 'yolov5l', 'yolov5x']]:
y = np.loadtxt(f, dtype=np.float32, usecols=[0, 1, 2, 3, 7, 8, 9], ndmin=2).T
x = np.arange(y.shape[1]) if x is None else np.array(x)
s = ['P', 'R', '[email protected]', '[email protected]:.95', 't_inference (ms/img)', 't_NMS (ms/img)', 't_total (ms/img)']
for i in range(7):
ax[i].plot(x, y[i], '.-', linewidth=2, markersize=8)
ax[i].set_title(s[i])
j = y[3].argmax() + 1
ax2.plot(y[6, :j], y[3, :j] * 1E2, '.-', linewidth=2, markersize=8,
label=f.stem.replace('study_coco_', '').replace('yolo', 'YOLO'))
ax2.plot(1E3 / np.array([209, 140, 97, 58, 35, 18]), [34.6, 40.5, 43.0, 47.5, 49.7, 51.5],
'k.-', linewidth=2, markersize=8, alpha=.25, label='EfficientDet')
ax2.grid()
ax2.set_yticks(np.arange(30, 60, 5))
ax2.set_xlim(0, 30)
ax2.set_ylim(29, 51)
ax2.set_xlabel('GPU Speed (ms/img)')
ax2.set_ylabel('COCO AP val')
ax2.legend(loc='lower right')
plt.savefig('test_study.png', dpi=300)
def plot_resultsf(start=0, stop=0, bucket='', id=(), labels=(), save_dir=''):
# Plot training 'results*.txt'. from utils.plots import *; plot_results(save_dir='runs/train/exp')
fig, ax = plt.subplots(2, 5, figsize=(12, 6), tight_layout=True)
ax = ax.ravel()
s = ['Box', 'Objectness', 'Classification', 'Precision', 'Recall',
'val Box', 'val Objectness', 'val Classification', '[email protected]', '[email protected]:0.95']
if bucket:
# files = ['https://storage.googleapis.com/%s/results%g.txt' % (bucket, x) for x in id]
files = ['results%g.txt' % x for x in id]
c = ('gsutil cp ' + '%s ' * len(files) + '.') % tuple('gs://%s/results%g.txt' % (bucket, x) for x in id)
os.system(c)
else:
files = list(Path(save_dir).glob('results*.txt'))
assert len(files), 'No results.txt files found in %s, nothing to plot.' % os.path.abspath(save_dir)
for fi, f in enumerate(files):
try:
results = np.loadtxt(f, usecols=[2, 3, 4, 8, 9, 12, 13, 14, 10, 11], ndmin=2).T
n = results.shape[1] # number of rows
x = range(start, min(stop, n) if stop else n)
for i in range(10):
y = results[i, x]
if i in [0, 1, 2, 5, 6, 7]:
y[y == 0] = np.nan # don't show zero loss values
# y /= y[0] # normalize
label = labels[fi] if len(labels) else f.stem
ax[i].plot(x, y, marker='.', label=label, linewidth=2, markersize=8)
ax[i].set_title(s[i])
# if i in [5, 6, 7]: # share train and val loss y axes
# ax[i].get_shared_y_axes().join(ax[i], ax[i - 5])
except Exception as e:
print('Warning: Plotting error for %s; %s' % (f, e))
ax[1].legend()
fig.savefig(Path(save_dir) / 'results.png', dpi=200)
def plot_one_box(x, img, color=None, label=None, line_thickness=None):
# Plots one bounding box on image img
tl = line_thickness or round(0.002 * (img.shape[0] + img.shape[1]) / 2) + 1 # line/font thickness
color = color or [random.randint(0, 255) for _ in range(3)]
c1, c2 = (int(x[0]), int(x[1])), (int(x[2]), int(x[3]))
cv2.rectangle(img, c1, c2, color, thickness=tl, lineType=cv2.LINE_AA)
if label:
tf = max(tl - 1, 1) # font thickness
t_size = cv2.getTextSize(label, 0, fontScale=tl / 3, thickness=tf)[0]
c2 = c1[0] + t_size[0], c1[1] - t_size[1] - 3
cv2.rectangle(img, c1, c2, color, -1, cv2.LINE_AA) # filled
cv2.putText(img, label, (c1[0], c1[1] - 2), 0, tl / 3, [225, 255, 255], thickness=tf, lineType=cv2.LINE_AA)
def plot_labelsf(labels, save_dir=Path(''), loggers=None):
# plot dataset labels
print('Plotting labels... ')
c, b = labels[:, 0], labels[:, 1:5].transpose() # classes, boxes
nc = int(c.max() + 1) # number of classes
x = pd.DataFrame(b.transpose(), columns=['x', 'y', 'width', 'height'])
# seaborn correlogram
sn.pairplot(x, corner=True, diag_kind='auto', kind='hist', diag_kws=dict(bins=50), plot_kws=dict(pmax=0.9))
plt.savefig(save_dir / 'labels_correlogram.jpg', dpi=200)
plt.close()
# matplotlib labels
matplotlib.use('svg') # faster
ax = plt.subplots(2, 2, figsize=(8, 8), tight_layout=True)[1].ravel()
ax[0].hist(c, bins=np.linspace(0, nc, nc + 1) - 0.5, rwidth=0.8)
ax[0].set_xlabel('classes')
sn.histplot(x, x='x', y='y', ax=ax[2], bins=50, pmax=0.9)
sn.histplot(x, x='width', y='height', ax=ax[3], bins=50, pmax=0.9)
# rectangles
labels[:, 1:3] = 0.5 # center
labels[:, 1:] = xywh2xyxy(labels[:, 1:]) * 2000
img = Image.fromarray(np.ones((2000, 2000, 3), dtype=np.uint8) * 255)
# for cls, *box in labels[:1000]:
# ImageDraw.Draw(img).rectangle(box, width=1, outline=colors[int(cls) % 10]) # plot
ax[1].imshow(img)
ax[1].axis('off')
for a in [0, 1, 2, 3]:
for s in ['top', 'right', 'left', 'bottom']:
ax[a].spines[s].set_visible(False)
plt.savefig(save_dir / 'labels.jpg', dpi=200)
matplotlib.use('Agg')
plt.close()
# loggers
for k, v in loggers.items() or {}:
if k == 'wandb' and v:
v.log({"Labels": [v.Image(str(x), caption=x.name) for x in save_dir.glob('*labels*.jpg')]})
def output_to_targetf(output):
# Convert model output to target format [batch_id, class_id, x, y, w, h, conf]
targets = []
for i, o in enumerate(output):
# for *box, conf, cls in o.cpu().numpy():
# targets.append([i, cls, *list(*xyxy2xywh(np.array(box)[None])), conf])
for item in o.cpu().numpy():
# (x, y, w, h), conf, *keypoints, cls
box = item[:4]
conf = item[4]
keypoints = item[5:15]
cls = item[15]
targets.append([i, cls, *list(*xyxy2xywh(np.array(box)[None])),*keypoints, conf])
return np.array(targets)
@threaded
def plot_imagesf(images, targets, paths=None, fname='images.jpg', names=None):
# Plot image grid with labels
if isinstance(images, torch.Tensor):
images = images.cpu().float().numpy()
if isinstance(targets, torch.Tensor):
targets = targets.cpu().numpy()
max_size = 1920 # max image size
max_subplots = 16 # max image subplots, i.e. 4x4
bs, _, h, w = images.shape # batch size, _, height, width
bs = min(bs, max_subplots) # limit plot images
ns = np.ceil(bs ** 0.5) # number of subplots (square)
if np.max(images[0]) <= 1:
images *= 255 # de-normalise (optional)
# Build Image
mosaic = np.full((int(ns * h), int(ns * w), 3), 255, dtype=np.uint8) # init
for i, im in enumerate(images):
if i == max_subplots: # if last batch has fewer images than we expect
break
x, y = int(w * (i // ns)), int(h * (i % ns)) # block origin
im = im.transpose(1, 2, 0)
mosaic[y:y + h, x:x + w, :] = im
# Resize (optional)
scale = max_size / ns / max(h, w)
if scale < 1:
h = math.ceil(scale * h)
w = math.ceil(scale * w)
mosaic = cv2.resize(mosaic, tuple(int(x * ns) for x in (w, h)))
# Annotate
fs = int((h + w) * ns * 0.01) # font size
annotator = Annotator(mosaic, line_width=round(fs / 10), font_size=fs, pil=True, example=names)
for i in range(i + 1):
x, y = int(w * (i // ns)), int(h * (i % ns)) # block origin
annotator.rectangle([x, y, x + w, y + h], None, (255, 255, 255), width=2) # borders
if paths:
annotator.text((x + 5, y + 5), text=Path(paths[i]).name[:40], txt_color=(220, 220, 220)) # filenames
if len(targets) > 0:
ti = targets[targets[:, 0] == i] # image targets
boxes = xywh2xyxy(ti[:, 2:6]).T
classes = ti[:, 1].astype('int')
labels = ti.shape[1] == 16 # labels if no conf column
conf = None if labels else ti[:, 16] # check for confidence presence (label vs pred)
keypoints = ti[:, 6:16].T
if boxes.shape[1]:
if boxes.max() <= 1.01: # if normalized with tolerance 0.01
boxes[[0, 2]] *= w # scale to pixels
boxes[[1, 3]] *= h
elif scale < 1: # absolute coords need scale if image scales
boxes *= scale
boxes[[0, 2]] += x
boxes[[1, 3]] += y
if keypoints.shape[1]:
if keypoints.max() <= 1.01: # if normalized with tolerance 0.01
keypoints[[0,2,4,6,8]] *= w # scale to pixels
keypoints[[1,3,5,7,9]] *= h
elif scale < 1: # absolute coords need scale if image scales
keypoints *= scale
keypoints[[0,2,4,6,8]] += x
keypoints[[1,3,5,7,9]] += y
for j, (box,kpy) in enumerate(zip(boxes.T.tolist(),keypoints.T.tolist())):
cls = classes[j]
color = colors(cls)
cls = names[cls] if names else cls
if labels or conf[j] > 0.25: # 0.25 conf thresh
label = f'{cls}' if labels else f'{cls} {conf[j]:.1f}'
annotator.box_label(box, label, color=color)
annotator.landmark(kpy)
annotator.im.save(fname) # saveutils/plots.py还有个函数需要修改
class Annotator:
# YOLOv5 Annotator for train/val mosaics and jpgs and detect/hub inference annotations
def __init__(self, im, line_width=None, font_size=None, font='Arial.ttf', pil=False, example='abc'):
assert im.data.contiguous, 'Image not contiguous. Apply np.ascontiguousarray(im) to Annotator() input images.'
non_ascii = not is_ascii(example) # non-latin labels, i.e. asian, arabic, cyrillic
self.pil = pil or non_ascii
if self.pil: # use PIL
self.im = im if isinstance(im, Image.Image) else Image.fromarray(im)
self.draw = ImageDraw.Draw(self.im)
self.font = check_pil_font(font='Arial.Unicode.ttf' if non_ascii else font,
size=font_size or max(round(sum(self.im.size) / 2 * 0.035), 12))
else: # use cv2
self.im = im
self.lw = line_width or max(round(sum(im.shape) / 2 * 0.003), 2) # line width
def box_label(self, box, label='', color=(128, 128, 128), txt_color=(255, 255, 255)):
# Add one xyxy box to image with label
if self.pil or not is_ascii(label):
self.draw.rectangle(box, width=self.lw, outline=color) # box
if label:
w, h = self.font.getsize(label) # text width, height
outside = box[1] - h >= 0 # label fits outside box
self.draw.rectangle(
(box[0], box[1] - h if outside else box[1], box[0] + w + 1,
box[1] + 1 if outside else box[1] + h + 1),
fill=color,
)
# self.draw.text((box[0], box[1]), label, fill=txt_color, font=self.font, anchor='ls') # for PIL>8.0
self.draw.text((box[0], box[1] - h if outside else box[1]), label, fill=txt_color, font=self.font)
else: # cv2
p1, p2 = (int(box[0]), int(box[1])), (int(box[2]), int(box[3]))
cv2.rectangle(self.im, p1, p2, color, thickness=self.lw, lineType=cv2.LINE_AA)
if label:
tf = max(self.lw - 1, 1) # font thickness
w, h = cv2.getTextSize(label, 0, fontScale=self.lw / 3, thickness=tf)[0] # text width, height
outside = p1[1] - h >= 3
p2 = p1[0] + w, p1[1] - h - 3 if outside else p1[1] + h + 3
cv2.rectangle(self.im, p1, p2, color, -1, cv2.LINE_AA) # filled
cv2.putText(self.im,
label, (p1[0], p1[1] - 2 if outside else p1[1] + h + 2),
0,
self.lw / 3,
txt_color,
thickness=tf,
lineType=cv2.LINE_AA)
# 新增landmark
def landmark(self,landmarks):
clors = [(255, 0, 0), (0, 255, 0), (0, 0, 255), (255, 255, 0), (0, 255, 255)]
if self.pil:
for i in range(len(landmarks)//2):
point_x = int(landmarks[2 * i])
point_y = int(landmarks[2 * i + 1])
self.draw.point((point_x,point_y),clors[i])
else: # cv2
h, w, c = self.im.shape
tl = 1 or round(0.002 * (h + w) / 2) + 1 # line/font thickness
for i in range(5):
point_x = int(landmarks[2 * i])
point_y = int(landmarks[2 * i + 1])
cv2.circle(self.im, (point_x, point_y), tl + 1, clors[i], -1)
def masks(self, masks, colors, im_gpu=None, alpha=0.5):
"""Plot masks at once.
Args:
masks (tensor): predicted masks on cuda, shape: [n, h, w]
colors (List[List[Int]]): colors for predicted masks, [[r, g, b] * n]
im_gpu (tensor): img is in cuda, shape: [3, h, w], range: [0, 1]
alpha (float): mask transparency: 0.0 fully transparent, 1.0 opaque
"""
if self.pil:
# convert to numpy first
self.im = np.asarray(self.im).copy()
if im_gpu is None:
# Add multiple masks of shape(h,w,n) with colors list([r,g,b], [r,g,b], ...)
if len(masks) == 0:
return
if isinstance(masks, torch.Tensor):
masks = torch.as_tensor(masks, dtype=torch.uint8)
masks = masks.permute(1, 2, 0).contiguous()
masks = masks.cpu().numpy()
# masks = np.ascontiguousarray(masks.transpose(1, 2, 0))
masks = scale_image(masks.shape[:2], masks, self.im.shape)
masks = np.asarray(masks, dtype=np.float32)
colors = np.asarray(colors, dtype=np.float32) # shape(n,3)
s = masks.sum(2, keepdims=True).clip(0, 1) # add all masks together
masks = (masks @ colors).clip(0, 255) # (h,w,n) @ (n,3) = (h,w,3)
self.im[:] = masks * alpha + self.im * (1 - s * alpha)
else:
if len(masks) == 0:
self.im[:] = im_gpu.permute(1, 2, 0).contiguous().cpu().numpy() * 255
colors = torch.tensor(colors, device=im_gpu.device, dtype=torch.float32) / 255.0
colors = colors[:, None, None] # shape(n,1,1,3)
masks = masks.unsqueeze(3) # shape(n,h,w,1)
masks_color = masks * (colors * alpha) # shape(n,h,w,3)
inv_alph_masks = (1 - masks * alpha).cumprod(0) # shape(n,h,w,1)
mcs = (masks_color * inv_alph_masks).sum(0) * 2 # mask color summand shape(n,h,w,3)
im_gpu = im_gpu.flip(dims=[0]) # flip channel
im_gpu = im_gpu.permute(1, 2, 0).contiguous() # shape(h,w,3)
im_gpu = im_gpu * inv_alph_masks[-1] + mcs
im_mask = (im_gpu * 255).byte().cpu().numpy()
self.im[:] = scale_image(im_gpu.shape, im_mask, self.im.shape)
if self.pil:
# convert im back to PIL and update draw
self.fromarray(self.im)
def rectangle(self, xy, fill=None, outline=None, width=1):
# Add rectangle to image (PIL-only)
self.draw.rectangle(xy, fill, outline, width)
def text(self, xy, text, txt_color=(255, 255, 255), anchor='top'):
# Add text to image (PIL-only)
if anchor == 'bottom': # start y from font bottom
w, h = self.font.getsize(text) # text width, height
xy[1] += 1 - h
self.draw.text(xy, text, fill=txt_color, font=self.font)
def fromarray(self, im):
# Update self.im from a numpy array
self.im = im if isinstance(im, Image.Image) else Image.fromarray(im)
self.draw = ImageDraw.Draw(self.im)
def result(self):
# Return annotated image as array
return np.asarray(self.im)utils/landmark/dataloaders.py
import glob
import logging
import math
import os
import random
import shutil
from itertools import repeat
from multiprocessing.pool import ThreadPool
from pathlib import Path
import cv2
import numpy as np
import torch
import torch.nn.functional as F
from PIL import Image, ExifTags
from torch.utils.data import Dataset
from tqdm import tqdm
from ..general import xyxy2xywh, xywh2xyxy
from ..torch_utils import torch_distributed_zero_first
# Parameters
help_url = 'https://github.com/ultralytics/yolov5/wiki/Train-Custom-Data'
img_formats = ['bmp', 'jpg', 'jpeg', 'png', 'tif', 'tiff', 'dng'] # acceptable image suffixes
vid_formats = ['mov', 'avi', 'mp4', 'mpg', 'mpeg', 'm4v', 'wmv', 'mkv'] # acceptable video suffixes
logger = logging.getLogger(__name__)
# Get orientation exif tag
for orientation in ExifTags.TAGS.keys():
if ExifTags.TAGS[orientation] == 'Orientation':
break
def get_hash(files):
# Returns a single hash value of a list of files
return sum(os.path.getsize(f) for f in files if os.path.isfile(f))
def img2label_paths(img_paths):
# Define label paths as a function of image paths
sa, sb = os.sep + 'images' + os.sep, os.sep + 'labels' + os.sep # /images/, /labels/ substrings
return [x.replace(sa, sb, 1).replace('.' + x.split('.')[-1], '.txt') for x in img_paths]
def exif_size(img):
# Returns exif-corrected PIL size
s = img.size # (width, height)
try:
rotation = dict(img._getexif().items())[orientation]
if rotation == 6: # rotation 270
s = (s[1], s[0])
elif rotation == 8: # rotation 90
s = (s[1], s[0])
except:
pass
return s
def create_dataloader(path, imgsz, batch_size, stride, opt, hyp=None, augment=False, cache=False, pad=0.0, rect=False,
rank=-1, world_size=1, workers=8, image_weights=False, quad=False, prefix=''):
# Make sure only the first process in DDP process the dataset first, and the following others can use the cache
with torch_distributed_zero_first(rank):
dataset = LoadFaceImagesAndLabels(path, imgsz, batch_size,
augment=augment, # augment images
hyp=hyp, # augmentation hyperparameters
rect=rect, # rectangular training
cache_images=cache,
single_cls=opt.single_cls,
stride=int(stride),
pad=pad,
image_weights=image_weights,
prefix=prefix)
batch_size = min(batch_size, len(dataset))
nw = min([os.cpu_count() // world_size, batch_size if batch_size > 1 else 0, workers]) # number of workers
sampler = torch.utils.data.distributed.DistributedSampler(dataset) if rank != -1 else None
loader = torch.utils.data.DataLoader if image_weights else InfiniteDataLoader
# Use torch.utils.data.DataLoader() if dataset.properties will update during training else InfiniteDataLoader()
dataloader = loader(dataset,
batch_size=batch_size,
num_workers=nw,
sampler=sampler,
pin_memory=True,
collate_fn=LoadFaceImagesAndLabels.collate_fn4 if quad else LoadFaceImagesAndLabels.collate_fn)
return dataloader, dataset
class InfiniteDataLoader(torch.utils.data.dataloader.DataLoader):
""" Dataloader that reuses workers
Uses same syntax as vanilla DataLoader
"""
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
object.__setattr__(self, 'batch_sampler', _RepeatSampler(self.batch_sampler))
self.iterator = super().__iter__()
def __len__(self):
return len(self.batch_sampler.sampler)
def __iter__(self):
for i in range(len(self)):
yield next(self.iterator)
class _RepeatSampler(object):
""" Sampler that repeats forever
Args:
sampler (Sampler)
"""
def __init__(self, sampler):
self.sampler = sampler
def __iter__(self):
while True:
yield from iter(self.sampler)
class LoadFaceImagesAndLabels(Dataset): # for training/testing
def __init__(self, path, img_size=640, batch_size=16, augment=False, hyp=None, rect=False, image_weights=False,
cache_images=False, single_cls=False, stride=32, pad=0.0, prefix=''):
self.img_size = img_size
self.augment = augment
self.hyp = hyp
self.image_weights = image_weights
self.rect = False if image_weights else rect
self.mosaic = self.augment and not self.rect # load 4 images at a time into a mosaic (only during training)
self.mosaic_border = [-img_size // 2, -img_size // 2]
self.stride = stride
try:
f = [] # image files
for p in path if isinstance(path, list) else [path]:
p = Path(p) # os-agnostic
if p.is_dir(): # dir
f += glob.glob(str(p / '**' / '*.*'), recursive=True)
elif p.is_file(): # file
with open(p, 'r') as t:
t = t.read().strip().splitlines()
parent = str(p.parent) + os.sep
f += [x.replace('./', parent) if x.startswith('./') else x for x in t] # local to global path
else:
raise FileNotFoundError(f'{prefix}{p} does not exist')
self.img_files = sorted([x.replace('/', os.sep) for x in f if x.split('.')[-1].lower() in img_formats])
assert self.img_files, f'{prefix}No images found'
except Exception as e:
raise Exception(f'{prefix}Error loading data from {path}: {e}\n{help_url}')
# Check cache
self.label_files = img2label_paths(self.img_files) # labels
cache_path = Path(self.label_files[0]).parent.with_suffix('.cache') # cached labels
if cache_path.is_file():
cache = torch.load(cache_path) # load
if cache['hash'] != get_hash(self.label_files + self.img_files) or 'results' not in cache: # changed
cache = self.cache_labels(cache_path) # re-cache
else:
cache = self.cache_labels(cache_path) # cache
# Display cache
[nf, nm, ne, nc, n] = cache.pop('results') # found, missing, empty, corrupted, total
desc = f"Scanning '{cache_path}' for images and labels... {nf} found, {nm} missing, {ne} empty, {nc} corrupted"
tqdm(None, desc=prefix + desc, total=n, initial=n)
assert nf > 0 or not augment, f'{prefix}No labels found in {cache_path}, can not start training. {help_url}'
# Read cache
cache.pop('hash') # remove hash
labels, shapes = zip(*cache.values())
self.labels = list(labels)
self.shapes = np.array(shapes, dtype=np.float64)
self.img_files = list(cache.keys()) # update
self.label_files = img2label_paths(cache.keys()) # update
if single_cls:
for x in self.labels:
x[:, 0] = 0
n = len(shapes) # number of images
bi = np.floor(np.arange(n) / batch_size).astype(np.int) # batch index
nb = bi[-1] + 1 # number of batches
self.batch = bi # batch index of image
self.n = n
self.indices = range(n)
# Rectangular Training
if self.rect:
# Sort by aspect ratio
s = self.shapes # wh
ar = s[:, 1] / s[:, 0] # aspect ratio
irect = ar.argsort()
self.img_files = [self.img_files[i] for i in irect]
self.label_files = [self.label_files[i] for i in irect]
self.labels = [self.labels[i] for i in irect]
self.shapes = s[irect] # wh
ar = ar[irect]
# Set training image shapes
shapes = [[1, 1]] * nb
for i in range(nb):
ari = ar[bi == i]
mini, maxi = ari.min(), ari.max()
if maxi < 1:
shapes[i] = [maxi, 1]
elif mini > 1:
shapes[i] = [1, 1 / mini]
self.batch_shapes = np.ceil(np.array(shapes) * img_size / stride + pad).astype(np.int) * stride
# Cache images into memory for faster training (WARNING: large datasets may exceed system RAM)
self.imgs = [None] * n
if cache_images:
gb = 0 # Gigabytes of cached images
self.img_hw0, self.img_hw = [None] * n, [None] * n
results = ThreadPool(8).imap(lambda x: load_image(*x), zip(repeat(self), range(n))) # 8 threads
pbar = tqdm(enumerate(results), total=n)
for i, x in pbar:
self.imgs[i], self.img_hw0[i], self.img_hw[i] = x # img, hw_original, hw_resized = load_image(self, i)
gb += self.imgs[i].nbytes
pbar.desc = f'{prefix}Caching images ({gb / 1E9:.1f}GB)'
def cache_labels(self, path=Path('./labels.cache'),prefix=''):
# Cache dataset labels, check images and read shapes
x = {} # dict
nm, nf, ne, nc = 0, 0, 0, 0 # number missing, found, empty, duplicate
desc = f"{prefix}Scanning '{path.parent / path.stem}' images and labels..."
pbar = tqdm(zip(self.img_files, self.label_files), desc=desc, total=len(self.img_files))
for i, (im_file, lb_file) in enumerate(pbar):
try:
# verify images
im = Image.open(im_file)
im.verify() # PIL verify
shape = exif_size(im) # image size
assert (shape[0] > 9) & (shape[1] > 9), 'image size <10 pixels'
# verify labels
if os.path.isfile(lb_file):
nf += 1 # label found
with open(lb_file, 'r') as f:
l = np.array([x.split() for x in f.read().strip().splitlines()], dtype=np.float32) # labels
if len(l):
assert l.shape[1] == 15, 'labels require 15 columns each'
assert (l >= -1).all(), 'negative labels'
assert (l[:, 1:] <= 1).all(), 'non-normalized or out of bounds coordinate labels'
assert np.unique(l, axis=0).shape[0] == l.shape[0], 'duplicate labels'
else:
ne += 1 # label empty
l = np.zeros((0, 15), dtype=np.float32)
else:
nm += 1 # label missing
l = np.zeros((0, 15), dtype=np.float32)
x[im_file] = [l, shape]
except Exception as e:
nc += 1
print(f'{prefix}WARNING: Ignoring corrupted image and/or label %s: %s' % (im_file, e))
pbar.desc = f"{prefix}Scanning '{path.parent / path.stem}' for images and labels... " \
f"{nf} found, {nm} missing, {ne} empty, {nc} corrupted"
if nf == 0:
print(f'{prefix}WARNING: No labels found in {path}. See {help_url}')
x['hash'] = get_hash(self.label_files + self.img_files)
x['results'] = [nf, nm, ne, nc, i + 1]
torch.save(x, path) # save for next time
logging.info(f"{prefix}New cache created: {path}")
return x
def __len__(self):
return len(self.img_files)
# def __iter__(self):
# self.count = -1
# print('ran dataset iter')
# #self.shuffled_vector = np.random.permutation(self.nF) if self.augment else np.arange(self.nF)
# return self
def __getitem__(self, index):
index = self.indices[index] # linear, shuffled, or image_weights
hyp = self.hyp
mosaic = self.mosaic and random.random() < hyp['mosaic']
if mosaic:
# Load mosaic
img, labels = load_mosaic_face(self, index)
shapes = None
# MixUp https://arxiv.org/pdf/1710.09412.pdf
if random.random() < hyp['mixup']:
img2, labels2 = load_mosaic_face(self, random.randint(0, self.n - 1))
r = np.random.beta(8.0, 8.0) # mixup ratio, alpha=beta=8.0
img = (img * r + img2 * (1 - r)).astype(np.uint8)
labels = np.concatenate((labels, labels2), 0)
else:
# Load image
img, (h0, w0), (h, w) = load_image(self, index)
# Letterbox
shape = self.batch_shapes[self.batch[index]] if self.rect else self.img_size # final letterboxed shape
img, ratio, pad = letterbox(img, shape, auto=False, scaleup=self.augment)
shapes = (h0, w0), ((h / h0, w / w0), pad) # for COCO mAP rescaling
# Load labels
labels = []
x = self.labels[index]
if x.size > 0:
# Normalized xywh to pixel xyxy format
labels = x.copy()
labels[:, 1] = ratio[0] * w * (x[:, 1] - x[:, 3] / 2) + pad[0] # pad width
labels[:, 2] = ratio[1] * h * (x[:, 2] - x[:, 4] / 2) + pad[1] # pad height
labels[:, 3] = ratio[0] * w * (x[:, 1] + x[:, 3] / 2) + pad[0]
labels[:, 4] = ratio[1] * h * (x[:, 2] + x[:, 4] / 2) + pad[1]
# labels[:, 5] = ratio[0] * w * x[:, 5] + pad[0] # pad width
labels[:, 5] = np.array(x[:, 5] > 0, dtype=np.int32) * (ratio[0] * w * x[:, 5] + pad[0]) + (
np.array(x[:, 5] > 0, dtype=np.int32) - 1)
labels[:, 6] = np.array(x[:, 6] > 0, dtype=np.int32) * (ratio[1] * h * x[:, 6] + pad[1]) + (
np.array(x[:, 6] > 0, dtype=np.int32) - 1)
labels[:, 7] = np.array(x[:, 7] > 0, dtype=np.int32) * (ratio[0] * w * x[:, 7] + pad[0]) + (
np.array(x[:, 7] > 0, dtype=np.int32) - 1)
labels[:, 8] = np.array(x[:, 8] > 0, dtype=np.int32) * (ratio[1] * h * x[:, 8] + pad[1]) + (
np.array(x[:, 8] > 0, dtype=np.int32) - 1)
labels[:, 9] = np.array(x[:, 5] > 0, dtype=np.int32) * (ratio[0] * w * x[:, 9] + pad[0]) + (
np.array(x[:, 9] > 0, dtype=np.int32) - 1)
labels[:, 10] = np.array(x[:, 5] > 0, dtype=np.int32) * (ratio[1] * h * x[:, 10] + pad[1]) + (
np.array(x[:, 10] > 0, dtype=np.int32) - 1)
labels[:, 11] = np.array(x[:, 11] > 0, dtype=np.int32) * (ratio[0] * w * x[:, 11] + pad[0]) + (
np.array(x[:, 11] > 0, dtype=np.int32) - 1)
labels[:, 12] = np.array(x[:, 12] > 0, dtype=np.int32) * (ratio[1] * h * x[:, 12] + pad[1]) + (
np.array(x[:, 12] > 0, dtype=np.int32) - 1)
labels[:, 13] = np.array(x[:, 13] > 0, dtype=np.int32) * (ratio[0] * w * x[:, 13] + pad[0]) + (
np.array(x[:, 13] > 0, dtype=np.int32) - 1)
labels[:, 14] = np.array(x[:, 14] > 0, dtype=np.int32) * (ratio[1] * h * x[:, 14] + pad[1]) + (
np.array(x[:, 14] > 0, dtype=np.int32) - 1)
if self.augment:
# Augment imagespace
if not mosaic:
img, labels = random_perspective(img, labels,
degrees=hyp['degrees'],
translate=hyp['translate'],
scale=hyp['scale'],
shear=hyp['shear'],
perspective=hyp['perspective'])
# Augment colorspace
augment_hsv(img, hgain=hyp['hsv_h'], sgain=hyp['hsv_s'], vgain=hyp['hsv_v'])
# Apply cutouts
# if random.random() < 0.9:
# labels = cutout(img, labels)
nL = len(labels) # number of labels
if nL:
labels[:, 1:5] = xyxy2xywh(labels[:, 1:5]) # convert xyxy to xywh
labels[:, [2, 4]] /= img.shape[0] # normalized height 0-1
labels[:, [1, 3]] /= img.shape[1] # normalized width 0-1
labels[:, [5, 7, 9, 11, 13]] /= img.shape[1] # normalized landmark x 0-1
labels[:, [5, 7, 9, 11, 13]] = np.where(labels[:, [5, 7, 9, 11, 13]] < 0, -1, labels[:, [5, 7, 9, 11, 13]])
labels[:, [6, 8, 10, 12, 14]] /= img.shape[0] # normalized landmark y 0-1
labels[:, [6, 8, 10, 12, 14]] = np.where(labels[:, [6, 8, 10, 12, 14]] < 0, -1,
labels[:, [6, 8, 10, 12, 14]])
if self.augment:
# flip up-down
if random.random() < hyp['flipud']:
img = np.flipud(img)
if nL:
labels[:, 2] = 1 - labels[:, 2]
labels[:, 6] = np.where(labels[:, 6] < 0, -1, 1 - labels[:, 6])
labels[:, 8] = np.where(labels[:, 8] < 0, -1, 1 - labels[:, 8])
labels[:, 10] = np.where(labels[:, 10] < 0, -1, 1 - labels[:, 10])
labels[:, 12] = np.where(labels[:, 12] < 0, -1, 1 - labels[:, 12])
labels[:, 14] = np.where(labels[:, 14] < 0, -1, 1 - labels[:, 14])
# flip left-right
if random.random() < hyp['fliplr']:
img = np.fliplr(img)
if nL:
labels[:, 1] = 1 - labels[:, 1]
labels[:, 5] = np.where(labels[:, 5] < 0, -1, 1 - labels[:, 5])
labels[:, 7] = np.where(labels[:, 7] < 0, -1, 1 - labels[:, 7])
labels[:, 9] = np.where(labels[:, 9] < 0, -1, 1 - labels[:, 9])
labels[:, 11] = np.where(labels[:, 11] < 0, -1, 1 - labels[:, 11])
labels[:, 13] = np.where(labels[:, 13] < 0, -1, 1 - labels[:, 13])
# 左右镜像的时候，左眼、右眼， 左嘴角、右嘴角无法区分, 应该交换位置，便于网络学习
eye_left = np.copy(labels[:, [5, 6]])
mouth_left = np.copy(labels[:, [11, 12]])
labels[:, [5, 6]] = labels[:, [7, 8]]
labels[:, [7, 8]] = eye_left
labels[:, [11, 12]] = labels[:, [13, 14]]
labels[:, [13, 14]] = mouth_left
labels_out = torch.zeros((nL, 16))
if nL:
labels_out[:, 1:] = torch.from_numpy(labels)
# showlabels(img, labels[:, 1:5], labels[:, 5:15])
# Convert
img = img[:, :, ::-1].transpose(2, 0, 1) # BGR to RGB, to 3x416x416
img = np.ascontiguousarray(img)
# print(index, ' --- labels_out: ', labels_out)
# if nL:
# print( ' : landmarks : ', torch.max(labels_out[:, 5:15]), ' --- ', torch.min(labels_out[:, 5:15]))
return torch.from_numpy(img), labels_out, self.img_files[index], shapes
@staticmethod
def collate_fn(batch):
img, label, path, shapes = zip(*batch) # transposed
for i, l in enumerate(label):
l[:, 0] = i # add target image index for build_targets()
return torch.stack(img, 0), torch.cat(label, 0), path, shapes
@staticmethod
def collate_fn4(batch):
im, label, path, shapes = zip(*batch) # transposed
n = len(shapes) // 4
im4, label4, path4, shapes4 = [], [], path[:n], shapes[:n]
ho = torch.tensor([[0.0, 0, 0, 1, 0, 0]])
wo = torch.tensor([[0.0, 0, 1, 0, 0, 0]])
s = torch.tensor([[1, 1, 0.5, 0.5, 0.5, 0.5]]) # scale
for i in range(n): # zidane torch.zeros(16,3,720,1280) # BCHW
i *= 4
if random.random() < 0.5:
im1 = F.interpolate(im[i].unsqueeze(0).float(), scale_factor=2.0, mode='bilinear',
align_corners=False)[0].type(im[i].type())
lb = label[i]
else:
im1 = torch.cat((torch.cat((im[i], im[i + 1]), 1), torch.cat((im[i + 2], im[i + 3]), 1)), 2)
lb = torch.cat((label[i], label[i + 1] + ho, label[i + 2] + wo, label[i + 3] + ho + wo), 0) * s
im4.append(im1)
label4.append(lb)
for i, lb in enumerate(label4):
lb[:, 0] = i # add target image index for build_targets()
return torch.stack(im4, 0), torch.cat(label4, 0), path4, shapes4
def showlabels(img, boxs, landmarks):
for box in boxs:
x, y, w, h = box[0] * img.shape[1], box[1] * img.shape[0], box[2] * img.shape[1], box[3] * img.shape[0]
# cv2.rectangle(image, (x,y), (x+w,y+h), (0,255,0), 2)
cv2.rectangle(img, (int(x - w / 2), int(y - h / 2)), (int(x + w / 2), int(y + h / 2)), (0, 255, 0), 2)
for landmark in landmarks:
# cv2.circle(img,(60,60),30,(0,0,255))
for i in range(5):
cv2.circle(img, (int(landmark[2 * i] * img.shape[1]), int(landmark[2 * i + 1] * img.shape[0])), 3,
(0, 0, 255), -1)
cv2.imshow('test', img)
cv2.waitKey(0)
def load_mosaic_face(self, index):
# loads images in a mosaic
labels4 = []
s = self.img_size
yc, xc = [int(random.uniform(-x, 2 * s + x)) for x in self.mosaic_border] # mosaic center x, y
indices = [index] + [self.indices[random.randint(0, self.n - 1)] for _ in range(3)] # 3 additional image indices
for i, index in enumerate(indices):
# Load image
img, _, (h, w) = load_image(self, index)
# place img in img4
if i == 0: # top left
img4 = np.full((s * 2, s * 2, img.shape[2]), 114, dtype=np.uint8) # base image with 4 tiles
x1a, y1a, x2a, y2a = max(xc - w, 0), max(yc - h, 0), xc, yc # xmin, ymin, xmax, ymax (large image)
x1b, y1b, x2b, y2b = w - (x2a - x1a), h - (y2a - y1a), w, h # xmin, ymin, xmax, ymax (small image)
elif i == 1: # top right
x1a, y1a, x2a, y2a = xc, max(yc - h, 0), min(xc + w, s * 2), yc
x1b, y1b, x2b, y2b = 0, h - (y2a - y1a), min(w, x2a - x1a), h
elif i == 2: # bottom left
x1a, y1a, x2a, y2a = max(xc - w, 0), yc, xc, min(s * 2, yc + h)
x1b, y1b, x2b, y2b = w - (x2a - x1a), 0, w, min(y2a - y1a, h)
elif i == 3: # bottom right
x1a, y1a, x2a, y2a = xc, yc, min(xc + w, s * 2), min(s * 2, yc + h)
x1b, y1b, x2b, y2b = 0, 0, min(w, x2a - x1a), min(y2a - y1a, h)
img4[y1a:y2a, x1a:x2a] = img[y1b:y2b, x1b:x2b] # img4[ymin:ymax, xmin:xmax]
padw = x1a - x1b
padh = y1a - y1b
# Labels
x = self.labels[index]
labels = x.copy()
if x.size > 0: # Normalized xywh to pixel xyxy format
# box, x1,y1,x2,y2
labels[:, 1] = w * (x[:, 1] - x[:, 3] / 2) + padw
labels[:, 2] = h * (x[:, 2] - x[:, 4] / 2) + padh
labels[:, 3] = w * (x[:, 1] + x[:, 3] / 2) + padw
labels[:, 4] = h * (x[:, 2] + x[:, 4] / 2) + padh
# 10 landmarks
labels[:, 5] = np.array(x[:, 5] > 0, dtype=np.int32) * (w * x[:, 5] + padw) + (
np.array(x[:, 5] > 0, dtype=np.int32) - 1)
labels[:, 6] = np.array(x[:, 6] > 0, dtype=np.int32) * (h * x[:, 6] + padh) + (
np.array(x[:, 6] > 0, dtype=np.int32) - 1)
labels[:, 7] = np.array(x[:, 7] > 0, dtype=np.int32) * (w * x[:, 7] + padw) + (
np.array(x[:, 7] > 0, dtype=np.int32) - 1)
labels[:, 8] = np.array(x[:, 8] > 0, dtype=np.int32) * (h * x[:, 8] + padh) + (
np.array(x[:, 8] > 0, dtype=np.int32) - 1)
labels[:, 9] = np.array(x[:, 9] > 0, dtype=np.int32) * (w * x[:, 9] + padw) + (
np.array(x[:, 9] > 0, dtype=np.int32) - 1)
labels[:, 10] = np.array(x[:, 10] > 0, dtype=np.int32) * (h * x[:, 10] + padh) + (
np.array(x[:, 10] > 0, dtype=np.int32) - 1)
labels[:, 11] = np.array(x[:, 11] > 0, dtype=np.int32) * (w * x[:, 11] + padw) + (
np.array(x[:, 11] > 0, dtype=np.int32) - 1)
labels[:, 12] = np.array(x[:, 12] > 0, dtype=np.int32) * (h * x[:, 12] + padh) + (
np.array(x[:, 12] > 0, dtype=np.int32) - 1)
labels[:, 13] = np.array(x[:, 13] > 0, dtype=np.int32) * (w * x[:, 13] + padw) + (
np.array(x[:, 13] > 0, dtype=np.int32) - 1)
labels[:, 14] = np.array(x[:, 14] > 0, dtype=np.int32) * (h * x[:, 14] + padh) + (
np.array(x[:, 14] > 0, dtype=np.int32) - 1)
labels4.append(labels)
# Concat/clip labels
if len(labels4):
labels4 = np.concatenate(labels4, 0)
np.clip(labels4[:, 1:5], 0, 2 * s, out=labels4[:, 1:5]) # use with random_perspective
# img4, labels4 = replicate(img4, labels4) # replicate
# landmarks
labels4[:, 5:] = np.where(labels4[:, 5:] < 0, -1, labels4[:, 5:])
labels4[:, 5:] = np.where(labels4[:, 5:] > 2 * s, -1, labels4[:, 5:])
labels4[:, 5] = np.where(labels4[:, 6] == -1, -1, labels4[:, 5])
labels4[:, 6] = np.where(labels4[:, 5] == -1, -1, labels4[:, 6])
labels4[:, 7] = np.where(labels4[:, 8] == -1, -1, labels4[:, 7])
labels4[:, 8] = np.where(labels4[:, 7] == -1, -1, labels4[:, 8])
labels4[:, 9] = np.where(labels4[:, 10] == -1, -1, labels4[:, 9])
labels4[:, 10] = np.where(labels4[:, 9] == -1, -1, labels4[:, 10])
labels4[:, 11] = np.where(labels4[:, 12] == -1, -1, labels4[:, 11])
labels4[:, 12] = np.where(labels4[:, 11] == -1, -1, labels4[:, 12])
labels4[:, 13] = np.where(labels4[:, 14] == -1, -1, labels4[:, 13])
labels4[:, 14] = np.where(labels4[:, 13] == -1, -1, labels4[:, 14])
# Augment
img4, labels4 = random_perspective(img4, labels4,
degrees=self.hyp['degrees'],
translate=self.hyp['translate'],
scale=self.hyp['scale'],
shear=self.hyp['shear'],
perspective=self.hyp['perspective'],
border=self.mosaic_border) # border to remove
return img4, labels4
# Ancillary functions --------------------------------------------------------------------------------------------------
def load_image(self, index):
# loads 1 image from dataset, returns img, original hw, resized hw
img = self.imgs[index]
if img is None: # not cached
path = self.img_files[index]
img = cv2.imread(path) # BGR
assert img is not None, 'Image Not Found ' + path
h0, w0 = img.shape[:2] # orig hw
r = self.img_size / max(h0, w0) # resize image to img_size
if r != 1: # always resize down, only resize up if training with augmentation
interp = cv2.INTER_AREA if r < 1 and not self.augment else cv2.INTER_LINEAR
img = cv2.resize(img, (int(w0 * r), int(h0 * r)), interpolation=interp)
return img, (h0, w0), img.shape[:2] # img, hw_original, hw_resized
else:
return self.imgs[index], self.img_hw0[index], self.img_hw[index] # img, hw_original, hw_resized
def augment_hsv(img, hgain=0.5, sgain=0.5, vgain=0.5):
r = np.random.uniform(-1, 1, 3) * [hgain, sgain, vgain] + 1 # random gains
hue, sat, val = cv2.split(cv2.cvtColor(img, cv2.COLOR_BGR2HSV))
dtype = img.dtype # uint8
x = np.arange(0, 256, dtype=np.int16)
lut_hue = ((x * r[0]) % 180).astype(dtype)
lut_sat = np.clip(x * r[1], 0, 255).astype(dtype)
lut_val = np.clip(x * r[2], 0, 255).astype(dtype)
img_hsv = cv2.merge((cv2.LUT(hue, lut_hue), cv2.LUT(sat, lut_sat), cv2.LUT(val, lut_val))).astype(dtype)
cv2.cvtColor(img_hsv, cv2.COLOR_HSV2BGR, dst=img) # no return needed
# Histogram equalization
# if random.random() < 0.2:
# for i in range(3):
# img[:, :, i] = cv2.equalizeHist(img[:, :, i])
def replicate(img, labels):
# Replicate labels
h, w = img.shape[:2]
boxes = labels[:, 1:].astype(int)
x1, y1, x2, y2 = boxes.T
s = ((x2 - x1) + (y2 - y1)) / 2 # side length (pixels)
for i in s.argsort()[:round(s.size * 0.5)]: # smallest indices
x1b, y1b, x2b, y2b = boxes[i]
bh, bw = y2b - y1b, x2b - x1b
yc, xc = int(random.uniform(0, h - bh)), int(random.uniform(0, w - bw)) # offset x, y
x1a, y1a, x2a, y2a = [xc, yc, xc + bw, yc + bh]
img[y1a:y2a, x1a:x2a] = img[y1b:y2b, x1b:x2b] # img4[ymin:ymax, xmin:xmax]
labels = np.append(labels, [[labels[i, 0], x1a, y1a, x2a, y2a]], axis=0)
return img, labels
def letterbox(img, new_shape=(640, 640), color=(114, 114, 114), auto=True, scaleFill=False, scaleup=True):
# Resize image to a 32-pixel-multiple rectangle https://github.com/ultralytics/yolov3/issues/232
shape = img.shape[:2] # current shape [height, width]
if isinstance(new_shape, int):
new_shape = (new_shape, new_shape)
# Scale ratio (new / old)
r = min(new_shape[0] / shape[0], new_shape[1] / shape[1])
if not scaleup: # only scale down, do not scale up (for better test mAP)
r = min(r, 1.0)
# Compute padding
ratio = r, r # width, height ratios
new_unpad = int(round(shape[1] * r)), int(round(shape[0] * r))
dw, dh = new_shape[1] - new_unpad[0], new_shape[0] - new_unpad[1] # wh padding
if auto: # minimum rectangle
dw, dh = np.mod(dw, 64), np.mod(dh, 64) # wh padding
elif scaleFill: # stretch
dw, dh = 0.0, 0.0
new_unpad = (new_shape[1], new_shape[0])
ratio = new_shape[1] / shape[1], new_shape[0] / shape[0] # width, height ratios
dw /= 2 # divide padding into 2 sides
dh /= 2
if shape[::-1] != new_unpad: # resize
img = cv2.resize(img, new_unpad, interpolation=cv2.INTER_LINEAR)
top, bottom = int(round(dh - 0.1)), int(round(dh + 0.1))
left, right = int(round(dw - 0.1)), int(round(dw + 0.1))
img = cv2.copyMakeBorder(img, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color) # add border
return img, ratio, (dw, dh)
def random_perspective(img, targets=(), degrees=10, translate=.1, scale=.1, shear=10, perspective=0.0, border=(0, 0)):
# torchvision.transforms.RandomAffine(degrees=(-10, 10), translate=(.1, .1), scale=(.9, 1.1), shear=(-10, 10))
# targets = [cls, xyxy]
height = img.shape[0] + border[0] * 2 # shape(h,w,c)
width = img.shape[1] + border[1] * 2
# Center
C = np.eye(3)
C[0, 2] = -img.shape[1] / 2 # x translation (pixels)
C[1, 2] = -img.shape[0] / 2 # y translation (pixels)
# Perspective
P = np.eye(3)
P[2, 0] = random.uniform(-perspective, perspective) # x perspective (about y)
P[2, 1] = random.uniform(-perspective, perspective) # y perspective (about x)
# Rotation and Scale
R = np.eye(3)
a = random.uniform(-degrees, degrees)
# a += random.choice([-180, -90, 0, 90]) # add 90deg rotations to small rotations
s = random.uniform(1 - scale, 1 + scale)
# s = 2 ** random.uniform(-scale, scale)
R[:2] = cv2.getRotationMatrix2D(angle=a, center=(0, 0), scale=s)
# Shear
S = np.eye(3)
S[0, 1] = math.tan(random.uniform(-shear, shear) * math.pi / 180) # x shear (deg)
S[1, 0] = math.tan(random.uniform(-shear, shear) * math.pi / 180) # y shear (deg)
# Translation
T = np.eye(3)
T[0, 2] = random.uniform(0.5 - translate, 0.5 + translate) * width # x translation (pixels)
T[1, 2] = random.uniform(0.5 - translate, 0.5 + translate) * height # y translation (pixels)
# Combined rotation matrix
M = T @ S @ R @ P @ C # order of operations (right to left) is IMPORTANT
if (border[0] != 0) or (border[1] != 0) or (M != np.eye(3)).any(): # image changed
if perspective:
img = cv2.warpPerspective(img, M, dsize=(width, height), borderValue=(114, 114, 114))
else: # affine
img = cv2.warpAffine(img, M[:2], dsize=(width, height), borderValue=(114, 114, 114))
# Visualize
# import matplotlib.pyplot as plt
# ax = plt.subplots(1, 2, figsize=(12, 6))[1].ravel()
# ax[0].imshow(img[:, :, ::-1]) # base
# ax[1].imshow(img2[:, :, ::-1]) # warped
# Transform label coordinates
n = len(targets)
if n:
# warp points
# xy = np.ones((n * 4, 3))
xy = np.ones((n * 9, 3))
xy[:, :2] = targets[:, [1, 2, 3, 4, 1, 4, 3, 2, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14]].reshape(n * 9,
2) # x1y1, x2y2, x1y2, x2y1
xy = xy @ M.T # transform
if perspective:
xy = (xy[:, :2] / xy[:, 2:3]).reshape(n, 18) # rescale
else: # affine
xy = xy[:, :2].reshape(n, 18)
# create new boxes
x = xy[:, [0, 2, 4, 6]]
y = xy[:, [1, 3, 5, 7]]
landmarks = xy[:, [8, 9, 10, 11, 12, 13, 14, 15, 16, 17]]
mask = np.array(targets[:, 5:] > 0, dtype=np.int32)
landmarks = landmarks * mask
landmarks = landmarks + mask - 1
landmarks = np.where(landmarks < 0, -1, landmarks)
landmarks[:, [0, 2, 4, 6, 8]] = np.where(landmarks[:, [0, 2, 4, 6, 8]] > width, -1,
landmarks[:, [0, 2, 4, 6, 8]])
landmarks[:, [1, 3, 5, 7, 9]] = np.where(landmarks[:, [1, 3, 5, 7, 9]] > height, -1,
landmarks[:, [1, 3, 5, 7, 9]])
landmarks[:, 0] = np.where(landmarks[:, 1] == -1, -1, landmarks[:, 0])
landmarks[:, 1] = np.where(landmarks[:, 0] == -1, -1, landmarks[:, 1])
landmarks[:, 2] = np.where(landmarks[:, 3] == -1, -1, landmarks[:, 2])
landmarks[:, 3] = np.where(landmarks[:, 2] == -1, -1, landmarks[:, 3])
landmarks[:, 4] = np.where(landmarks[:, 5] == -1, -1, landmarks[:, 4])
landmarks[:, 5] = np.where(landmarks[:, 4] == -1, -1, landmarks[:, 5])
landmarks[:, 6] = np.where(landmarks[:, 7] == -1, -1, landmarks[:, 6])
landmarks[:, 7] = np.where(landmarks[:, 6] == -1, -1, landmarks[:, 7])
landmarks[:, 8] = np.where(landmarks[:, 9] == -1, -1, landmarks[:, 8])
landmarks[:, 9] = np.where(landmarks[:, 8] == -1, -1, landmarks[:, 9])
targets[:, 5:] = landmarks
xy = np.concatenate((x.min(1), y.min(1), x.max(1), y.max(1))).reshape(4, n).T
# # apply angle-based reduction of bounding boxes
# radians = a * math.pi / 180
# reduction = max(abs(math.sin(radians)), abs(math.cos(radians))) ** 0.5
# x = (xy[:, 2] + xy[:, 0]) / 2
# y = (xy[:, 3] + xy[:, 1]) / 2
# w = (xy[:, 2] - xy[:, 0]) * reduction
# h = (xy[:, 3] - xy[:, 1]) * reduction
# xy = np.concatenate((x - w / 2, y - h / 2, x + w / 2, y + h / 2)).reshape(4, n).T
# clip boxes
xy[:, [0, 2]] = xy[:, [0, 2]].clip(0, width)
xy[:, [1, 3]] = xy[:, [1, 3]].clip(0, height)
# filter candidates
i = box_candidates(box1=targets[:, 1:5].T * s, box2=xy.T)
targets = targets[i]
targets[:, 1:5] = xy[i]
return img, targets
def box_candidates(box1, box2, wh_thr=2, ar_thr=20, area_thr=0.1): # box1(4,n), box2(4,n)
# Compute candidate boxes: box1 before augment, box2 after augment, wh_thr (pixels), aspect_ratio_thr, area_ratio
w1, h1 = box1[2] - box1[0], box1[3] - box1[1]
w2, h2 = box2[2] - box2[0], box2[3] - box2[1]
ar = np.maximum(w2 / (h2 + 1e-16), h2 / (w2 + 1e-16)) # aspect ratio
return (w2 > wh_thr) & (h2 > wh_thr) & (w2 * h2 / (w1 * h1 + 1e-16) > area_thr) & (ar < ar_thr) # candidates
def cutout(image, labels):
# Applies image cutout augmentation https://arxiv.org/abs/1708.04552
h, w = image.shape[:2]
def bbox_ioa(box1, box2):
# Returns the intersection over box2 area given box1, box2. box1 is 4, box2 is nx4. boxes are x1y1x2y2
box2 = box2.transpose()
# Get the coordinates of bounding boxes
b1_x1, b1_y1, b1_x2, b1_y2 = box1[0], box1[1], box1[2], box1[3]
b2_x1, b2_y1, b2_x2, b2_y2 = box2[0], box2[1], box2[2], box2[3]
# Intersection area
inter_area = (np.minimum(b1_x2, b2_x2) - np.maximum(b1_x1, b2_x1)).clip(0) * \
(np.minimum(b1_y2, b2_y2) - np.maximum(b1_y1, b2_y1)).clip(0)
# box2 area
box2_area = (b2_x2 - b2_x1) * (b2_y2 - b2_y1) + 1e-16
# Intersection over box2 area
return inter_area / box2_area
# create random masks
scales = [0.5] * 1 + [0.25] * 2 + [0.125] * 4 + [0.0625] * 8 + [0.03125] * 16 # image size fraction
for s in scales:
mask_h = random.randint(1, int(h * s))
mask_w = random.randint(1, int(w * s))
# box
xmin = max(0, random.randint(0, w) - mask_w // 2)
ymin = max(0, random.randint(0, h) - mask_h // 2)
xmax = min(w, xmin + mask_w)
ymax = min(h, ymin + mask_h)
# apply random color mask
image[ymin:ymax, xmin:xmax] = [random.randint(64, 191) for _ in range(3)]
# return unobscured labels
if len(labels) and s > 0.03:
box = np.array([xmin, ymin, xmax, ymax], dtype=np.float32)
ioa = bbox_ioa(box, labels[:, 1:5]) # intersection over area
labels = labels[ioa < 0.60] # remove >60% obscured labels
return labels
def create_folder(path='./new'):
# Create folder
if os.path.exists(path):
shutil.rmtree(path) # delete output folder
os.makedirs(path) # make new output folder
def flatten_recursive(path='../coco128'):
# Flatten a recursive directory by bringing all files to top level
new_path = Path(path + '_flat')
create_folder(new_path)
for file in tqdm(glob.glob(str(Path(path)) + '/**/*.*', recursive=True)):
shutil.copyfile(file, new_path / Path(file).name)
def extract_boxes(path='../coco128/'): # from utils.datasets import *; extract_boxes('../coco128')
# Convert detection dataset into classification dataset, with one directory per class
path = Path(path) # images dir
shutil.rmtree(path / 'classifier') if (path / 'classifier').is_dir() else None # remove existing
files = list(path.rglob('*.*'))
n = len(files) # number of files
for im_file in tqdm(files, total=n):
if im_file.suffix[1:] in img_formats:
# image
im = cv2.imread(str(im_file))[..., ::-1] # BGR to RGB
h, w = im.shape[:2]
# labels
lb_file = Path(img2label_paths([str(im_file)])[0])
if Path(lb_file).exists():
with open(lb_file, 'r') as f:
lb = np.array([x.split() for x in f.read().strip().splitlines()], dtype=np.float32) # labels
for j, x in enumerate(lb):
c = int(x[0]) # class
f = (path / 'classifier') / f'{c}' / f'{path.stem}_{im_file.stem}_{j}.jpg' # new filename
if not f.parent.is_dir():
f.parent.mkdir(parents=True)
b = x[1:] * [w, h, w, h] # box
# b[2:] = b[2:].max() # rectangle to square
b[2:] = b[2:] * 1.2 + 3 # pad
b = xywh2xyxy(b.reshape(-1, 4)).ravel().astype(np.int)
b[[0, 2]] = np.clip(b[[0, 2]], 0, w) # clip boxes outside of image
b[[1, 3]] = np.clip(b[[1, 3]], 0, h)
assert cv2.imwrite(str(f), im[b[1]:b[3], b[0]:b[2]]), f'box failure in {f}'
def autosplit(path='../coco128', weights=(0.9, 0.1, 0.0)): # from utils.datasets import *; autosplit('../coco128')
""" Autosplit a dataset into train/val/test splits and save path/autosplit_*.txt files
# Arguments
path: Path to images directory
weights: Train, val, test weights (list)
"""
path = Path(path) # images dir
files = list(path.rglob('*.*'))
n = len(files) # number of files
indices = random.choices([0, 1, 2], weights=weights, k=n) # assign each image to a split
txt = ['autosplit_train.txt', 'autosplit_val.txt', 'autosplit_test.txt'] # 3 txt files
[(path / x).unlink() for x in txt if (path / x).exists()] # remove existing
for i, img in tqdm(zip(indices, files), total=n):
if img.suffix[1:] in img_formats:
with open(path / txt[i], 'a') as f:
f.write(str(img) + '\n') # add image to txt file
models文件夹
models/yolo.py
class LmkDetect(nn.Module):
stride = None # strides computed during build
export = False # onnx export cat output
def __init__(self, nc=80, anchors=(), ch=()): # detection layer
super(LmkDetect, self).__init__()
self.nc = nc # number of classes
# self.no = nc + 5 # number of outputs per anchor
self.no = nc + 5 + 10 # number of outputs per anchor
self.nl = len(anchors) # number of detection layers
self.na = len(anchors[0]) // 2 # number of anchors
self.grid = [torch.zeros(1)] * self.nl # init grid
a = torch.tensor(anchors).float().view(self.nl, -1, 2)
self.register_buffer('anchors', a) # shape(nl,na,2)
self.register_buffer('anchor_grid', a.clone().view(self.nl, 1, -1, 1, 1, 2)) # shape(nl,1,na,1,1,2)
self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch) # output conv
def forward(self, x):
# x = x.copy() # for profiling
z = [] # inference output
if self.export:
for i in range(self.nl):
x[i] = self.m[i](x[i]) # conv
bs, _, ny, nx = x[i].shape # x(bs,255,20,20) to x(bs,3,20,20,85)
x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()
if self.grid[i].shape[2:4] != x[i].shape[2:4]:
# self.grid[i] = self._make_grid(nx, ny).to(x[i].device)
self.grid[i], self.anchor_grid[i] = self._make_grid_new(nx, ny, i)
y = torch.full_like(x[i], 0)
y = y + torch.cat((x[i][:, :, :, :, 0:5].sigmoid(),
torch.cat((x[i][:, :, :, :, 5:15], x[i][:, :, :, :, 15:15 + self.nc].sigmoid()), 4)),
4)
box_xy = (y[:, :, :, :, 0:2] * 2. - 0.5 + self.grid[i].to(x[i].device)) * self.stride[i] # xy
box_wh = (y[:, :, :, :, 2:4] * 2) ** 2 * self.anchor_grid[i] # wh
# box_conf = torch.cat((box_xy, torch.cat((box_wh, y[:, :, :, :, 4:5]), 4)), 4)
landm1 = y[:, :, :, :, 5:7] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[
i] # landmark x1 y1
landm2 = y[:, :, :, :, 7:9] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[
i] # landmark x2 y2
landm3 = y[:, :, :, :, 9:11] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[
i] # landmark x3 y3
landm4 = y[:, :, :, :, 11:13] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[
i] # landmark x4 y4
landm5 = y[:, :, :, :, 13:15] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[
i] # landmark x5 y5
# landm = torch.cat((landm1, torch.cat((landm2, torch.cat((landm3, torch.cat((landm4, landm5), 4)), 4)), 4)), 4)
# y = torch.cat((box_conf, torch.cat((landm, y[:, :, :, :, 15:15+self.nc]), 4)), 4)
y = torch.cat([box_xy, box_wh, y[:, :, :, :, 4:5], landm1, landm2, landm3, landm4, landm5,
y[:, :, :, :, 15:15 + self.nc]], -1)
z.append(y.view(bs, -1, self.no))
return torch.cat(z, 1)
for i in range(self.nl):
x[i] = self.m[i](x[i]) # conv
bs, _, ny, nx = x[i].shape # x(bs,255,20,20) to x(bs,3,20,20,85)
x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()
if not self.training: # inference
if self.grid[i].shape[2:4] != x[i].shape[2:4]:
self.grid[i] = self._make_grid(nx, ny).to(x[i].device)
y = torch.full_like(x[i], 0)
class_range = list(range(5)) + list(range(15, 15 + self.nc))
y[..., class_range] = x[i][..., class_range].sigmoid()
y[..., 5:15] = x[i][..., 5:15]
# y = x[i].sigmoid()
y[..., 0:2] = (y[..., 0:2] * 2. - 0.5 + self.grid[i].to(x[i].device)) * self.stride[i] # xy
y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # wh
# y[..., 5:15] = y[..., 5:15] * 8 - 4
y[..., 5:7] = y[..., 5:7] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[
i] # landmark x1 y1
y[..., 7:9] = y[..., 7:9] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[
i] # landmark x2 y2
y[..., 9:11] = y[..., 9:11] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[
i] # landmark x3 y3
y[..., 11:13] = y[..., 11:13] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[
i] # landmark x4 y4
y[..., 13:15] = y[..., 13:15] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[
i] # landmark x5 y5
# y[..., 5:7] = (y[..., 5:7] * 2 -1) * self.anchor_grid[i] # landmark x1 y1
# y[..., 7:9] = (y[..., 7:9] * 2 -1) * self.anchor_grid[i] # landmark x2 y2
# y[..., 9:11] = (y[..., 9:11] * 2 -1) * self.anchor_grid[i] # landmark x3 y3
# y[..., 11:13] = (y[..., 11:13] * 2 -1) * self.anchor_grid[i] # landmark x4 y4
# y[..., 13:15] = (y[..., 13:15] * 2 -1) * self.anchor_grid[i] # landmark x5 y5
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()
def _make_grid_new(self, nx=20, ny=20, i=0):
d = self.anchors[i].device
if '1.10.0' in torch.__version__: # torch>=1.10.0 meshgrid workaround for torch>=0.7 compatibility
yv, xv = torch.meshgrid([torch.arange(ny).to(d), torch.arange(nx).to(d)], indexing='ij')
else:
yv, xv = torch.meshgrid([torch.arange(ny).to(d), torch.arange(nx).to(d)])
grid = torch.stack((xv, yv), 2).expand((1, self.na, ny, nx, 2)).float()
anchor_grid = (self.anchors[i].clone() * self.stride[i]).view((1, self.na, 1, 1, 2)).expand(
(1, self.na, ny, nx, 2)).float()
return grid, anchor_grid同时在以下两个函数中增加LmkDetect
1、DetectionModel中
if isinstance(m, (Detect, Segment, LmkDetect))
2、parse_model中
elif m in {Detect, Segment, LmkDetect}
models/experimental.py添加
def attempt_loadf(weights, device=None, inplace=True, fuse=True):
# Loads an ensemble of models weights=[a,b,c] or a single model weights=[a] or weights=a
from models.yolo import Detect, Model, LmkDetect
model = Ensemble()
for w in weights if isinstance(weights, list) else [weights]:
ckpt = torch.load(attempt_download(w), map_location='cpu') # load
ckpt = (ckpt.get('ema') or ckpt['model']).to(device).float() # FP32 model
# Model compatibility updates
if not hasattr(ckpt, 'stride'):
ckpt.stride = torch.tensor([32.])
if hasattr(ckpt, 'names') and isinstance(ckpt.names, (list, tuple)):
ckpt.names = dict(enumerate(ckpt.names)) # convert to dict
model.append(ckpt.fuse().eval() if fuse and hasattr(ckpt, 'fuse') else ckpt.eval()) # model in eval mode
# Module compatibility updates
for m in model.modules():
t = type(m)
if t in (nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6, nn.SiLU, Detect, Model,LmkDetect):
m.inplace = inplace # torch 1.7.0 compatibility
if t is Detect and not isinstance(m.anchor_grid, list):
delattr(m, 'anchor_grid')
setattr(m, 'anchor_grid', [torch.zeros(1)] * m.nl)
elif t is LmkDetect and not isinstance(m.anchor_grid, list):
delattr(m, 'anchor_grid')
setattr(m, 'anchor_grid', [torch.zeros(1)] * m.nl)
elif t is nn.Upsample and not hasattr(m, 'recompute_scale_factor'):
m.recompute_scale_factor = None # torch 1.11.0 compatibility
# Return model
if len(model) == 1:
return model[-1]
# Return detection ensemble
print(f'Ensemble created with {weights}\n')
for k in 'names', 'nc', 'yaml':
setattr(model, k, getattr(model[0], k))
model.stride = model[torch.argmax(torch.tensor([m.stride.max() for m in model])).int()].stride # max stride
assert all(model[0].nc == m.nc for m in model), f'Models have different class counts: {[m.nc for m in model]}'
return model模型配置文件
models/landmark/yolov5n-landmark.yaml
# YOLOv5 by Ultralytics, GPL-3.0 license
# Parameters
nc: 80 # number of classes
depth_multiple: 0.33 # model depth multiple
width_multiple: 0.25 # layer channel multiple
anchors:
- [4,5, 8,10, 13,16] # P3/8
- [23,29, 43,55, 73,105] # P4/16
- [146,217, 231,300, 335,433] # P5/32
# YOLOv5 v6.0 backbone
backbone: #channel * (1/width_multiple)
# [from, number, module, args] input size [N, 3, 640, 640]
[[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [N, 16, 320, 320]
[-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [N, 32, 160, 160]
[-1, 3, C3, [128]], # [N, 32, 160, 160]
[-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [N, 64, 80, 80]
[-1, 6, C3, [256]], # [N, 64, 80, 80]
[-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [N, 128, 40, 40]
[-1, 9, C3, [512]], # [N, 128, 40, 40]
[-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [N, 256, 20, 20]
[-1, 1, SPPF, [1024, 5]], # 9 [N, 256, 20, 20]
[-1, 3, C3, [1024, False]],# [N, 256, 20, 20]
]
# YOLOv5 v6.0 head
head:
[[-1, 1, Conv, [512, 1, 1]], # [N, 128, 20, 20]
[-1, 1, nn.Upsample, [None, 2, 'nearest']], # [N, 128, 40, 40]
[[-1, 6], 1, Concat, [1]], # cat backbone P4 [N, 256, 40, 40]
[-1, 3, C3, [512, False]], # 13 [N, 128, 40, 40]
[-1, 1, Conv, [256, 1, 1]], # [N, 64, 40, 40]
[-1, 1, nn.Upsample, [None, 2, 'nearest']], # [N, 64, 80, 80]
[[-1, 4], 1, Concat, [1]], # cat backbone P3 [N, 128, 80, 80]
[-1, 3, C3, [256, False]], # 17 (P3/8-small) [N, 64, 80, 80]
[-1, 1, Conv, [256, 3, 2]], # [N, 64, 40, 40]
[[-1, 14], 1, Concat, [1]], # cat head P4 [N, 128, 40, 40]
[-1, 3, C3, [512, False]], # 20 (P4/16-medium) [N, 128, 40, 40]
[-1, 1, Conv, [512, 3, 2]], # [N, 128, 20, 20]
[[-1, 10], 1, Concat, [1]], # cat head P5 [N, 256, 20, 20]
[-1, 3, C3, [1024, False]], # 23 (P5/32-large) [N, 256, 20, 20]
[[17, 20, 23], 1, LmkDetect, [nc, anchors]], # Detect(P3, P4, P5)
]
#YOLOv5n summary: 270 layers, 1872157 parameters, 1872157 gradients, 4.5 GFLOPs超参 data/hyps/hyp.scratch.yaml
# Hyperparameters for COCO training from scratch
# python train.py --batch 40 --cfg yolov5m.yaml --weights '' --data coco.yaml --img 640 --epochs 300
# See tutorials for hyperparameter evolution https://github.com/ultralytics/yolov5#tutorials
lr0: 0.01 # initial learning rate (SGD=1E-2, Adam=1E-3)
lrf: 0.2 # final OneCycleLR learning rate (lr0 * lrf)
momentum: 0.937 # SGD momentum/Adam beta1
weight_decay: 0.0005 # optimizer weight decay 5e-4
warmup_epochs: 3.0 # warmup epochs (fractions ok)
warmup_momentum: 0.8 # warmup initial momentum
warmup_bias_lr: 0.1 # warmup initial bias lr
box: 0.05 # box loss gain
cls: 0.5 # cls loss gain
landmark: 0.005 # landmark loss gain
cls_pw: 1.0 # cls BCELoss positive_weight
obj: 1.0 # obj loss gain (scale with pixels)
obj_pw: 1.0 # obj BCELoss positive_weight
iou_t: 0.20 # IoU training threshold
anchor_t: 4.0 # anchor-multiple threshold
# anchors: 3 # anchors per output layer (0 to ignore)
fl_gamma: 0.0 # focal loss gamma (efficientDet default gamma=1.5)
hsv_h: 0.015 # image HSV-Hue augmentation (fraction)
hsv_s: 0.7 # image HSV-Saturation augmentation (fraction)
hsv_v: 0.4 # image HSV-Value augmentation (fraction)
degrees: 0.0 # image rotation (+/- deg)
translate: 0.1 # image translation (+/- fraction)
scale: 0.5 # image scale (+/- gain)
shear: 0.5 # image shear (+/- deg)
perspective: 0.0 # image perspective (+/- fraction), range 0-0.001
flipud: 0.0 # image flip up-down (probability)
fliplr: 0.5 # image flip left-right (probability)
mosaic: 0.5 # image mosaic (probability)
mixup: 0.0 # image mixup (probability)
检测结果
五个关键点（眼睛，鼻子，嘴）
完整代码：GitHub