def on_train_batch_end(self, ni, model, imgs, targets, paths, plots):
# Callback runs on train batch end
if plots:
if ni == 0:
if not self.opt.sync_bn: # --sync known issue https://github.com/ultralytics/yolov5/issues/3754
with warnings.catch_warnings():
warnings.simplefilter('ignore') # suppress jit trace warning
self.tb.add_graph(torch.jit.trace(de_parallel(model), imgs[0:1], strict=False), [])
if ni < 3:
f = self.save_dir / f'train_batch{ni}.jpg' # filename
plot_images(imgs, targets, paths, f)
if self.wandb and ni == 10:
files = sorted(self.save_dir.glob('train*.jpg'))
self.wandb.log({'Mosaics': [wandb.Image(str(f), caption=f.name) for f in files if f.exists()]})
def test(self, epoch, state='test'):
# initialize dataset
self.data_loader.initialize(self.sess, state=state)
# initialize tqdm
tt = tqdm(range(self.data_loader.num_iterations_test),
total=self.data_loader.num_iterations_test,
desc="Test-{}-".format(epoch))
loss_per_epoch = AverageMeter()
# Iterate over batches
for _ in tt:
# One Train step on the current batch
loss = self.sess.run([self.loss_node],
feed_dict={self.training: False})[0]
# update metrics returned from train_step func
loss_per_epoch.update(loss)
# summarize
summaries_dict = {'test/loss_per_epoch': loss_per_epoch.val}
self.summarizer.summarize(
self.model.global_step_tensor.eval(self.sess), summaries_dict)
print("""
Test-{} loss:{:.4f}
""".format(epoch, loss_per_epoch.val))
# generate images
filename = 'epoch-' + str(self.model.global_step_tensor.eval(
self.sess))
plot_images(self.sample_x(),
filename,
self.config.etc_dir,
shape=(32, 32),
n_rows=8)
tt.close()
def test(data,
weights=None,
batch_size=16,
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_conf=False,
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)
save_txt = opt.save_txt # save *.txt labels
# 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, map_location=device) # 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
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
whwh = torch.Tensor([width, height, width, height]).to(device)
# Disable gradients
with torch.no_grad():
# Run model
t = time_synchronized()
inf_out, train_out = model(img, augment=augment) # inference and training outputs
t0 += time_synchronized() - t
# Compute loss
if training: # if model has loss hyperparameters
loss += compute_loss([x.float() for x in train_out], targets, model)[1][:3] # box, obj, cls
# Run NMS
t = time_synchronized()
output = non_max_suppression(inf_out, conf_thres=conf_thres, iou_thres=iou_thres)
t1 += time_synchronized() - t
# Statistics per image
for si, pred in enumerate(output):
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))
continue
# Predictions
predn = pred.clone()
scale_coords(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[5])] if is_coco else int(p[5]),
'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]) * whwh
scale_coords(img[si].shape[1:], tbox, shapes[si][0], shapes[si][1]) # native-space labels
# 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:
f = save_dir / f'test_batch{batch_i}_labels.jpg' # filename
plot_images(img, targets, paths, f, names) # labels
f = save_dir / f'test_batch{batch_i}_pred.jpg'
plot_images(img, output_to_target(output, width, height), paths, f, names) # predictions
# 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_class(*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)
# W&B logging
if plots and wandb and wandb.run:
wandb.log({"Images": wandb_images})
wandb.log({"Validation": [wandb.Image(str(x), caption=x.name) for x in sorted(save_dir.glob('test*.jpg'))]})
# 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)
# 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 = glob.glob('../coco/annotations/instances_val*.json')[0] # 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('ERROR: pycocotools unable to run: %s' % e)
# Return results
if not training:
print('Results saved to %s' % save_dir)
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
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
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, names = (1, ['item']) if opt.single_cls else (int(
data_dict['nc']), data_dict['names']) # number classes, names
assert len(names) == nc, '%g names found for nc=%g dataset in %s' % (
len(names), nc, opt.data) # check
# Model
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:
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)
# 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()')
# Exponential moving average
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)
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,
hyp=hyp,
cache=opt.cache_images
and not opt.notest,
rect=True,
rank=-1,
world_size=opt.world_size,
workers=opt.workers)[0] # testloader
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))
plot_labels(labels, save_dir=save_dir)
if tb_writer:
# tb_writer.add_hparams(hyp, {}) # causes duplicate https://github.com/ultralytics/yolov5/pull/384
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) # 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 # 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(4, device=device) # mean losses
if rank != -1:
dataloader.sampler.set_epoch(epoch)
pbar = enumerate(dataloader)
logger.info(
('\n' + '%10s' * 8) % ('Epoch', 'gpu_mem', 'box', 'obj', 'cls',
'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' * 6) % ('%g/%g' % (epoch, epochs - 1), mem,
*mloss, targets.shape[0], imgs.shape[-1])
pbar.set_description(s)
# Plot
if ni < 3:
f = str(save_dir / f'train_batch{ni}.jpg') # filename
result = plot_images(images=imgs,
targets=targets,
paths=paths,
fname=f)
# if tb_writer and result is not None:
# tb_writer.add_image(f, result, dataformats='HWC', global_step=epoch)
# tb_writer.add_graph(model, imgs) # add model to tensorboard
# end batch ------------------------------------------------------------------------------------------------
# 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]:
# mAP
if ema:
ema.update_attr(
model,
include=['yaml', 'nc', 'hyp', 'gr', 'names', 'stride'])
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=epoch == 0 or final_epoch, # plot first and last
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
n = opt.name if opt.name.isnumeric() else ''
fresults, flast, fbest = save_dir / f'results{n}.txt', wdir / f'last{n}.pt', wdir / f'best{n}.pt'
for f1, f2 in zip([wdir / 'last.pt', wdir / 'best.pt', results_file],
[flast, fbest, fresults]):
if f1.exists():
os.rename(f1, f2) # rename
if str(f2).endswith('.pt'): # is *.pt
strip_optimizer(f2) # strip optimizer
os.system(
'gsutil cp %s gs://%s/weights' %
(f2, opt.bucket)) if opt.bucket else None # upload
# Finish
if not opt.evolve:
plot_results(save_dir=save_dir) # save as results.png
logger.info('%g epochs completed in %.3f hours.\n' %
(epoch - start_epoch + 1, (time.time() - t0) / 3600))
dist.destroy_process_group() if rank not in [-1, 0] else None
torch.cuda.empty_cache()
return results
def test(self, epoch, state='test'):
# initialize dataset
self.data_loader.initialize(self.sess, state=state)
# initialize tqdm
tt = tqdm(range(self.data_loader.num_iterations_test),
total=self.data_loader.num_iterations_test,
desc="Test-{}-".format(epoch))
loss_per_epoch = AverageMeter()
cross_entropy_per_epoch = AverageMeter()
acc_per_epoch = AverageMeter()
ensemble_cross_entropy_per_epoch = AverageMeter()
ensemble_acc_per_epoch = AverageMeter()
# Iterate over batches
for _ in tt:
# One Train step on the current batch
loss, cross_entropy, acc = self.sess.run(
[self.loss_node, self.cross_entropy_node, self.acc_node],
feed_dict={self.training: False})
# update metrics returned from train_step func
loss_per_epoch.update(loss)
acc_per_epoch.update(acc)
cross_entropy_per_epoch.update(cross_entropy)
y_sample, sampled_softmax = self.sess.run(
[self.y, self.model.sampled_softmaxs])
ensemble_softmax = np.mean(sampled_softmax, axis=1)
ensemble_softmax[ensemble_softmax < 1e-8] = 1e-8
#for _ in range(self.config.num_test_sample):
# y_sample, softmax_sample = self.sess.run([self.y, self.model.out_softmax], feed_dict={self.training: False})
# ensemble_softmax += np.array(softmax_sample)/self.config.num_test_sample
onehot_y_sample = np.eye(self.config.num_classes)[y_sample]
ensemble_cross_entropy = -np.mean(np.sum(
onehot_y_sample * np.log(ensemble_softmax), axis=1),
axis=0)
ensemble_acc = np.mean(
np.equal(np.argmax(ensemble_softmax, axis=1), y_sample))
ensemble_cross_entropy_per_epoch.update(ensemble_cross_entropy)
ensemble_acc_per_epoch.update(ensemble_acc)
# summarize
summaries_dict = {
'test/loss_per_epoch':
loss_per_epoch.val,
'test/cross_entropy_per_epoch':
cross_entropy_per_epoch.val,
'test/acc_per_epoch':
acc_per_epoch.val,
'test/ensemble_cross_entropy_per_epoch':
ensemble_cross_entropy_per_epoch.val,
'test/ensemble_acc_per_epoch':
ensemble_acc_per_epoch.val
}
self.summarizer.summarize(
self.model.global_step_tensor.eval(self.sess), summaries_dict)
print("""
Test-{} loss:{:.4f} -- cross entropy:{:.4f} -- acc:{:.4f} -- ensemble cross entropy:{:.4f} -- ensemble acc:{:.4f}
""".format(epoch, loss_per_epoch.val, cross_entropy_per_epoch.val,
acc_per_epoch.val, ensemble_cross_entropy_per_epoch.val,
ensemble_acc_per_epoch.val))
tt.close()
# generate images
filename = 'epoch-' + str(self.model.global_step_tensor.eval(
self.sess))
plot_images(self.sample_x(),
filename,
self.config.etc_dir,
shape=(32, 32),
n_rows=8)
tt.close()
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, names = (
(1, ["item"]) if opt.single_cls else (int(data_dict["nc"]), data_dict["names"])
) # number classes, names
assert len(names) == nc, "%g names found for nc=%g dataset in %s" % (
len(names),
nc,
opt.data,
) # check
# Model
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,
)
# 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,
)
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,
hyp=hyp,
cache=opt.cache_images and not opt.notest,
rect=True,
rank=-1,
world_size=opt.world_size,
workers=opt.workers,
)[
0
] # testloader
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_labels(labels, save_dir=save_dir)
if tb_writer:
tb_writer.add_histogram("classes", c, 0)
if wandb:
wandb.log(
{
"Labels": [
wandb.Image(str(x), caption=x.name)
for x in save_dir.glob("*labels*.png")
]
}
)
# Anchors
if not opt.noautoanchor:
check_anchors(dataset, model=model, thr=hyp["anchor_t"], imgsz=imgsz)
# Model parameters
hyp["cls"] *= nc / 80.0 # 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
) # 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
) # 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(4, device=device) # mean losses
if rank != -1:
dataloader.sampler.set_epoch(epoch)
pbar = enumerate(dataloader)
logger.info(
("\n" + "%10s" * 8)
% ("Epoch", "gpu_mem", "box", "obj", "cls", "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" * 6) % (
"%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
plot_images(images=imgs, targets=targets, paths=paths, fname=f)
# 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]:
# mAP
if ema:
ema.update_attr(
model, include=["yaml", "nc", "hyp", "gr", "names", "stride"]
)
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=plots and final_epoch,
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
n = opt.name if opt.name.isnumeric() else ""
fresults, flast, fbest = (
save_dir / f"results{n}.txt",
wdir / f"last{n}.pt",
wdir / f"best{n}.pt",
)
for f1, f2 in zip(
[wdir / "last.pt", wdir / "best.pt", results_file], [flast, fbest, fresults]
):
if f1.exists():
os.rename(f1, f2) # rename
if str(f2).endswith(".pt"): # is *.pt
strip_optimizer(f2) # strip optimizer
os.system(
"gsutil cp %s gs://%s/weights" % (f2, opt.bucket)
) if opt.bucket else None # upload
# Finish
if plots:
plot_results(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()
]
}
)
logger.info(
"%g epochs completed in %.3f hours.\n"
% (epoch - start_epoch + 1, (time.time() - t0) / 3600)
)
else:
dist.destroy_process_group()
wandb.run.finish() if wandb and wandb.run else None
torch.cuda.empty_cache()
return results
def run(
data,
weights=None, # model.pt path(s)
batch_size=32, # batch size
imgsz=640, # inference size (pixels)
conf_thres=0.001, # confidence threshold
iou_thres=0.6, # NMS IoU threshold
task='val', # train, val, test, speed or study
device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu
workers=8, # max dataloader workers (per RANK in DDP mode)
single_cls=False, # treat as single-class dataset
augment=False, # augmented inference
verbose=False, # verbose output
save_txt=False, # save results to *.txt
save_hybrid=False, # save label+prediction hybrid results to *.txt
save_conf=False, # save confidences in --save-txt labels
save_json=False, # save a COCO-JSON results file
project=ROOT / 'runs/val', # save to project/name
name='exp', # save to project/name
exist_ok=False, # existing project/name ok, do not increment
half=True, # use FP16 half-precision inference
dnn=False, # use OpenCV DNN for ONNX inference
model=None,
dataloader=None,
save_dir=Path(''),
plots=True,
callbacks=Callbacks(),
compute_loss=None,
):
# Initialize/load model and set device
training = model is not None
if training: # called by train.py
device, pt, jit, engine = next(model.parameters()).device, True, False, False # get model device, PyTorch model
half &= device.type != 'cpu' # half precision only supported on CUDA
model.half() if half else model.float()
else: # called directly
device = select_device(device, batch_size=batch_size)
# 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
model = DetectMultiBackend(weights, device=device, dnn=dnn, data=data, fp16=half)
stride, pt, jit, engine = model.stride, model.pt, model.jit, model.engine
imgsz = check_img_size(imgsz, s=stride) # check image size
half = model.fp16 # FP16 supported on limited backends with CUDA
if engine:
batch_size = model.batch_size
else:
device = model.device
if not (pt or jit):
batch_size = 1 # export.py models default to batch-size 1
LOGGER.info(f'Forcing --batch-size 1 square inference (1,3,{imgsz},{imgsz}) for non-PyTorch models')
# Data
data = check_dataset(data) # check
# Configure
model.eval()
cuda = device.type != 'cpu'
is_coco = isinstance(data.get('val'), str) and data['val'].endswith(f'coco{os.sep}val2017.txt') # COCO dataset
nc = 1 if single_cls else int(data['nc']) # number of classes
iouv = torch.linspace(0.5, 0.95, 10, device=device) # iou vector for [email protected]:0.95
niou = iouv.numel()
# Dataloader
if not training:
if pt and not single_cls: # check --weights are trained on --data
ncm = model.model.nc
assert ncm == nc, f'{weights[0]} ({ncm} classes) trained on different --data than what you passed ({nc} ' \
f'classes). Pass correct combination of --weights and --data that are trained together.'
model.warmup(imgsz=(1 if pt else batch_size, 3, imgsz, imgsz)) # warmup
pad = 0.0 if task in ('speed', 'benchmark') else 0.5
rect = False if task == 'benchmark' else pt # square inference for benchmarks
task = task if task in ('train', 'val', 'test') else 'val' # path to train/val/test images
dataloader = create_dataloader(data[task],
imgsz,
batch_size,
stride,
single_cls,
pad=pad,
rect=rect,
workers=workers,
prefix=colorstr(f'{task}: '))[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)}
class_map = coco80_to_coco91_class() if is_coco else list(range(1000))
s = ('%20s' + '%11s' * 6) % ('Class', 'Images', 'Labels', 'P', 'R', '[email protected]', '[email protected]:.95')
dt, p, r, f1, mp, mr, map50, map = [0.0, 0.0, 0.0], 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0
loss = torch.zeros(3, device=device)
jdict, stats, ap, ap_class = [], [], [], []
callbacks.run('on_val_start')
pbar = tqdm(dataloader, desc=s, bar_format='{l_bar}{bar:10}{r_bar}{bar:-10b}') # progress bar
for batch_i, (im, targets, paths, shapes) in enumerate(pbar):
callbacks.run('on_val_batch_start')
t1 = time_sync()
if cuda:
im = im.to(device, non_blocking=True)
targets = targets.to(device)
im = im.half() if half else im.float() # uint8 to fp16/32
im /= 255 # 0 - 255 to 0.0 - 1.0
nb, _, height, width = im.shape # batch size, channels, height, width
t2 = time_sync()
dt[0] += t2 - t1
# Inference
out, train_out = model(im) if training else model(im, augment=augment, val=True) # inference, loss outputs
dt[1] += time_sync() - t2
# Loss
if compute_loss:
loss += compute_loss([x.float() for x in train_out], targets)[1] # box, obj, cls
# NMS
targets[:, 2:] *= torch.tensor((width, height, width, height), device=device) # to pixels
lb = [targets[targets[:, 0] == i, 1:] for i in range(nb)] if save_hybrid else [] # for autolabelling
t3 = time_sync()
out = non_max_suppression(out, conf_thres, iou_thres, labels=lb, multi_label=True, agnostic=single_cls)
dt[2] += time_sync() - t3
# Metrics
for si, pred in enumerate(out):
labels = targets[targets[:, 0] == si, 1:]
nl, npr = labels.shape[0], pred.shape[0] # number of labels, predictions
path, shape = Path(paths[si]), shapes[si][0]
correct = torch.zeros(npr, niou, dtype=torch.bool, device=device) # init
seen += 1
if npr == 0:
if nl:
stats.append((correct, *torch.zeros((3, 0), device=device)))
continue
# Predictions
if single_cls:
pred[:, 5] = 0
predn = pred.clone()
scale_coords(im[si].shape[1:], predn[:, :4], shape, shapes[si][1]) # native-space pred
# Evaluate
if nl:
tbox = xywh2xyxy(labels[:, 1:5]) # target boxes
scale_coords(im[si].shape[1:], tbox, shape, shapes[si][1]) # native-space labels
labelsn = torch.cat((labels[:, 0:1], tbox), 1) # native-space labels
correct = process_batch(predn, labelsn, iouv)
if plots:
confusion_matrix.process_batch(predn, labelsn)
stats.append((correct, pred[:, 4], pred[:, 5], labels[:, 0])) # (correct, conf, pcls, tcls)
# Save/log
if save_txt:
save_one_txt(predn, save_conf, shape, file=save_dir / 'labels' / (path.stem + '.txt'))
if save_json:
save_one_json(predn, jdict, path, class_map) # append to COCO-JSON dictionary
callbacks.run('on_val_image_end', pred, predn, path, names, im[si])
# Plot images
if plots and batch_i < 3:
plot_images(im, targets, paths, save_dir / f'val_batch{batch_i}_labels.jpg', names) # labels
plot_images(im, output_to_target(out), paths, save_dir / f'val_batch{batch_i}_pred.jpg', names) # pred
callbacks.run('on_val_batch_end')
# Compute metrics
stats = [torch.cat(x, 0).cpu().numpy() for x in zip(*stats)] # to numpy
if len(stats) and stats[0].any():
tp, fp, p, r, f1, ap, ap_class = ap_per_class(*stats, plot=plots, save_dir=save_dir, names=names)
ap50, ap = ap[:, 0], ap.mean(1) # [email protected], [email protected]:0.95
mp, mr, map50, map = p.mean(), r.mean(), ap50.mean(), ap.mean()
nt = np.bincount(stats[3].astype(int), minlength=nc) # number of targets per class
else:
nt = torch.zeros(1)
# Print results
pf = '%20s' + '%11i' * 2 + '%11.3g' * 4 # print format
LOGGER.info(pf % ('all', seen, nt.sum(), mp, mr, map50, map))
# Print results per class
if (verbose or (nc < 50 and not training)) and nc > 1 and len(stats):
for i, c in enumerate(ap_class):
LOGGER.info(pf % (names[c], seen, nt[c], p[i], r[i], ap50[i], ap[i]))
# Print speeds
t = tuple(x / seen * 1E3 for x in dt) # speeds per image
if not training:
shape = (batch_size, 3, imgsz, imgsz)
LOGGER.info(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {shape}' % t)
# Plots
if plots:
confusion_matrix.plot(save_dir=save_dir, names=list(names.values()))
callbacks.run('on_val_end')
# 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 = str(Path(data.get('path', '../coco')) / 'annotations/instances_val2017.json') # annotations json
pred_json = str(save_dir / f"{w}_predictions.json") # predictions json
LOGGER.info(f'\nEvaluating pycocotools mAP... saving {pred_json}...')
with open(pred_json, 'w') as f:
json.dump(jdict, f)
try: # https://github.com/cocodataset/cocoapi/blob/master/PythonAPI/pycocoEvalDemo.ipynb
check_requirements(['pycocotools'])
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.im_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:
LOGGER.info(f'pycocotools unable to run: {e}')
# Return results
model.float() # for training
if not training:
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}")
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
def train(hyp, opt, device, tb_writer=None):
train_batches_uploaded = False
logger.info('hyperparameters', extra=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.safe_dump(hyp, f, sort_keys=False)
with open(save_dir / 'opt.yaml', 'w') as f:
yaml.safe_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.safe_load(f) # data dict
is_coco = opt.data.endswith('coco.yaml')
# Logging- Doing this before checking the dataset. Might update data_dict
loggers = {'wandb': None} # loggers dict
if rank in [-1, 0]:
opt.hyp = hyp # add hyperparameters
run_id = torch.load(weights).get('wandb_id') if weights.endswith('.pt') and os.path.isfile(weights) else None
wandb_logger = WandbLogger(opt, save_dir.stem, run_id, data_dict)
loggers['wandb'] = wandb_logger.wandb
data_dict = wandb_logger.data_dict
if wandb_logger.wandb:
weights, epochs, hyp = opt.weights, opt.epochs, opt.hyp # WandbLogger might update weights, epochs if resuming
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
colors = data_dict.get('colors', None)
# Model
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
model = Model(opt.cfg or ckpt['model'].yaml, ch=3, nc=nc, anchors=hyp.get('anchors')).to(device) # create
exclude = ['anchor'] if (opt.cfg or hyp.get('anchors')) and not opt.resume 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, anchors=hyp.get('anchors')).to(device) # create
with torch_distributed_zero_first(rank):
check_dataset(data_dict) # check
train_path = data_dict['train']
test_path = data_dict['val']
# 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
logger.info(f"Scaled weight_decay = {hyp['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
if opt.linear_lr:
lf = lambda x: (1 - x / (epochs - 1)) * (1.0 - hyp['lrf']) + hyp['lrf'] # linear
else:
lf = one_cycle(1, hyp['lrf'], epochs) # cosine 1->hyp['lrf']
scheduler = lr_scheduler.LambdaLR(optimizer, lr_lambda=lf)
# plot_lr_scheduler(optimizer, scheduler, epochs)
# EMA
ema = ModelEMA(model) if rank in [-1, 0] else None
# 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']
# EMA
if ema and ckpt.get('ema'):
ema.ema.load_state_dict(ckpt['ema'].float().state_dict())
ema.updates = ckpt['updates']
# Results
if ckpt.get('training_results') is not None:
results_file.write_text(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 = max(int(model.stride.max()), 32) # grid size (max stride)
nl = model.model[-1].nl # number of detection layers (used for scaling hyp['obj'])
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()')
# 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, quad=opt.quad, 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]:
testloader = create_dataloader(test_path, imgsz_test, batch_size * 2, 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_labels(labels, names, save_dir, loggers)
if opt.sly:
upload_label_vis()
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.half().float() # pre-reduce anchor precision
# DDP mode
if cuda and rank != -1:
model = DDP(model, device_ids=[opt.local_rank], output_device=opt.local_rank,
# nn.MultiheadAttention incompatibility with DDP https://github.com/pytorch/pytorch/issues/26698
find_unused_parameters=any(isinstance(layer, nn.MultiheadAttention) for layer in model.modules()))
# Model parameters
hyp['box'] *= 3. / nl # scale to layers
hyp['cls'] *= nc / 80. * 3. / nl # scale to classes and layers
hyp['obj'] *= (imgsz / 640) ** 2 * 3. / nl # scale to image size and layers
hyp['label_smoothing'] = opt.label_smoothing
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
model.colors = colors
model.img_size = opt.img_size
# 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)
compute_loss = ComputeLoss(model) # init loss class
logger.info(f'Image sizes {imgsz} train, {imgsz_test} test\n'
f'Using {dataloader.num_workers} dataloader workers\n'
f'Logging results to {save_dir}\n'
f'Starting training for {epochs} epochs...')
sly_epoch_progress = sly.Progress("Epoch", 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(4, device=device) # mean losses
if rank != -1:
dataloader.sampler.set_epoch(epoch)
pbar = enumerate(dataloader)
#logger.info(('\n' + '%10s' * 8) % ('Epoch', 'gpu_mem', 'box', 'obj', 'cls', 'total', 'labels', 'img_size'))
if rank in [-1, 0]:
#pbar = tqdm(pbar, total=nb) # progress bar
pass
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)) # loss scaled by batch_size
if rank != -1:
loss *= opt.world_size # gradient averaged between devices in DDP mode
if opt.quad:
loss *= 4.
# 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' * 6) % (
'%g/%g' % (epoch, epochs - 1), mem, *mloss, targets.shape[0], imgs.shape[-1])
#pbar.set_description(s)
#'Epoch', 'gpu_mem', 'box', 'obj', 'cls', 'total', 'labels', 'img_size'
logger.info("Training", extra={
"epoch": epoch,
"epochs_count": epochs - 1,
"gpu_mem": mem,
"mbox_loss": float(mloss[0].cpu().numpy()),
"mobj_loss": float(mloss[1].cpu().numpy()),
"mcls_loss": float(mloss[2].cpu().numpy()),
"mtotal_loss": float(mloss[3].cpu().numpy()),
"labels": targets.shape[0],
"img_size": imgs.shape[-1]
})
# Plot
if plots and ni < 3:
f = save_dir / f'train_batch{ni}.jpg' # filename
plot_images(imgs, targets, paths, f)
#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(torch.jit.trace(model, imgs, strict=False), []) # add model graph
elif plots and ni == 10 and wandb_logger.wandb:
wandb_logger.log({"Mosaics": [wandb_logger.wandb.Image(str(x), caption=x.name) for x in
save_dir.glob('train*.jpg') if x.exists()]})
if plots and ni == 10 and opt.sly:
train_batches_uploaded = True
upload_train_data_vis()
# 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]:
# mAP
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
wandb_logger.current_epoch = epoch + 1
results, maps, times = test.test(data_dict,
batch_size=batch_size * 2,
imgsz=imgsz_test,
model=ema.ema,
single_cls=opt.single_cls,
dataloader=testloader,
save_dir=save_dir,
verbose=nc < 50 and final_epoch,
plots=plots and final_epoch,
wandb_logger=wandb_logger,
compute_loss=compute_loss,
is_coco=is_coco,
opt_sly=opt.sly)
# Write
with open(results_file, 'a') as f:
f.write(s + '%10.4g' * 7 % results + '\n') # append metrics, val_loss
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
if opt.sly:
send_epoch_log(epoch, epochs, sly_epoch_progress)
metrics = {}
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_logger.wandb:
wandb_logger.log({tag: x}) # W&B
if opt.sly:
if torch.is_tensor(x):
x = float(x.cpu().numpy())
metrics[tag] = x
if opt.sly:
send_metrics(epoch, epochs, metrics, opt.metrics_period)
# 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
wandb_logger.end_epoch(best_result=best_fitness == fi)
# Save model
if (not opt.nosave) or (final_epoch and not opt.evolve): # if save
model.img_size = opt.img_size
ema.ema.img_size = opt.img_size
ckpt = {'epoch': epoch,
'best_fitness': best_fitness,
'training_results': results_file.read_text(),
'model': deepcopy(model.module if is_parallel(model) else model).half(),
'ema': deepcopy(ema.ema).half(),
'updates': ema.updates,
'optimizer': optimizer.state_dict(),
'wandb_id': wandb_logger.wandb_run.id if wandb_logger.wandb else None}
# Save last, best and delete
torch.save(ckpt, last)
if best_fitness == fi:
torch.save(ckpt, best)
if wandb_logger.wandb:
if ((epoch + 1) % opt.save_period == 0 and not final_epoch) and opt.save_period != -1:
wandb_logger.log_model(
last.parent, opt, epoch, fi, best_model=best_fitness == fi)
del ckpt
# end epoch ----------------------------------------------------------------------------------------------------
if plots and opt.sly and train_batches_uploaded is False:
train_batches_uploaded = True
upload_train_data_vis()
# end training
if rank in [-1, 0]:
# Plots
if plots:
plot_results(save_dir=save_dir) # save as results.png
if wandb_logger.wandb:
files = ['results.png', 'confusion_matrix.png', *[f'{x}_curve.png' for x in ('F1', 'PR', 'P', 'R')]]
wandb_logger.log({"Results": [wandb_logger.wandb.Image(str(save_dir / f), caption=f) for f in files
if (save_dir / f).exists()]})
# Test best.pt
logger.info('%g epochs completed in %.3f hours.\n' % (epoch - start_epoch + 1, (time.time() - t0) / 3600))
if opt.data.endswith('coco.yaml') and nc == 80: # if COCO
for m in (last, best) if best.exists() else (last): # speed, mAP tests
results, _, _ = test.test(opt.data,
batch_size=batch_size * 2,
imgsz=imgsz_test,
conf_thres=0.001,
iou_thres=0.7,
model=attempt_load(m, device).half(),
single_cls=opt.single_cls,
dataloader=testloader,
save_dir=save_dir,
save_json=True,
plots=False,
is_coco=is_coco)
# 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
if wandb_logger.wandb and not opt.evolve: # Log the stripped model
wandb_logger.wandb.log_artifact(str(final), type='model',
name='run_' + wandb_logger.wandb_run.id + '_model',
aliases=['last', 'best', 'stripped'])
wandb_logger.finish_run()
else:
dist.destroy_process_group()
torch.cuda.empty_cache()
return results
