def model_type(p='path/to/model.pt'):
# Return model type from model path, i.e. path='path/to/model.onnx' -> type=onnx
from export import export_formats
suffixes = list(export_formats().Suffix) + ['.xml'] # export suffixes
check_suffix(p, suffixes) # checks
p = Path(p).name # eliminate trailing separators
pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, xml2 = (
s in p for s in suffixes)
xml |= xml2 # *_openvino_model or *.xml
tflite &= not edgetpu # *.tflite
return pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs
def __init__(self,
weights='yolov5s.pt',
device=None,
dnn=False,
data=None):
# Usage:
# PyTorch: weights = *.pt
# TorchScript: *.torchscript
# CoreML: *.mlmodel
# OpenVINO: *.xml
# TensorFlow: *_saved_model
# TensorFlow: *.pb
# TensorFlow Lite: *.tflite
# TensorFlow Edge TPU: *_edgetpu.tflite
# ONNX Runtime: *.onnx
# OpenCV DNN: *.onnx with dnn=True
# TensorRT: *.engine
from models.experimental import attempt_download, attempt_load # scoped to avoid circular import
super().__init__()
w = str(weights[0] if isinstance(weights, list) else weights)
suffix = Path(w).suffix.lower()
suffixes = [
'.pt', '.torchscript', '.onnx', '.engine', '.tflite', '.pb', '',
'.mlmodel', '.xml'
]
check_suffix(w, suffixes) # check weights have acceptable suffix
pt, jit, onnx, engine, tflite, pb, saved_model, coreml, xml = (
suffix == x for x in suffixes) # backends
stride, names = 64, [f'class{i}'
for i in range(1000)] # assign defaults
w = attempt_download(w) # download if not local
if data: # data.yaml path (optional)
with open(data, errors='ignore') as f:
names = yaml.safe_load(f)['names'] # class names
if pt: # PyTorch
model = attempt_load(weights if isinstance(weights, list) else w,
map_location=device)
stride = max(int(model.stride.max()), 32) # model stride
names = model.module.names if hasattr(
model, 'module') else model.names # get class names
self.model = model # explicitly assign for to(), cpu(), cuda(), half()
elif jit: # TorchScript
LOGGER.info(f'Loading {w} for TorchScript inference...')
extra_files = {'config.txt': ''} # model metadata
model = torch.jit.load(w, _extra_files=extra_files)
if extra_files['config.txt']:
d = json.loads(extra_files['config.txt']) # extra_files dict
stride, names = int(d['stride']), d['names']
elif dnn: # ONNX OpenCV DNN
LOGGER.info(f'Loading {w} for ONNX OpenCV DNN inference...')
check_requirements(('opencv-python>=4.5.4', ))
net = cv2.dnn.readNetFromONNX(w)
elif onnx: # ONNX Runtime
LOGGER.info(f'Loading {w} for ONNX Runtime inference...')
cuda = torch.cuda.is_available()
check_requirements(
('onnx', 'onnxruntime-gpu' if cuda else 'onnxruntime'))
import onnxruntime
providers = ['CUDAExecutionProvider', 'CPUExecutionProvider'
] if cuda else ['CPUExecutionProvider']
session = onnxruntime.InferenceSession(w, providers=providers)
elif xml: # OpenVINO
LOGGER.info(f'Loading {w} for OpenVINO inference...')
check_requirements(
('openvino-dev', )
) # requires openvino-dev: https://pypi.org/project/openvino-dev/
import openvino.inference_engine as ie
core = ie.IECore()
network = core.read_network(
model=w,
weights=Path(w).with_suffix('.bin')) # *.xml, *.bin paths
executable_network = core.load_network(network,
device_name='CPU',
num_requests=1)
elif engine: # TensorRT
LOGGER.info(f'Loading {w} for TensorRT inference...')
import tensorrt as trt # https://developer.nvidia.com/nvidia-tensorrt-download
check_version(trt.__version__, '7.0.0',
hard=True) # require tensorrt>=7.0.0
Binding = namedtuple('Binding',
('name', 'dtype', 'shape', 'data', 'ptr'))
logger = trt.Logger(trt.Logger.INFO)
with open(w, 'rb') as f, trt.Runtime(logger) as runtime:
model = runtime.deserialize_cuda_engine(f.read())
bindings = OrderedDict()
for index in range(model.num_bindings):
name = model.get_binding_name(index)
dtype = trt.nptype(model.get_binding_dtype(index))
shape = tuple(model.get_binding_shape(index))
data = torch.from_numpy(np.empty(
shape, dtype=np.dtype(dtype))).to(device)
bindings[name] = Binding(name, dtype, shape, data,
int(data.data_ptr()))
binding_addrs = OrderedDict(
(n, d.ptr) for n, d in bindings.items())
context = model.create_execution_context()
batch_size = bindings['images'].shape[0]
elif coreml: # CoreML
LOGGER.info(f'Loading {w} for CoreML inference...')
import coremltools as ct
model = ct.models.MLModel(w)
else: # TensorFlow (SavedModel, GraphDef, Lite, Edge TPU)
if saved_model: # SavedModel
LOGGER.info(
f'Loading {w} for TensorFlow SavedModel inference...')
import tensorflow as tf
model = tf.keras.models.load_model(w)
elif pb: # GraphDef https://www.tensorflow.org/guide/migrate#a_graphpb_or_graphpbtxt
LOGGER.info(
f'Loading {w} for TensorFlow GraphDef inference...')
import tensorflow as tf
def wrap_frozen_graph(gd, inputs, outputs):
x = tf.compat.v1.wrap_function(
lambda: tf.compat.v1.import_graph_def(gd, name=""),
[]) # wrapped
return x.prune(
tf.nest.map_structure(x.graph.as_graph_element,
inputs),
tf.nest.map_structure(x.graph.as_graph_element,
outputs))
graph_def = tf.Graph().as_graph_def()
graph_def.ParseFromString(open(w, 'rb').read())
frozen_func = wrap_frozen_graph(gd=graph_def,
inputs="x:0",
outputs="Identity:0")
elif tflite: # https://www.tensorflow.org/lite/guide/python#install_tensorflow_lite_for_python
try:
import tflite_runtime.interpreter as tfl # prefer tflite_runtime if installed
except ImportError:
import tensorflow.lite as tfl
if 'edgetpu' in w.lower(
): # Edge TPU https://coral.ai/software/#edgetpu-runtime
LOGGER.info(
f'Loading {w} for TensorFlow Lite Edge TPU inference...'
)
delegate = {
'Linux': 'libedgetpu.so.1',
'Darwin': 'libedgetpu.1.dylib',
'Windows': 'edgetpu.dll'
}[platform.system()]
interpreter = tfl.Interpreter(
model_path=w,
experimental_delegates=[tfl.load_delegate(delegate)])
else: # Lite
LOGGER.info(
f'Loading {w} for TensorFlow Lite inference...')
interpreter = tfl.Interpreter(
model_path=w) # load TFLite model
interpreter.allocate_tensors() # allocate
input_details = interpreter.get_input_details() # inputs
output_details = interpreter.get_output_details() # outputs
self.__dict__.update(locals()) # assign all variables to self
def run(
weights=ROOT / 'yolov5s.pt', # model.pt path(s)
source=ROOT / 'data/images', # file/dir/URL/glob, 0 for webcam
imgsz=640, # inference size (pixels)
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/detect', # 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
):
source = str(source)
save_img = not nosave and not source.endswith(
'.txt') # save inference images
webcam = source.isnumeric() or source.endswith(
'.txt') or source.lower().startswith(
('rtsp://', 'rtmp://', 'http://', 'https://'))
# 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
# Initialize
set_logging()
device = select_device(device)
half &= device.type != 'cpu' # half precision only supported on CUDA
# Load model
w = str(weights[0] if isinstance(weights, list) else weights)
classify, suffix, suffixes = False, Path(w).suffix.lower(), [
'.pt', '.onnx', '.tflite', '.pb', ''
]
check_suffix(w, suffixes) # check weights have acceptable suffix
pt, onnx, tflite, pb, saved_model = (suffix == x
for x in suffixes) # backend booleans
stride, names = 64, [f'class{i}' for i in range(1000)] # assign defaults
if pt:
model = torch.jit.load(w) if 'torchscript' in w else attempt_load(
weights, map_location=device, fuse=False)
stride = int(model.stride.max()) # model stride
names = model.module.names if hasattr(
model, 'module') else model.names # get class names
"""
for _, param in enumerate(model.named_parameters()):
print("====>", param[0], param[1].shape)
torch.save(model.state_dict(), 'new_params.pt')
for k, v in model.state_dict().items():
print(k, v.shape)
exit()
"""
if half:
model.half() # to FP16
if classify: # second-stage classifier
modelc = load_classifier(name='resnet50', n=2) # initialize
modelc.load_state_dict(
torch.load('resnet50.pt',
map_location=device)['model']).to(device).eval()
elif onnx:
if dnn:
# check_requirements(('opencv-python>=4.5.4',))
net = cv2.dnn.readNetFromONNX(w)
else:
check_requirements(('onnx', 'onnxruntime'))
import onnxruntime
session = onnxruntime.InferenceSession(w, None)
else: # TensorFlow models
check_requirements(('tensorflow>=2.4.1', ))
import tensorflow as tf
if pb: # https://www.tensorflow.org/guide/migrate#a_graphpb_or_graphpbtxt
def wrap_frozen_graph(gd, inputs, outputs):
x = tf.compat.v1.wrap_function(
lambda: tf.compat.v1.import_graph_def(gd, name=""),
[]) # wrapped import
return x.prune(
tf.nest.map_structure(x.graph.as_graph_element, inputs),
tf.nest.map_structure(x.graph.as_graph_element, outputs))
graph_def = tf.Graph().as_graph_def()
graph_def.ParseFromString(open(w, 'rb').read())
frozen_func = wrap_frozen_graph(gd=graph_def,
inputs="x:0",
outputs="Identity:0")
elif saved_model:
model = tf.keras.models.load_model(w)
elif tflite:
interpreter = tf.lite.Interpreter(
model_path=w) # load TFLite model
interpreter.allocate_tensors() # allocate
input_details = interpreter.get_input_details() # inputs
output_details = interpreter.get_output_details() # outputs
int8 = input_details[0][
'dtype'] == np.uint8 # is TFLite quantized uint8 model
imgsz = check_img_size(imgsz, s=stride) # check image size
# Dataloader
if webcam:
view_img = check_imshow()
cudnn.benchmark = True # set True to speed up constant image size inference
dataset = LoadStreams(source, img_size=imgsz, stride=stride, auto=pt)
bs = len(dataset) # batch_size
else:
dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt)
bs = 1 # batch_size
vid_path, vid_writer = [None] * bs, [None] * bs
# Run inference
if pt and device.type != 'cpu':
model(
torch.zeros(1, 3, *imgsz).to(device).type_as(
next(model.parameters()))) # run once
dt, seen = [0.0, 0.0, 0.0], 0
for path, img, im0s, vid_cap in dataset:
t1 = time_sync()
if onnx:
img = img.astype('float32')
else:
img = torch.from_numpy(img).to(device)
img = img.half() if half else img.float() # uint8 to fp16/32
img = img / 255.0 # 0 - 255 to 0.0 - 1.0
if len(img.shape) == 3:
img = img[None] # expand for batch dim
t2 = time_sync()
dt[0] += t2 - t1
# Inference
if pt:
visualize = increment_path(save_dir / Path(path).stem,
mkdir=True) if visualize else False
pred = model(img, augment=augment, visualize=visualize)[0]
anchor_grid = model.model[-1].anchors * model.model[-1].stride[
..., None, None]
delattr(model.model[-1],
'anchor_grid') # model.model[-1] is detect layer
model.model[-1].register_buffer("anchor_grid", anchor_grid)
model.to(device).eval()
wts_file = "generated.wts"
with open(wts_file, 'w') as f:
f.write('{}\n'.format(len(model.state_dict().keys())))
for k, v in model.state_dict().items():
if len(v.shape) == 0:
continue
print(k, v.shape)
vr = v.reshape(-1).cpu().numpy()
f.write('{} {} {} {}'.format(
k, len(vr), v.shape[0],
v.shape[1] if len(v.shape) > 1 else 0))
for vv in vr:
f.write(' ')
f.write(struct.pack('>f', float(vv)).hex())
f.write('\n')
exit()
elif onnx:
if dnn:
net.setInput(img)
pred = torch.tensor(net.forward())
else:
pred = torch.tensor(
session.run([session.get_outputs()[0].name],
{session.get_inputs()[0].name: img}))
else: # tensorflow model (tflite, pb, saved_model)
imn = img.permute(0, 2, 3, 1).cpu().numpy() # image in numpy
if pb:
pred = frozen_func(x=tf.constant(imn)).numpy()
elif saved_model:
pred = model(imn, training=False).numpy()
elif tflite:
if int8:
scale, zero_point = input_details[0]['quantization']
imn = (imn / scale + zero_point).astype(
np.uint8) # de-scale
interpreter.set_tensor(input_details[0]['index'], imn)
interpreter.invoke()
pred = interpreter.get_tensor(output_details[0]['index'])
if int8:
scale, zero_point = output_details[0]['quantization']
pred = (pred.astype(np.float32) -
zero_point) * scale # re-scale
pred[..., 0] *= imgsz[1] # x
pred[..., 1] *= imgsz[0] # y
pred[..., 2] *= imgsz[1] # w
pred[..., 3] *= imgsz[0] # h
pred = torch.tensor(pred)
t3 = time_sync()
dt[1] += t3 - t2
# NMS
pred = non_max_suppression(pred,
conf_thres,
iou_thres,
classes,
agnostic_nms,
max_det=max_det)
dt[2] += time_sync() - t3
# Second-stage classifier (optional)
if classify:
pred = apply_classifier(pred, modelc, img, im0s)
# Process predictions
for i, det in enumerate(pred): # per image
seen += 1
if webcam: # batch_size >= 1
p, s, im0, frame = path[i], f'{i}: ', im0s[i].copy(
), dataset.count
else:
p, s, im0, frame = path, '', im0s.copy(), getattr(
dataset, 'frame', 0)
p = Path(p) # to Path
save_path = str(save_dir / p.name) # img.jpg
txt_path = str(save_dir / 'labels' / p.stem) + (
'' if dataset.mode == 'image' else f'_{frame}') # img.txt
s += '%gx%g ' % img.shape[2:] # print string
gn = torch.tensor(im0.shape)[[1, 0, 1,
0]] # normalization gain whwh
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_coords(img.shape[2:], det[:, :4],
im0.shape).round()
# Print results
for c in det[:, -1].unique():
n = (det[:, -1] == c).sum() # detections per class
s += f"{n} {names[int(c)]}{'s' * (n > 1)}, " # add to string
# Write results
for *xyxy, conf, cls in reversed(det):
if save_txt: # Write to file
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(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))
if save_crop:
save_one_box(xyxy,
imc,
file=save_dir / 'crops' / names[c] /
f'{p.stem}.jpg',
BGR=True)
# Print time (inference-only)
print(f'{s}Done. ({t3 - t2:.3f}s)')
# Stream results
im0 = annotator.result()
if view_img:
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 += '.mp4'
vid_writer[i] = cv2.VideoWriter(
save_path, cv2.VideoWriter_fourcc(*'mp4v'), fps,
(w, h))
vid_writer[i].write(im0)
# Print results
t = tuple(x / seen * 1E3 for x in dt) # speeds per image
print(
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 ''
print(f"Results saved to {colorstr('bold', save_dir)}{s}")
if update:
strip_optimizer(weights) # update model (to fix SourceChangeWarning)
def train(
hyp, # path/to/hyp.yaml or hyp dictionary
opt,
device,
callbacks):
save_dir, epochs, batch_size, weights, single_cls, evolve, data, cfg, resume, noval, nosave, workers, freeze = \
Path(opt.save_dir), opt.epochs, opt.batch_size, opt.weights, opt.single_cls, opt.evolve, opt.data, opt.cfg, \
opt.resume, opt.noval, opt.nosave, opt.workers, opt.freeze
# Directories
w = save_dir / 'weights' # weights dir
(w.parent if evolve else w).mkdir(parents=True, exist_ok=True) # make dir
last, best = w / 'last.pt', w / 'best.pt'
# Hyperparameters
if isinstance(hyp, str):
with open(hyp, errors='ignore') as f:
hyp = yaml.safe_load(f) # load hyps dict
LOGGER.info(
colorstr('hyperparameters: ') + ', '.join(f'{k}={v}'
for k, v in hyp.items()))
# Save run settings
if not evolve:
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)
# Loggers
data_dict = None
if RANK in [-1, 0]:
loggers = Loggers(save_dir, weights, opt, hyp,
LOGGER) # loggers instance
if loggers.wandb:
data_dict = loggers.wandb.data_dict
if resume:
weights, epochs, hyp, batch_size = opt.weights, opt.epochs, opt.hyp, opt.batch_size
# Register actions
for k in methods(loggers):
callbacks.register_action(k, callback=getattr(loggers, k))
# Config
plots = not evolve # create plots
cuda = device.type != 'cpu'
init_seeds(1 + RANK)
with torch_distributed_zero_first(LOCAL_RANK):
data_dict = data_dict or check_dataset(data) # check if None
train_path, val_path = data_dict['train'], data_dict['val']
nc = 1 if single_cls else int(data_dict['nc']) # number of classes
names = ['item'] if single_cls and len(
data_dict['names']) != 1 else data_dict['names'] # class names
assert len(
names
) == nc, f'{len(names)} names found for nc={nc} dataset in {data}' # check
is_coco = isinstance(val_path, str) and val_path.endswith(
'coco/val2017.txt') # COCO dataset
# Model
check_suffix(weights, '.pt') # check weights
pretrained = weights.endswith('.pt')
if pretrained:
with torch_distributed_zero_first(LOCAL_RANK):
weights = attempt_download(
weights) # download if not found locally
ckpt = torch.load(weights, map_location='cpu'
) # load checkpoint to CPU to avoid CUDA memory leak
model = Model(cfg or ckpt['model'].yaml,
ch=3,
nc=nc,
anchors=hyp.get('anchors')).to(device) # create
exclude = [
'anchor'
] if (cfg or hyp.get('anchors')) and not resume else [] # exclude keys
csd = ckpt['model'].float().state_dict(
) # checkpoint state_dict as FP32
csd = intersect_dicts(csd, model.state_dict(),
exclude=exclude) # intersect
model.load_state_dict(csd, strict=False) # load
LOGGER.info(
f'Transferred {len(csd)}/{len(model.state_dict())} items from {weights}'
) # report
else:
model = Model(cfg, ch=3, nc=nc,
anchors=hyp.get('anchors')).to(device) # create
# Freeze
freeze = [
f'model.{x}.'
for x in (freeze if len(freeze) > 1 else range(freeze[0]))
] # layers to freeze
for k, v in model.named_parameters():
v.requires_grad = True # train all layers
if any(x in k for x in freeze):
LOGGER.info(f'freezing {k}')
v.requires_grad = False
# Image size
gs = max(int(model.stride.max()), 32) # grid size (max stride)
imgsz = check_img_size(opt.imgsz, gs,
floor=gs * 2) # verify imgsz is gs-multiple
# Batch size
if RANK == -1 and batch_size == -1: # single-GPU only, estimate best batch size
batch_size = check_train_batch_size(model, imgsz)
loggers.on_params_update({"batch_size": batch_size})
# Optimizer
nbs = 64 # nominal batch size
accumulate = max(round(nbs / batch_size),
1) # accumulate loss before optimizing
hyp['weight_decay'] *= batch_size * accumulate / nbs # scale weight_decay
LOGGER.info(f"Scaled weight_decay = {hyp['weight_decay']}")
g0, g1, g2 = [], [], [] # optimizer parameter groups
for v in model.modules():
if hasattr(v, 'bias') and isinstance(v.bias, nn.Parameter): # bias
g2.append(v.bias)
if isinstance(v, nn.BatchNorm2d): # weight (no decay)
g0.append(v.weight)
elif hasattr(v, 'weight') and isinstance(
v.weight, nn.Parameter): # weight (with decay)
g1.append(v.weight)
if opt.optimizer == 'Adam':
optimizer = Adam(g0, lr=hyp['lr0'],
betas=(hyp['momentum'],
0.999)) # adjust beta1 to momentum
elif opt.optimizer == 'AdamW':
optimizer = AdamW(g0, lr=hyp['lr0'],
betas=(hyp['momentum'],
0.999)) # adjust beta1 to momentum
else:
optimizer = SGD(g0,
lr=hyp['lr0'],
momentum=hyp['momentum'],
nesterov=True)
optimizer.add_param_group({
'params': g1,
'weight_decay': hyp['weight_decay']
}) # add g1 with weight_decay
optimizer.add_param_group({'params': g2}) # add g2 (biases)
LOGGER.info(
f"{colorstr('optimizer:')} {type(optimizer).__name__} with parameter groups "
f"{len(g0)} weight (no decay), {len(g1)} weight, {len(g2)} bias")
del g0, g1, g2
# Scheduler
if opt.cos_lr:
lf = one_cycle(1, hyp['lrf'], epochs) # cosine 1->hyp['lrf']
else:
lf = lambda x: (1 - x / epochs) * (1.0 - hyp['lrf']) + hyp['lrf'
] # linear
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']
# Epochs
start_epoch = ckpt['epoch'] + 1
if resume:
assert start_epoch > 0, f'{weights} training to {epochs} epochs is finished, nothing to resume.'
if epochs < start_epoch:
LOGGER.info(
f"{weights} has been trained for {ckpt['epoch']} epochs. Fine-tuning for {epochs} more epochs."
)
epochs += ckpt['epoch'] # finetune additional epochs
del ckpt, csd
# DP mode
if cuda and RANK == -1 and torch.cuda.device_count() > 1:
LOGGER.warning(
'WARNING: DP not recommended, use torch.distributed.run for best DDP Multi-GPU results.\n'
'See Multi-GPU Tutorial at https://github.com/ultralytics/yolov5/issues/475 to get started.'
)
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
train_loader, dataset = create_dataloader(
train_path,
imgsz,
batch_size // WORLD_SIZE,
gs,
single_cls,
hyp=hyp,
augment=True,
cache=None if opt.cache == 'val' else opt.cache,
rect=opt.rect,
rank=LOCAL_RANK,
workers=workers,
image_weights=opt.image_weights,
quad=opt.quad,
prefix=colorstr('train: '),
shuffle=True)
mlc = int(np.concatenate(dataset.labels, 0)[:, 0].max()) # max label class
nb = len(train_loader) # number of batches
assert mlc < nc, f'Label class {mlc} exceeds nc={nc} in {data}. Possible class labels are 0-{nc - 1}'
# Process 0
if RANK in [-1, 0]:
val_loader = create_dataloader(val_path,
imgsz,
batch_size // WORLD_SIZE * 2,
gs,
single_cls,
hyp=hyp,
cache=None if noval else opt.cache,
rect=True,
rank=-1,
workers=workers * 2,
pad=0.5,
prefix=colorstr('val: '))[0]
if not 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)
# Anchors
if not opt.noautoanchor:
check_anchors(dataset,
model=model,
thr=hyp['anchor_t'],
imgsz=imgsz)
model.half().float() # pre-reduce anchor precision
callbacks.run('on_pretrain_routine_end')
# DDP mode
if cuda and RANK != -1:
model = DDP(model, device_ids=[LOCAL_RANK], output_device=LOCAL_RANK)
# Model attributes
nl = de_parallel(
model).model[-1].nl # number of detection layers (to scale hyps)
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.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
last_opt_step = -1
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)
stopper = EarlyStopping(patience=opt.patience)
compute_loss = ComputeLoss(model) # init loss class
LOGGER.info(
f'Image sizes {imgsz} train, {imgsz} val\n'
f'Using {train_loader.num_workers * WORLD_SIZE} dataloader workers\n'
f"Logging results to {colorstr('bold', save_dir)}\n"
f'Starting training for {epochs} epochs...')
for epoch in range(
start_epoch, epochs
): # epoch ------------------------------------------------------------------
model.train()
# Update image weights (optional, single-GPU only)
if opt.image_weights:
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
# Update mosaic border (optional)
# 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(3, device=device) # mean losses
if RANK != -1:
train_loader.sampler.set_epoch(epoch)
pbar = enumerate(train_loader)
LOGGER.info(
('\n' + '%10s' * 7) %
('Epoch', 'gpu_mem', 'box', 'obj', 'cls', 'labels', 'img_size'))
if RANK in [-1, 0]:
pbar = tqdm(
pbar, total=nb,
bar_format='{l_bar}{bar:10}{r_bar}{bar:-10b}') # 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 # uint8 to float32, 0-255 to 0.0-1.0
# Warmup
if ni <= nw:
xi = [0, nw] # x interp
# compute_loss.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 / 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 = nn.functional.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 *= WORLD_SIZE # gradient averaged between devices in DDP mode
if opt.quad:
loss *= 4.
# Backward
scaler.scale(loss).backward()
# Optimize
if ni - last_opt_step >= accumulate:
scaler.step(optimizer) # optimizer.step
scaler.update()
optimizer.zero_grad()
if ema:
ema.update(model)
last_opt_step = ni
# Log
if RANK in [-1, 0]:
mloss = (mloss * i + loss_items) / (i + 1
) # update mean losses
mem = f'{torch.cuda.memory_reserved() / 1E9 if torch.cuda.is_available() else 0:.3g}G' # (GB)
pbar.set_description(('%10s' * 2 + '%10.4g' * 5) %
(f'{epoch}/{epochs - 1}', mem, *mloss,
targets.shape[0], imgs.shape[-1]))
callbacks.run('on_train_batch_end', ni, model, imgs, targets,
paths, plots, opt.sync_bn)
if callbacks.stop_training:
return
# end batch ------------------------------------------------------------------------------------------------
# Scheduler
lr = [x['lr'] for x in optimizer.param_groups] # for loggers
scheduler.step()
if RANK in [-1, 0]:
# mAP
callbacks.run('on_train_epoch_end', epoch=epoch)
ema.update_attr(model,
include=[
'yaml', 'nc', 'hyp', 'names', 'stride',
'class_weights'
])
final_epoch = (epoch + 1 == epochs) or stopper.possible_stop
if not noval or final_epoch: # Calculate mAP
results, maps, _ = val.run(data_dict,
batch_size=batch_size //
WORLD_SIZE * 2,
imgsz=imgsz,
model=ema.ema,
single_cls=single_cls,
dataloader=val_loader,
save_dir=save_dir,
plots=False,
callbacks=callbacks,
compute_loss=compute_loss)
# 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
log_vals = list(mloss) + list(results) + lr
callbacks.run('on_fit_epoch_end', log_vals, epoch, best_fitness,
fi)
# Save model
if (not nosave) or (final_epoch and not evolve): # if save
ckpt = {
'epoch': epoch,
'best_fitness': best_fitness,
'model': deepcopy(de_parallel(model)).half(),
'ema': deepcopy(ema.ema).half(),
'updates': ema.updates,
'optimizer': optimizer.state_dict(),
'wandb_id':
loggers.wandb.wandb_run.id if loggers.wandb else None,
'date': datetime.now().isoformat()
}
# Save last, best and delete
torch.save(ckpt, last)
if best_fitness == fi:
torch.save(ckpt, best)
if (epoch > 0) and (opt.save_period >
0) and (epoch % opt.save_period == 0):
torch.save(ckpt, w / f'epoch{epoch}.pt')
del ckpt
callbacks.run('on_model_save', last, epoch, final_epoch,
best_fitness, fi)
# Stop Single-GPU
if RANK == -1 and stopper(epoch=epoch, fitness=fi):
break
# Stop DDP TODO: known issues shttps://github.com/ultralytics/yolov5/pull/4576
# stop = stopper(epoch=epoch, fitness=fi)
# if RANK == 0:
# dist.broadcast_object_list([stop], 0) # broadcast 'stop' to all ranks
# Stop DPP
# with torch_distributed_zero_first(RANK):
# if stop:
# break # must break all DDP ranks
# end epoch ----------------------------------------------------------------------------------------------------
# end training -----------------------------------------------------------------------------------------------------
if RANK in [-1, 0]:
LOGGER.info(
f'\n{epoch - start_epoch + 1} epochs completed in {(time.time() - t0) / 3600:.3f} hours.'
)
for f in last, best:
if f.exists():
strip_optimizer(f) # strip optimizers
if f is best:
LOGGER.info(f'\nValidating {f}...')
results, _, _ = val.run(
data_dict,
batch_size=batch_size // WORLD_SIZE * 2,
imgsz=imgsz,
model=attempt_load(f, device).half(),
iou_thres=0.65 if is_coco else
0.60, # best pycocotools results at 0.65
single_cls=single_cls,
dataloader=val_loader,
save_dir=save_dir,
save_json=is_coco,
verbose=True,
plots=True,
callbacks=callbacks,
compute_loss=compute_loss) # val best model with plots
if is_coco:
callbacks.run('on_fit_epoch_end',
list(mloss) + list(results) + lr, epoch,
best_fitness, fi)
callbacks.run('on_train_end', last, best, plots, epoch, results)
LOGGER.info(f"Results saved to {colorstr('bold', save_dir)}")
torch.cuda.empty_cache()
return results
def __init__(self, weights='yolov5s.pt', device=None, dnn=True):
# Usage:
# PyTorch: weights = *.pt
# TorchScript: *.torchscript.pt
# CoreML: *.mlmodel
# TensorFlow: *_saved_model
# TensorFlow: *.pb
# TensorFlow Lite: *.tflite
# ONNX Runtime: *.onnx
# OpenCV DNN: *.onnx with dnn=True
# TensorRT: *.engine
super().__init__()
w = str(weights[0] if isinstance(weights, list) else weights)
suffix, suffixes = Path(w).suffix.lower(), [
'.pt', '.onnx', '.engine', '.tflite', '.pb', '', '.mlmodel'
]
check_suffix(w, suffixes) # check weights have acceptable suffix
pt, onnx, engine, tflite, pb, saved_model, coreml = (
suffix == x for x in suffixes) # backend booleans
jit = pt and 'torchscript' in w.lower()
stride, names = 64, [f'class{i}'
for i in range(1000)] # assign defaults
if jit: # TorchScript
LOGGER.info(f'Loading {w} for TorchScript inference...')
extra_files = {'config.txt': ''} # model metadata
model = torch.jit.load(w, _extra_files=extra_files)
if extra_files['config.txt']:
d = json.loads(extra_files['config.txt']) # extra_files dict
stride, names = int(d['stride']), d['names']
elif pt: # PyTorch
from models.experimental import attempt_load # scoped to avoid circular import
model = torch.jit.load(w) if 'torchscript' in w else attempt_load(
weights, map_location=device)
stride = int(model.stride.max()) # model stride
names = model.module.names if hasattr(
model, 'module') else model.names # get class names
elif coreml: # CoreML *.mlmodel
import coremltools as ct
model = ct.models.MLModel(w)
elif dnn: # ONNX OpenCV DNN
LOGGER.info(f'Loading {w} for ONNX OpenCV DNN inference...')
check_requirements(('opencv-python>=4.5.4', ))
net = cv2.dnn.readNetFromONNX(w)
elif onnx: # ONNX Runtime
LOGGER.info(f'Loading {w} for ONNX Runtime inference...')
check_requirements(
('onnx',
'onnxruntime-gpu' if torch.has_cuda else 'onnxruntime'))
import onnxruntime
session = onnxruntime.InferenceSession(w, None)
elif engine: # TensorRT
LOGGER.info(f'Loading {w} for TensorRT inference...')
import tensorrt as trt # https://developer.nvidia.com/nvidia-tensorrt-download
Binding = namedtuple('Binding',
('name', 'dtype', 'shape', 'data', 'ptr'))
logger = trt.Logger(trt.Logger.INFO)
with open(w, 'rb') as f, trt.Runtime(logger) as runtime:
model = runtime.deserialize_cuda_engine(f.read())
bindings = OrderedDict()
for index in range(model.num_bindings):
name = model.get_binding_name(index)
dtype = trt.nptype(model.get_binding_dtype(index))
shape = tuple(model.get_binding_shape(index))
data = torch.from_numpy(np.empty(
shape, dtype=np.dtype(dtype))).to(device)
bindings[name] = Binding(name, dtype, shape, data,
int(data.data_ptr()))
binding_addrs = OrderedDict(
(n, d.ptr) for n, d in bindings.items())
context = model.create_execution_context()
batch_size = bindings['images'].shape[0]
else: # TensorFlow model (TFLite, pb, saved_model)
if pb: # https://www.tensorflow.org/guide/migrate#a_graphpb_or_graphpbtxt
LOGGER.info(f'Loading {w} for TensorFlow *.pb inference...')
import tensorflow as tf
def wrap_frozen_graph(gd, inputs, outputs):
x = tf.compat.v1.wrap_function(
lambda: tf.compat.v1.import_graph_def(gd, name=""),
[]) # wrapped
return x.prune(
tf.nest.map_structure(x.graph.as_graph_element,
inputs),
tf.nest.map_structure(x.graph.as_graph_element,
outputs))
graph_def = tf.Graph().as_graph_def()
graph_def.ParseFromString(open(w, 'rb').read())
frozen_func = wrap_frozen_graph(gd=graph_def,
inputs="x:0",
outputs="Identity:0")
elif saved_model:
LOGGER.info(
f'Loading {w} for TensorFlow saved_model inference...')
import tensorflow as tf
model = tf.keras.models.load_model(w)
elif tflite: # https://www.tensorflow.org/lite/guide/python#install_tensorflow_lite_for_python
if 'edgetpu' in w.lower():
LOGGER.info(
f'Loading {w} for TensorFlow Lite Edge TPU inference...'
)
import tflite_runtime.interpreter as tfli
delegate = {
'Linux':
'libedgetpu.so.1', # install https://coral.ai/software/#edgetpu-runtime
'Darwin': 'libedgetpu.1.dylib',
'Windows': 'edgetpu.dll'
}[platform.system()]
interpreter = tfli.Interpreter(
model_path=w,
experimental_delegates=[tfli.load_delegate(delegate)])
else:
LOGGER.info(
f'Loading {w} for TensorFlow Lite inference...')
import tensorflow as tf
interpreter = tf.lite.Interpreter(
model_path=w) # load TFLite model
interpreter.allocate_tensors() # allocate
input_details = interpreter.get_input_details() # inputs
output_details = interpreter.get_output_details() # outputs
self.__dict__.update(locals()) # assign all variables to self
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
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
classes=None, # filter by class: --class 0, or --class 0 2 3
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
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 = next(model.parameters()).device # get model device
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
check_suffix(weights, '.pt')
model = attempt_load(weights, map_location=device) # load FP32 model
gs = max(int(model.stride.max()), 32) # grid size (max stride)
imgsz = check_img_size(imgsz, s=gs) # check image 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)
# Data
data = check_dataset(data) # check
# Half
half &= device.type != 'cpu' # half precision only supported on CUDA
model.half() if half else model.float()
# Configure
model.eval()
is_coco = isinstance(data.get('val'), str) and data['val'].endswith(
'coco/val2017.txt') # COCO dataset
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()
# Dataloader
if not training:
if device.type != 'cpu':
model(
torch.zeros(1, 3, imgsz, imgsz).to(device).type_as(
next(model.parameters()))) # run once
task = task if task in (
'train', 'val', 'test') else 'val' # path to train/val/test images
dataloader = create_dataloader(data[task],
imgsz,
batch_size,
gs,
single_cls,
pad=0.5,
rect=True,
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)
}
#
# names = {(k - 1): v for k, v in names.items() if v != "head"}
#
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 = [], [], [], []
for batch_i, (img, targets, paths,
shapes) in enumerate(tqdm(dataloader, desc=s)):
t1 = time_sync()
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
t2 = time_sync()
dt[0] += t2 - t1
# Run model
out, train_out = model(
img, augment=augment) # inference and training outputs
dt[1] += time_sync() - t2
# Compute loss
if compute_loss:
loss += compute_loss([x.float() for x in train_out],
targets)[1] # box, obj, cls
# Run NMS
targets[:, 2:] *= 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
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
# if out:
# for i, pred in enumerate(out):
# if len(pred) == 0:
# continue
# only_person_pred = pred[pred[:, 5] != 0]
# only_person_pred[:, 5] = 0
# out[i] = only_person_pred
#
# Statistics per image
for si, pred in enumerate(out):
labels = targets[targets[:, 0] == si, 1:]
nl = len(labels)
tcls = labels[:, 0].tolist() if nl else [] # target class
path, shape = Path(paths[si]), shapes[si][0]
seen += 1
if len(pred) == 0:
if nl:
stats.append((torch.zeros(0, niou, dtype=torch.bool),
torch.Tensor(), torch.Tensor(), tcls))
continue
# Predictions
if single_cls:
pred[:, 5] = 0
predn = pred.clone()
scale_coords(img[si].shape[1:], predn[:, :4], shape,
shapes[si][1]) # native-space pred
# Evaluate
if nl:
tbox = xywh2xyxy(labels[:, 1:5]) # target boxes
scale_coords(img[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)
else:
correct = torch.zeros(pred.shape[0], niou, dtype=torch.bool)
stats.append((correct.cpu(), pred[:, 4].cpu(), pred[:, 5].cpu(),
tcls)) # (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,
img[si])
# Plot images
if plots and batch_i < 3:
f = save_dir / f'val_batch{batch_i}_labels.jpg' # labels
Thread(target=plot_images,
args=(img, targets, paths, f, names),
daemon=True).start()
f = save_dir / f'val_batch{batch_i}_pred.jpg' # predictions
Thread(target=plot_images,
args=(img, output_to_target(out), paths, f, names),
daemon=True).start()
# 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)
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(np.int64),
minlength=nc) # number of targets per class
else:
nt = torch.zeros(1)
# Print results
pf = '%20s' + '%11i' * 2 + '%11.3g' * 4 # print format
print(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):
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 dt) # speeds per image
if not training:
shape = (batch_size, 3, imgsz, imgsz)
print(
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
print(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.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
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 ''
print(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 run(
weights=ROOT / 'yolov5s.pt', # model.pt path(s) 训练的权重
source=ROOT /
'data/images', # file/dir/URL/glob, 0 for webcam 测试数据，图片/视频路径，'0'摄像头，rtsp视频流
imgsz=640, # inference size (pixels) 网络输入图片大小
conf_thres=0.25, # confidence threshold 置信度阈值
iou_thres=0.45, # NMS IOU threshold nms的iou阈值
max_det=1000, # maximum detections per image 分类数
device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu 设备
view_img=True, # show results 是否展示预测之后的图片/视频
save_txt=False, # save results to *.txt 是否将预测的框坐标保持txt格式，默认false
# save_conf=False, # save confidences in --save-txt labels 置信度保存
save_crop=False, # save cropped prediction boxes
nosave=False, # do not save images/videos 不保存
classes=None, # filter by class: --class 0, or --class 0 2 3 设置只保留某一部分类别
agnostic_nms=False, # class-agnostic NMS 进行nms是否也去除不同类别之间的框
augment=False, # augmented inference 图像增强
visualize=False, # visualize features 可视化
# update=False, # update all models 若ture，则对所有模型进行strip_optimizer操作，去除pt文件中的优化器等信息，默认false
project=ROOT / 'runs/detect', # 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
):
source = str(source)
save_img = not nosave and not source.endswith(
'.txt') # save inference images
webcam = source.isnumeric() or source.endswith(
'.txt') or source.lower().startswith(
('rtsp://', 'rtmp://', 'http://', 'https://'))
# 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
# Initialize
set_logging()
device = select_device(device)
half &= device.type != 'cpu' # half precision only supported on CUDA
# Load model
w = weights[0] if isinstance(weights, list) else weights
classify, suffix, suffixes = False, Path(w).suffix.lower(), [
'.pt', '.onnx', '.tflite', '.pb', ''
]
check_suffix(w, suffixes) # check weights have acceptable suffix
pt, onnx, tflite, pb, saved_model = (suffix == x
for x in suffixes) # backend booleans
stride, names = 64, [f'class{i}' for i in range(1000)] # assign defaults
if pt:
model = attempt_load(
weights,
map_location=device) # load FP32 model 加载float32模型，确保图片分辨率能整除32
stride = int(model.stride.max()) # model stride
names = model.module.names if hasattr(
model, 'module') else model.names # get class names
#设置Float16
if half:
model.half() # to FP16
# 设置2次分类
if classify: # second-stage classifier
modelc = load_classifier(name='resnet50', n=2) # initialize
modelc.load_state_dict(
torch.load('resnet50.pt',
map_location=device)['model']).to(device).eval()
# elif onnx:
# check_requirements(('onnx', 'onnxruntime'))
# import onnxruntime
# session = onnxruntime.InferenceSession(w, None)
else: # TensorFlow models
check_requirements(('tensorflow>=2.4.1', ))
import tensorflow as tf
if pb: # https://www.tensorflow.org/guide/migrate#a_graphpb_or_graphpbtxt
def wrap_frozen_graph(gd, inputs, outputs):
x = tf.compat.v1.wrap_function(
lambda: tf.compat.v1.import_graph_def(gd, name=""),
[]) # wrapped import
return x.prune(
tf.nest.map_structure(x.graph.as_graph_element, inputs),
tf.nest.map_structure(x.graph.as_graph_element, outputs))
graph_def = tf.Graph().as_graph_def()
graph_def.ParseFromString(open(w, 'rb').read())
frozen_func = wrap_frozen_graph(gd=graph_def,
inputs="x:0",
outputs="Identity:0")
elif saved_model:
model = tf.keras.models.load_model(w)
elif tflite:
interpreter = tf.lite.Interpreter(
model_path=w) # load TFLite model
interpreter.allocate_tensors() # allocate
input_details = interpreter.get_input_details() # inputs
output_details = interpreter.get_output_details() # outputs
int8 = input_details[0][
'dtype'] == np.uint8 # is TFLite quantized uint8 model
imgsz = check_img_size(imgsz, s=stride) # check image size
# Dataloader
# 通过不同的输入源来设置不同的数据加载方式
# 摄像头
if webcam:
view_img = check_imshow()
cudnn.benchmark = True # set True to speed up constant image size inference
dataset = LoadStreams(source, img_size=imgsz, stride=stride, auto=pt)
bs = len(dataset) # batch_size
# 图片或视频
else:
dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt)
bs = 1 # batch_size
vid_path, vid_writer = [None] * bs, [None] * bs
# Run inference
if pt and device.type != 'cpu':
# 进行一次前向推理，测试程序是否正常
model(
torch.zeros(1, 3, *imgsz).to(device).type_as(
next(model.parameters()))) # run once
dt, seen = [0.0, 0.0, 0.0], 0
'''
path 图片/视频路径
img 进行resize+pad之后的图片，如(3,640,512) 格式(c,h,w)
img0s 原size图片，如(1080,810,3)
cap 当读取图片时为None,读取视频时为视频源
'''
for path, img, im0s, vid_cap in dataset:
t1 = time_sync()
if onnx:
img = img.astype('float32')
else:
img = torch.from_numpy(img).to(device)
# 图片也设置为Float16或者32
img = img.half() if half else img.float() # uint8 to fp16/32
img = img / 255.0 # 0 - 255 to 0.0 - 1.0
# 没有batch_size时，在最前面添加一个轴
if len(img.shape) == 3:
img = img[None] # expand for batch dim
t2 = time_sync()
dt[0] += t2 - t1
# Inference
if pt:
visualize = increment_path(save_dir / Path(path).stem,
mkdir=True) if visualize else False
'''
前向传播，返回pred的shape是(1,num_boxes,5+num_class)
h,w为传入网络图片的高和宽，注意dataset在检测时使用了矩形推理，所以h不一定等于w
num_boxes = (h/32*w/32+h/16*w/16+h/8*w/8)*3
例如：图片大小720，1280 -> 15120个boxes = (20*12 + 40*24 + 80*48 = 5040)*3
pred[...,0:4]为预测框坐标；预测框坐标为xywh
pred[...,4]为objectness置信度
pred[...,5:-1]为分类结果
'''
pred = model(img, augment=augment, visualize=visualize)[0]
# elif onnx:
# pred = torch.tensor(session.run([session.get_outputs()[0].name], {session.get_inputs()[0].name: img}))
else: # tensorflow model (tflite, pb, saved_model)
imn = img.permute(0, 2, 3, 1).cpu().numpy() # image in numpy
if pb:
pred = frozen_func(x=tf.constant(imn)).numpy()
elif saved_model:
pred = model(imn, training=False).numpy()
elif tflite:
if int8:
scale, zero_point = input_details[0]['quantization']
imn = (imn / scale + zero_point).astype(
np.uint8) # de-scale
interpreter.set_tensor(input_details[0]['index'], imn)
interpreter.invoke()
pred = interpreter.get_tensor(output_details[0]['index'])
if int8:
scale, zero_point = output_details[0]['quantization']
pred = (pred.astype(np.float32) -
zero_point) * scale # re-scale
pred[..., 0] *= imgsz[1] # x
pred[..., 1] *= imgsz[0] # y
pred[..., 2] *= imgsz[1] # w
pred[..., 3] *= imgsz[0] # h
pred = torch.tensor(pred)
t3 = time_sync()
dt[1] += t3 - t2
# NMS
'''
pred:前向传播的输出
conf_thres:置信度阈值
iou_thres:iou阈值
classes：是否只保留特定的类别
agnostic_nmsL进行nms是否也去除不同类别之间的框
经过nms后预测框格式，xywh->xyxy(左上角右上角)
pred是一个列表list[torch.tensor],长度为nms后目标框个数
每一个torch.tensor的shape为(num_boxes,6),内容为box(4个值)+cunf+cls
'''
pred = non_max_suppression(pred,
conf_thres,
iou_thres,
classes,
agnostic_nms,
max_det=max_det)
dt[2] += time_sync() - t3
# Second-stage classifier (optional)
# 添加二级分类，默认false
if classify:
pred = apply_classifier(pred, modelc, img, im0s)
# Process predictions
# 对每一张图片处理
for i, det in enumerate(pred): # per image
seen += 1
# 如果输入源是webcam，则batch_size不为1，取出dataset中的一张图片
if webcam: # batch_size >= 1
p, s, im0, frame = path[i], f'{i}: ', im0s[i].copy(
), dataset.count
else:
p, s, im0, frame = path, '', im0s.copy(), getattr(
dataset, 'frame', 0)
p = Path(p) # to Path
# 设置保存图片或视频的路径
# p是原图片路径
save_path = str(save_dir / p.name) # img.jpg
#设置保存框坐标txt文件的路径
txt_path = str(save_dir / 'labels' / p.stem) + (
'' if dataset.mode == 'image' else f'_{frame}') # img.txt
# 设置打印信息（图片宽高），s如'640*512'
s += '%gx%g ' % img.shape[2:] # print string
gn = torch.tensor(im0.shape)[[1, 0, 1,
0]] # normalization gain whwh
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
# 调整预测框坐标，基于resize+pad的图片坐标->基于原size图片坐标
# 此时坐标格式为xyxy
det[:, :4] = scale_coords(img.shape[2:], det[:, :4],
im0.shape).round()
# Print results
# 打印检测到的类别数量
for c in det[:, -1].unique():
n = (det[:, -1] == c).sum() # detections per class
s += f"{n} {names[int(c)]}{'s' * (n > 1)}, " # add to string
# Write results
# 保存预测结果
for *xyxy, conf, cls in reversed(det):
# if save_txt: # Write to file
# # 将xyxy格式转为xywh格式，并除上我w，h作归一化，转化为列表再保存
# 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(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))
if save_crop:
save_one_box(xyxy,
imc,
file=save_dir / 'crops' / names[c] /
f'{p.stem}.jpg',
BGR=True)
# Print time (inference-only)
# print(f'{pred[0][0][0].tolist()} {pred[0][0][1].tolist()} {s}Done. ({t3 - t2:.3f}s)')
# Stream results
im0 = annotator.result()
# xxx = (pred[0][0][0].tolist()+pred[0][0][2].tolist())/2
# yyy = (pred[0][0][1].tolist()+pred[0][0][3].tolist())/2
if view_img:
# + / 2 +
cv2.imshow(str(p), im0)
cv2.moveWindow(str(p), 0, 0)
# pyautogui.moveTo(xxx, yyy)
cv2.waitKey(1000) # 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 += '.mp4'
# vid_writer[i] = cv2.VideoWriter(save_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (w, h))
# vid_writer[i].write(im0)
# Print results
t = tuple(x / seen * 1E3 for x in dt) # speeds per image
print(
f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {(1, 3, *imgsz)}'
% t)
def __init__(self, weights='yolov3.pt', device=None, dnn=True):
# Usage:
# PyTorch: weights = *.pt
# TorchScript: *.torchscript.pt
# CoreML: *.mlmodel
# TensorFlow: *_saved_model
# TensorFlow: *.pb
# TensorFlow Lite: *.tflite
# ONNX Runtime: *.onnx
# OpenCV DNN: *.onnx with dnn=True
super().__init__()
w = str(weights[0] if isinstance(weights, list) else weights)
suffix, suffixes = Path(w).suffix.lower(), ['.pt', '.onnx', '.tflite', '.pb', '', '.mlmodel']
check_suffix(w, suffixes) # check weights have acceptable suffix
pt, onnx, tflite, pb, saved_model, coreml = (suffix == x for x in suffixes) # backend booleans
jit = pt and 'torchscript' in w.lower()
stride, names = 64, [f'class{i}' for i in range(1000)] # assign defaults
if jit: # TorchScript
LOGGER.info(f'Loading {w} for TorchScript inference...')
extra_files = {'config.txt': ''} # model metadata
model = torch.jit.load(w, _extra_files=extra_files)
if extra_files['config.txt']:
d = json.loads(extra_files['config.txt']) # extra_files dict
stride, names = int(d['stride']), d['names']
elif pt: # PyTorch
from models.experimental import attempt_load # scoped to avoid circular import
model = torch.jit.load(w) if 'torchscript' in w else attempt_load(weights, map_location=device)
stride = int(model.stride.max()) # model stride
names = model.module.names if hasattr(model, 'module') else model.names # get class names
elif coreml: # CoreML *.mlmodel
import coremltools as ct
model = ct.models.MLModel(w)
elif dnn: # ONNX OpenCV DNN
LOGGER.info(f'Loading {w} for ONNX OpenCV DNN inference...')
check_requirements(('opencv-python>=4.5.4',))
net = cv2.dnn.readNetFromONNX(w)
elif onnx: # ONNX Runtime
LOGGER.info(f'Loading {w} for ONNX Runtime inference...')
check_requirements(('onnx', 'onnxruntime-gpu' if torch.has_cuda else 'onnxruntime'))
import onnxruntime
session = onnxruntime.InferenceSession(w, None)
else: # TensorFlow model (TFLite, pb, saved_model)
import tensorflow as tf
if pb: # https://www.tensorflow.org/guide/migrate#a_graphpb_or_graphpbtxt
def wrap_frozen_graph(gd, inputs, outputs):
x = tf.compat.v1.wrap_function(lambda: tf.compat.v1.import_graph_def(gd, name=""), []) # wrapped
return x.prune(tf.nest.map_structure(x.graph.as_graph_element, inputs),
tf.nest.map_structure(x.graph.as_graph_element, outputs))
LOGGER.info(f'Loading {w} for TensorFlow *.pb inference...')
graph_def = tf.Graph().as_graph_def()
graph_def.ParseFromString(open(w, 'rb').read())
frozen_func = wrap_frozen_graph(gd=graph_def, inputs="x:0", outputs="Identity:0")
elif saved_model:
LOGGER.info(f'Loading {w} for TensorFlow saved_model inference...')
model = tf.keras.models.load_model(w)
elif tflite: # https://www.tensorflow.org/lite/guide/python#install_tensorflow_lite_for_python
if 'edgetpu' in w.lower():
LOGGER.info(f'Loading {w} for TensorFlow Edge TPU inference...')
import tflite_runtime.interpreter as tfli
delegate = {'Linux': 'libedgetpu.so.1', # install https://coral.ai/software/#edgetpu-runtime
'Darwin': 'libedgetpu.1.dylib',
'Windows': 'edgetpu.dll'}[platform.system()]
interpreter = tfli.Interpreter(model_path=w, experimental_delegates=[tfli.load_delegate(delegate)])
else:
LOGGER.info(f'Loading {w} for TensorFlow Lite inference...')
interpreter = tf.lite.Interpreter(model_path=w) # load TFLite model
interpreter.allocate_tensors() # allocate
input_details = interpreter.get_input_details() # inputs
output_details = interpreter.get_output_details() # outputs
self.__dict__.update(locals()) # assign all variables to self
def run(
weights=ROOT / 'yolov5s.pt', # model.pt path(s) 训练的权重
imgsz=[640, 640], # inference size (pixels) 网络输入图片大小
conf_thres=0.25, # confidence threshold 置信度阈值
iou_thres=0.45, # NMS IOU threshold nms的iou阈值
max_det=1000, # maximum detections per image 分类数
device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu 设备
view_img=True, # show results 是否展示预测之后的图片/视频
classes=None, # filter by class: --class 0, or --class 0 2 3 设置只保留某一部分类别
agnostic_nms=False, # class-agnostic NMS 进行nms是否也去除不同类别之间的框
augment=False, # augmented inference 图像增强
visualize=False, # visualize features 可视化
line_thickness=3, # bounding box thickness (pixels)
hide_labels=False, # hide labels
hide_conf=False, # hide confidences
half=False, # use FP16 half-precision inference
):
# Initialize
set_logging()
device = select_device(device)
half &= device.type != 'cpu' # half precision only supported on CUDA
# Load model
w = weights[0] if isinstance(weights, list) else weights
classify, suffix, suffixes = False, Path(w).suffix.lower(), [
'.pt', '.onnx', '.tflite', '.pb', ''
]
check_suffix(w, suffixes) # check weights have acceptable suffix
pt, onnx, tflite, pb, saved_model = (suffix == x
for x in suffixes) # backend booleans
stride, names = 64, [f'class{i}' for i in range(1000)] # assign defaults
if pt:
model = attempt_load(
weights,
map_location=device) # load FP32 model 加载float32模型，确保图片分辨率能整除32
stride = int(model.stride.max()) # model stride
names = model.module.names if hasattr(
model, 'module') else model.names # get class names
#设置Float16
if half:
model.half() # to FP16
# 设置2次分类
if classify: # second-stage classifier
modelc = load_classifier(name='resnet50', n=2) # initialize
modelc.load_state_dict(
torch.load('resnet50.pt',
map_location=device)['model']).to(device).eval()
else: # TensorFlow models
check_requirements(('tensorflow>=2.4.1', ))
import tensorflow as tf
if pb: # https://www.tensorflow.org/guide/migrate#a_graphpb_or_graphpbtxt
def wrap_frozen_graph(gd, inputs, outputs):
x = tf.compat.v1.wrap_function(
lambda: tf.compat.v1.import_graph_def(gd, name=""),
[]) # wrapped import
return x.prune(
tf.nest.map_structure(x.graph.as_graph_element, inputs),
tf.nest.map_structure(x.graph.as_graph_element, outputs))
graph_def = tf.Graph().as_graph_def()
graph_def.ParseFromString(open(w, 'rb').read())
frozen_func = wrap_frozen_graph(gd=graph_def,
inputs="x:0",
outputs="Identity:0")
elif saved_model:
model = tf.keras.models.load_model(w)
elif tflite:
interpreter = tf.lite.Interpreter(
model_path=w) # load TFLite model
interpreter.allocate_tensors() # allocate
input_details = interpreter.get_input_details() # inputs
output_details = interpreter.get_output_details() # outputs
int8 = input_details[0][
'dtype'] == np.uint8 # is TFLite quantized uint8 model
imgsz = check_img_size(imgsz, s=stride) # check image size
# Dataloader
# 图片或视频
tmp = False
tmp2 = False
mon = {'top': 0, 'left': 0, 'width': 960, 'height': 960}
while True:
im = np.array(mss().grab(mon))
screen = cv2.cvtColor(im, cv2.COLOR_BGRA2BGR)
dataset = LoadImages(screen, img_size=imgsz, stride=stride, auto=pt)
dt, seen = [0.0, 0.0, 0.0], 0
'''
path 图片/视频路径
img 进行resize+pad之后的图片，如(3,640,512) 格式(c,h,w)
img0s 原size图片，如(1080,810,3)
cap 当读取图片时为None,读取视频时为视频源
'''
for img, im0s, vid_cap in dataset:
t1 = time_sync()
if onnx:
img = img.astype('float32')
else:
img = torch.from_numpy(img).to(device)
# print(img)
# 图片也设置为Float16或者32
img = img.half() if half else img.float() # uint8 to fp16/32
img = img / 255.0 # 0 - 255 to 0.0 - 1.0
# 没有batch_size时，在最前面添加一个轴
if len(img.shape) == 3:
img = img[None] # expand for batch dim
t2 = time_sync()
dt[0] += t2 - t1
# Inference
if pt:
'''
前向传播，返回pred的shape是(1,num_boxes,5+num_class)
h,w为传入网络图片的高和宽，注意dataset在检测时使用了矩形推理，所以h不一定等于w
num_boxes = (h/32*w/32+h/16*w/16+h/8*w/8)*3
例如：图片大小720，1280 -> 15120个boxes = (20*12 + 40*24 + 80*48 = 5040)*3
pred[...,0:4]为预测框坐标；预测框坐标为xywh
pred[...,4]为objectness置信度
pred[...,5:-1]为分类结果
'''
pred = model(img, augment=augment, visualize=visualize)[0]
else: # tensorflow model (tflite, pb, saved_model)
imn = img.permute(0, 2, 3, 1).cpu().numpy() # image in numpy
if pb:
pred = frozen_func(x=tf.constant(imn)).numpy()
elif saved_model:
pred = model(imn, training=False).numpy()
elif tflite:
if int8:
scale, zero_point = input_details[0]['quantization']
imn = (imn / scale + zero_point).astype(
np.uint8) # de-scale
interpreter.set_tensor(input_details[0]['index'], imn)
interpreter.invoke()
pred = interpreter.get_tensor(output_details[0]['index'])
if int8:
scale, zero_point = output_details[0]['quantization']
pred = (pred.astype(np.float32) -
zero_point) * scale # re-scale
pred[..., 0] *= imgsz[1] # x
pred[..., 1] *= imgsz[0] # y
pred[..., 2] *= imgsz[1] # w
pred[..., 3] *= imgsz[0] # h
pred = torch.tensor(pred)
t3 = time_sync()
dt[1] += t3 - t2
# NMS
'''
pred:前向传播的输出
conf_thres:置信度阈值
iou_thres:iou阈值
classes：是否只保留特定的类别
agnostic_nmsL进行nms是否也去除不同类别之间的框
经过nms后预测框格式，xywh->xyxy(左上角右上角)
pred是一个列表list[torch.tensor],长度为nms后目标框个数
每一个torch.tensor的shape为(num_boxes,6),内容为box(4个值)+cunf+cls
'''
pred = non_max_suppression(pred,
conf_thres,
iou_thres,
classes,
agnostic_nms,
max_det=max_det)
dt[2] += time_sync() - t3
# Second-stage classifier (optional)
# 添加二级分类，默认false
# if classify:
# pred = apply_classifier(pred, modelc, img, im0s)
# Process predictions
# 对每一张图片处理
for i, det in enumerate(pred): # per image
seen += 1
s, im0 = '', im0s.copy()
# 设置打印信息（图片宽高），s如'640*512'
s += '%gx%g ' % img.shape[2:] # print string
annotator = Annotator(im0,
line_width=line_thickness,
example=str(names))
if len(det):
# Rescale boxes from img_size to im0 size
# 调整预测框坐标，基于resize+pad的图片坐标->基于原size图片坐标
# 此时坐标格式为xyxy
det[:, :4] = scale_coords(img.shape[2:], det[:, :4],
im0.shape).round()
# Print results
# 打印检测到的类别数量
for c in det[:, -1].unique():
n = (det[:, -1] == c).sum() # detections per class
s += f"{n} {names[int(c)]}{'s' * (n > 1)}, " # add to string
# Write results
# 保存预测结果
for *xyxy, conf, cls in reversed(det):
if 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))
# Stream results
im0 = annotator.result()
cv2.imshow('a crop of the screen', im0)
cv2.moveWindow('a crop of the screen', 960, 0)
if cv2.waitKey(1) & 0xff == ord('q'):
tmp = True
break
if tmp:
tmp2 = True
break
if tmp2:
break