def postprocess(outputs, nested_test_image, args):
image_shape = nested_test_image.tensors.shape[2:]
postprocessor = PostProcess()
results = postprocessor(outputs, dg.to_variable([image_shape]))
image = plot_results(nested_test_image.tensors[0], results[0])
output_path = args.output_dir + '/' + args.demo_image.split('/')[-1]
cv2.imwrite(output_path, image)
def invoice(pretrained=False, num_classes=91, return_postprocessor=False):
"""
DETR R50 with 6 encoder and 6 decoder layers.
"""
model = _make_detr("resnet50", dilation=False, num_classes=num_classes)
if pretrained:
checkpoint = torch.hub.load_state_dict_from_url(
url="invoice_large_transfer/checkpoint.pth",
map_location="cpu",
check_hash=True)
model.load_state_dict(checkpoint["model"])
if return_postprocessor:
return model, PostProcess()
return model
def detr_resnet101(pretrained=False, num_classes=91, return_postprocessor=False):
"""
DETR-DC5 R101 with 6 encoder and 6 decoder layers.
Achieves 43.5/63.8 AP/AP50 on COCO val5k.
"""
model = _make_detr("resnet101", dilation=False, num_classes=num_classes)
if pretrained:
checkpoint = torch.hub.load_state_dict_from_url(
url="https://dl.fbaipublicfiles.com/detr/detr-r101-2c7b67e5.pth", map_location="cpu", check_hash=True
)
model.load_state_dict(checkpoint["model"])
if return_postprocessor:
return model, PostProcess()
return model
def detr_resnet101(pretrained=False,
num_classes=91,
return_postprocessor=False):
"""
DETR-DC5 R101 with 6 encoder and 6 decoder layers.
Achieves 43.5/63.8 AP/AP50 on COCO val5k.
"""
model = _make_detr('resnet101', dilation=False, num_classes=num_classes)
if pretrained:
checkpoint = paddle.load(
'https://dl.fbaipublicfiles.com/detr/detr-r101-2c7b67e5.pdiparams')
model.load_state_dict(checkpoint['model'])
if return_postprocessor:
return model, PostProcess()
return model
def detr_resnet50(pretrained=False,
num_classes=91,
return_postprocessor=False):
"""
DETR R50 with 6 encoder and 6 decoder layers.
Achieves 42/62.4 AP/AP50 on COCO val5k.
"""
model = _make_detr('resnet50', dilation=False, num_classes=num_classes)
if pretrained:
checkpoint = paddle.load(
'https://dl.fbaipublicfiles.com/detr/detr-r50-e632da11.pdiparams')
model.load_state_dict(checkpoint['model'])
if return_postprocessor:
return model, PostProcess()
return model
def detr_resnet101_dc5(pretrained=False, num_classes=91, return_postprocessor=False):
"""
DETR-DC5 R101 with 6 encoder and 6 decoder layers.
The last block of ResNet-101 has dilation to increase
output resolution.
Achieves 44.9/64.7 AP/AP50 on COCO val5k.
"""
model = _make_detr("resnet101", dilation=True, num_classes=num_classes)
if pretrained:
checkpoint = torch.hub.load_state_dict_from_url(
url="https://dl.fbaipublicfiles.com/detr/detr-r101-dc5-a2e86def.pth", map_location="cpu", check_hash=True
)
model.load_state_dict(checkpoint["model"])
if return_postprocessor:
return model, PostProcess()
return model
def detr_resnet50(pretrained=False,
num_classes=91,
return_postprocessor=False):
"""
DETR R50 with 6 encoder and 6 decoder layers.
Achieves 42/62.4 AP/AP50 on COCO val5k.
"""
model = _make_detr("resnet50", dilation=False, num_classes=num_classes)
if pretrained:
checkpoint = torch.hub.load_state_dict_from_url(
url="https://dl.fbaipublicfiles.com/detr/detr-r50-e632da11.pth",
map_location="cpu",
check_hash=False)
model.load_state_dict(checkpoint["model"])
if return_postprocessor:
return model, PostProcess()
return model
def detr_resnet101_dc5(pretrained=False,
num_classes=91,
return_postprocessor=False):
"""
DETR-DC5 R101 with 6 encoder and 6 decoder layers.
The last block of ResNet-101 has dilation to increase
output resolution.
Achieves 44.9/64.7 AP/AP50 on COCO val5k.
"""
model = _make_detr('resnet101', dilation=True, num_classes=num_classes)
if pretrained:
checkpoint = paddle.load(
'https://dl.fbaipublicfiles.com/detr/detr-r101-dc5-a2e86def.pdiparams'
)
model.load_state_dict(checkpoint['model'])
if return_postprocessor:
return model, PostProcess()
return model
pre_norm = False
num_classes = 3
num_queries = 100
aux_loss = True
backbone = build_backbone(lr_backbone, masks, backbone, dilation, hidden_dim, position_embedding)
transformer = build_transformer(hidden_dim, dropout, nheads, dim_feedforward, enc_layers, dec_layers, pre_norm)
model = DETR(
backbone,
transformer,
num_classes=num_classes,
num_queries=num_queries,
aux_loss=aux_loss,
)
transform = make_coco_transforms()
postprocessors = PostProcess()
checkpoint = torch.load('/home/palm/PycharmProjects/detr/snapshots/1/checkpoint00295.pth')
model.load_state_dict(checkpoint['model'])
train_ints, valid_ints, labels, max_box_per_image = create_csv_training_instances(
'/home/palm/PycharmProjects/algea/dataset/train_annotations',
'/home/palm/PycharmProjects/algea/dataset/test_annotations',
'/home/palm/PycharmProjects/algea/dataset/classes',
)
# os.listdir()
all_detections = []
all_annotations = []
model.cuda()
for instance in valid_ints:
t = time.time()
aux_loss=args.aux_loss,
)
matcher = HungarianMatcher(cost_class=args.set_cost_class, cost_bbox=args.set_cost_bbox, cost_giou=args.set_cost_giou)
weight_dict = {'loss_ce': 1, 'loss_bbox': args.bbox_loss_coef}
weight_dict['loss_giou'] = args.giou_loss_coef
if args.aux_loss:
aux_weight_dict = {}
for i in range(args.dec_layers - 1):
aux_weight_dict.update({k + f'_{i}': v for k, v in weight_dict.items()})
weight_dict.update(aux_weight_dict)
losses = ['labels', 'boxes', 'cardinality']
criterion = SetCriterion(args.num_classes, matcher=matcher, weight_dict=weight_dict,
eos_coef=args.eos_coef, losses=losses)
postprocessors = {'bbox': PostProcess()}
criterion.to(device)
model.to(device)
# %% set distributed model
model_without_ddp = model
if args.distributed:
model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[args.gpu])
model_without_ddp = model.module
n_parameters = sum(p.numel() for p in model.parameters() if p.requires_grad)
print('number of params:', n_parameters)
# %% set optimizer
param_dicts = [
{"params": [p for n, p in model_without_ddp.named_parameters() if "backbone" not in n and p.requires_grad]},
def build_model(args):
num_classes = 20 if args.dataset_file != 'coco' else 91
if args.dataset_file == "coco_panoptic":
num_classes = 250
if args.dataset_file == "ImageNet":
num_classes = 2 # feel free to this num_classes, positive integer larger than 1 is OK.
device = torch.device(args.device)
backbone = build_backbone(args)
transformer = build_transformer(args)
if args.dataset_file == "ImageNet":
model = UPDETR(backbone,
transformer,
num_classes=num_classes,
num_queries=args.num_queries,
aux_loss=args.aux_loss,
num_patches=args.num_patches,
feature_recon=args.feature_recon,
query_shuffle=args.query_shuffle)
else:
model = DETR(
backbone,
transformer,
num_classes=num_classes,
num_queries=args.num_queries,
aux_loss=args.aux_loss,
)
if args.masks:
model = DETRsegm(model,
freeze_detr=(args.frozen_weights is not None))
matcher = build_matcher(args)
weight_dict = {
'loss_ce': 1,
'loss_bbox': args.bbox_loss_coef,
'loss_giou': args.giou_loss_coef
}
if args.dataset_file == 'ImageNet' and args.feature_recon:
weight_dict['loss_feature'] = 1
if args.masks:
weight_dict["loss_mask"] = args.mask_loss_coef
weight_dict["loss_dice"] = args.dice_loss_coef
# TODO this is a hack
if args.aux_loss:
aux_weight_dict = {}
for i in range(args.dec_layers - 1):
aux_weight_dict.update(
{k + f'_{i}': v
for k, v in weight_dict.items()})
weight_dict.update(aux_weight_dict)
losses = ['labels', 'boxes', 'cardinality']
if args.dataset_file == 'ImageNet' and args.feature_recon:
losses += ['feature']
if args.masks:
losses += ["masks"]
criterion = SetCriterion(num_classes,
matcher=matcher,
weight_dict=weight_dict,
eos_coef=args.eos_coef,
losses=losses)
criterion.to(device)
postprocessors = {'bbox': PostProcess()}
if args.masks:
postprocessors['segm'] = PostProcessSegm()
if args.dataset_file == "coco_panoptic":
is_thing_map = {i: i <= 90 for i in range(201)}
postprocessors["panoptic"] = PostProcessPanoptic(is_thing_map,
threshold=0.85)
return model, criterion, postprocessors