Python maskrcnn_benchmark.modeling.detector.build_detection_model() Examples

The following are 30 code examples of maskrcnn_benchmark.modeling.detector.build_detection_model(). You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may also want to check out all available functions/classes of the module maskrcnn_benchmark.modeling.detector , or try the search function .
Example #1
Source File: inference_engine.py    From R2CNN.pytorch with MIT License 6 votes vote down vote up 
def __init__(
        self,
        cfg,
        weights,
        confidence_threshold=0.5,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(weights)

        self.transforms = self.build_transform()

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
Example #2
Source File: predictor.py    From sampling-free with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.color_list = colormap()
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = torch.tensor(confidence_threshold)
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #3
Source File: predictor.py    From RRPN_pytorch with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #4
Source File: predictor.py    From RRPN_pytorch with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #5
Source File: predictor.py    From training with Apache License 2.0 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #6
Source File: predictor.py    From RRPN_pytorch with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #7
Source File: predictor.py    From retinamask with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        checkpointer = DetectronCheckpointer(cfg, self.model)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #8
Source File: predictor.py    From FreeAnchor with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        checkpointer = DetectronCheckpointer(cfg, self.model)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.checkpointer = checkpointer
        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim

        self.CATEGORIES = COCO_CATEGORIES if cfg.DATASETS.TEST[0][:4] == 'coco' else VOC_CATEGORIES
Example #9
Source File: test_detectors.py    From maskrcnn-benchmark with MIT License 5 votes vote down vote up 
def create_model(cfg, device):
    cfg = copy.deepcopy(cfg)
    cfg.freeze()
    model = build_detection_model(cfg)
    model = model.to(device)
    return model
Example #10
Source File: predictor.py    From maskrcnn-benchmark with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
        weight_loading = None
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)
        
        if weight_loading:
            print('Loading weight from {}.'.format(weight_loading))
            _ = checkpointer._load_model(torch.load(weight_loading))
        
        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #11
Source File: predictor.py    From DF-Traffic-Sign-Identification with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #12
Source File: predictor.py    From HRNet-MaskRCNN-Benchmark with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #13
Source File: test_detectors.py    From sampling-free with MIT License 5 votes vote down vote up 
def create_model(cfg, device):
    cfg = copy.deepcopy(cfg)
    cfg.freeze()
    model = build_detection_model(cfg)
    model = model.to(device)
    return model
Example #14
Source File: predictor.py    From argoverse_baselinetracker with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #15
Source File: test_detectors.py    From DF-Traffic-Sign-Identification with MIT License 5 votes vote down vote up 
def create_model(cfg, device):
    cfg = copy.deepcopy(cfg)
    cfg.freeze()
    model = build_detection_model(cfg)
    model = model.to(device)
    return model
Example #16
Source File: predictor.py    From remote_sensing_object_detection_2019 with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #17
Source File: predictor.py    From remote_sensing_object_detection_2019 with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #18
Source File: predictor.py    From remote_sensing_object_detection_2019 with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #19
Source File: test_detectors.py    From DetNAS with MIT License 5 votes vote down vote up 
def create_model(cfg, device):
    cfg = copy.deepcopy(cfg)
    cfg.freeze()
    model = build_detection_model(cfg)
    model = model.to(device)
    return model
Example #20
Source File: predictor.py    From DetNAS with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
        weight_loading = None
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)
        
        if weight_loading:
            print('Loading weight from {}.'.format(weight_loading))
            _ = checkpointer._load_model(torch.load(weight_loading))
        
        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #21
Source File: predictor.py    From R2CNN.pytorch with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #22
Source File: predictor.py    From Res2Net-maskrcnn with MIT License 5 votes vote down vote up 
def __init__(
        self,
        cfg,
        confidence_threshold=0.7,
        show_mask_heatmaps=False,
        masks_per_dim=2,
        min_image_size=224,
    ):
        self.cfg = cfg.clone()
        self.model = build_detection_model(cfg)
        self.model.eval()
        self.device = torch.device(cfg.MODEL.DEVICE)
        self.model.to(self.device)
        self.min_image_size = min_image_size

        save_dir = cfg.OUTPUT_DIR
        checkpointer = DetectronCheckpointer(cfg, self.model, save_dir=save_dir)
        _ = checkpointer.load(cfg.MODEL.WEIGHT)

        self.transforms = self.build_transform()

        mask_threshold = -1 if show_mask_heatmaps else 0.5
        self.masker = Masker(threshold=mask_threshold, padding=1)

        # used to make colors for each class
        self.palette = torch.tensor([2 ** 25 - 1, 2 ** 15 - 1, 2 ** 21 - 1])

        self.cpu_device = torch.device("cpu")
        self.confidence_threshold = confidence_threshold
        self.show_mask_heatmaps = show_mask_heatmaps
        self.masks_per_dim = masks_per_dim
Example #23
Source File: train_net.py    From TinyBenchmark with MIT License 4 votes vote down vote up 
def train(cfg, local_rank, distributed):
    model = build_detection_model(cfg)
    device = torch.device(cfg.MODEL.DEVICE)
    model.to(device)

    optimizer = make_optimizer(cfg, model)
    scheduler = make_lr_scheduler(cfg, optimizer)

    if distributed:
        model = torch.nn.parallel.DistributedDataParallel(
            model, device_ids=[local_rank], output_device=local_rank,
            # this should be removed if we update BatchNorm stats
            broadcast_buffers=False,
        )

    arguments = {}
    arguments["iteration"] = 0

    output_dir = cfg.OUTPUT_DIR

    save_to_disk = get_rank() == 0
    checkpointer = DetectronCheckpointer(
        cfg, model, optimizer, scheduler, output_dir, save_to_disk
    )
    extra_checkpoint_data = checkpointer.load(cfg.MODEL.WEIGHT)
    arguments.update(extra_checkpoint_data)

    data_loader = make_data_loader(
        cfg,
        is_train=True,
        is_distributed=distributed,
        start_iter=arguments["iteration"],
    )

    checkpoint_period = cfg.SOLVER.CHECKPOINT_PERIOD

    do_train(
        model,
        data_loader,
        optimizer,
        scheduler,
        checkpointer,
        device,
        checkpoint_period,
        arguments,
    )

    return model
Example #24
Source File: train_net.py    From RRPN_pytorch with MIT License 4 votes vote down vote up 
def train(cfg, local_rank, distributed):
    model = build_detection_model(cfg)
    device = torch.device(cfg.MODEL.DEVICE)
    model.to(device)

    optimizer = make_optimizer(cfg, model)
    scheduler = make_lr_scheduler(cfg, optimizer)

    if distributed:
        model = torch.nn.parallel.DistributedDataParallel(
            model, device_ids=[local_rank], output_device=local_rank,
            # this should be removed if we update BatchNorm stats
            broadcast_buffers=False,
        )

    arguments = {}
    arguments["iteration"] = 0

    output_dir = cfg.OUTPUT_DIR

    save_to_disk = get_rank() == 0
    checkpointer = DetectronCheckpointer(
        cfg, model, optimizer, scheduler, output_dir, save_to_disk
    )



    extra_checkpoint_data = checkpointer.load(cfg.MODEL.WEIGHT)
    arguments.update(extra_checkpoint_data)

    data_loader = make_data_loader(
        cfg,
        is_train=True,
        is_distributed=distributed,
        start_iter=arguments["iteration"],
    )

    checkpoint_period = cfg.SOLVER.CHECKPOINT_PERIOD

    do_train(
        model,
        data_loader,
        optimizer,
        scheduler,
        checkpointer,
        device,
        checkpoint_period,
        arguments,
    )

    return model
Example #25
Source File: train_net.py    From DF-Traffic-Sign-Identification with MIT License 4 votes vote down vote up 
def train(cfg, local_rank, distributed):
    model = build_detection_model(cfg)
    device = torch.device(cfg.MODEL.DEVICE)
    model.to(device)

    optimizer = make_optimizer(cfg, model)
    scheduler = make_lr_scheduler(cfg, optimizer)

    # Initialize mixed-precision training
    use_mixed_precision = cfg.DTYPE == "float16"
    amp_opt_level = 'O1' if use_mixed_precision else 'O0'
    model, optimizer = amp.initialize(model, optimizer, opt_level=amp_opt_level)

    if distributed:
        model = torch.nn.parallel.DistributedDataParallel(
            model, device_ids=[local_rank], output_device=local_rank,
            # this should be removed if we update BatchNorm stats
            broadcast_buffers=False,
        )

    arguments = {}
    arguments["iteration"] = 0

    output_dir = cfg.OUTPUT_DIR

    save_to_disk = get_rank() == 0
    checkpointer = DetectronCheckpointer(
        cfg, model, optimizer, scheduler, output_dir, save_to_disk
    )
    extra_checkpoint_data = checkpointer.load(cfg.MODEL.WEIGHT)
    arguments.update(extra_checkpoint_data)

    data_loader = make_data_loader(
        cfg,
        is_train=True,
        is_distributed=distributed,
        start_iter=arguments["iteration"],
    )

    checkpoint_period = cfg.SOLVER.CHECKPOINT_PERIOD

    do_train(
        model,
        data_loader,
        optimizer,
        scheduler,
        checkpointer,
        device,
        checkpoint_period,
        arguments,
    )

    return model
Example #26
Source File: train_net.py    From maskscoring_rcnn with MIT License 4 votes vote down vote up 
def train(cfg, local_rank, distributed):
    model = build_detection_model(cfg)
    device = torch.device(cfg.MODEL.DEVICE)
    model.to(device)

    optimizer = make_optimizer(cfg, model)
    scheduler = make_lr_scheduler(cfg, optimizer)

    if distributed:
        model = torch.nn.parallel.deprecated.DistributedDataParallel(
            model, device_ids=[local_rank], output_device=local_rank,
            # this should be removed if we update BatchNorm stats
            broadcast_buffers=False,
        )

    arguments = {}
    arguments["iteration"] = 0

    output_dir = cfg.OUTPUT_DIR

    save_to_disk = get_rank() == 0
    checkpointer = DetectronCheckpointer(
        cfg, model, optimizer, scheduler, output_dir, save_to_disk
    )
    extra_checkpoint_data = checkpointer.load(cfg.MODEL.WEIGHT)
    arguments.update(extra_checkpoint_data)

    data_loader = make_data_loader(
        cfg,
        is_train=True,
        is_distributed=distributed,
        start_iter=arguments["iteration"],
    )

    checkpoint_period = cfg.SOLVER.CHECKPOINT_PERIOD

    do_train(
        model,
        data_loader,
        optimizer,
        scheduler,
        checkpointer,
        device,
        checkpoint_period,
        arguments,
    )

    return model
Example #27
Source File: train_net.py    From training with Apache License 2.0 4 votes vote down vote up 
def train(cfg, local_rank, distributed):
    model = build_detection_model(cfg)
    device = torch.device(cfg.MODEL.DEVICE)
    model.to(device)

    optimizer = make_optimizer(cfg, model)
    scheduler = make_lr_scheduler(cfg, optimizer)

    if distributed:
        model = torch.nn.parallel.DistributedDataParallel(
            model, device_ids=[local_rank], output_device=local_rank,
            # this should be removed if we update BatchNorm stats
            broadcast_buffers=False,
        )

    arguments = {}
    arguments["iteration"] = 0

    output_dir = cfg.OUTPUT_DIR

    save_to_disk = get_rank() == 0
    checkpointer = DetectronCheckpointer(
        cfg, model, optimizer, scheduler, output_dir, save_to_disk
    )
    extra_checkpoint_data = checkpointer.load(cfg.MODEL.WEIGHT)
    arguments.update(extra_checkpoint_data)

    data_loader = make_data_loader(
        cfg,
        is_train=True,
        is_distributed=distributed,
        start_iter=arguments["iteration"],
    )

    checkpoint_period = cfg.SOLVER.CHECKPOINT_PERIOD

    do_train(
        model,
        data_loader,
        optimizer,
        scheduler,
        checkpointer,
        device,
        checkpoint_period,
        arguments,
    )

    return model
Example #28
Source File: train_net.py    From retinamask with MIT License 4 votes vote down vote up 
def train(cfg, local_rank, distributed):
    model = build_detection_model(cfg)
    device = torch.device(cfg.MODEL.DEVICE)
    model.to(device)

    optimizer = make_optimizer(cfg, model)
    scheduler = make_lr_scheduler(cfg, optimizer)

    if distributed:
        model = torch.nn.parallel.deprecated.DistributedDataParallel(
            model, device_ids=[local_rank], output_device=local_rank,
            # this should be removed if we update BatchNorm stats
            broadcast_buffers=False,
        )

    arguments = {}
    arguments["iteration"] = 0

    output_dir = cfg.OUTPUT_DIR

    save_to_disk = get_rank() == 0
    checkpointer = DetectronCheckpointer(
        cfg, model, optimizer, scheduler, output_dir, save_to_disk
    )
    extra_checkpoint_data = checkpointer.load(cfg.MODEL.WEIGHT)
    arguments.update(extra_checkpoint_data)

    data_loader = make_data_loader(
        cfg,
        is_train=True,
        is_distributed=distributed,
        start_iter=arguments["iteration"],
    )

    checkpoint_period = cfg.SOLVER.CHECKPOINT_PERIOD

    do_train(
        model,
        data_loader,
        optimizer,
        scheduler,
        checkpointer,
        device,
        checkpoint_period,
        arguments,
    )

    return model
Example #29
Source File: train_net.py    From FreeAnchor with MIT License 4 votes vote down vote up 
def train(cfg, local_rank, distributed):
    model = build_detection_model(cfg)
    device = torch.device(cfg.MODEL.DEVICE)
    model.to(device)

    optimizer = make_optimizer(cfg, model)
    scheduler = make_lr_scheduler(cfg, optimizer)

    if distributed:
        model = torch.nn.parallel.DistributedDataParallel(
            model, device_ids=[local_rank], output_device=local_rank,
            # this should be removed if we update BatchNorm stats
            broadcast_buffers=False,
        )

    arguments = {}
    arguments["iteration"] = 0

    output_dir = cfg.OUTPUT_DIR

    save_to_disk = get_rank() == 0
    checkpointer = DetectronCheckpointer(
        cfg, model, optimizer, scheduler, output_dir, save_to_disk
    )
    extra_checkpoint_data = checkpointer.load(cfg.MODEL.WEIGHT)
    arguments.update(extra_checkpoint_data)

    data_loader = make_data_loader(
        cfg,
        is_train=True,
        is_distributed=distributed,
        start_iter=arguments["iteration"],
    )

    checkpoint_period = cfg.SOLVER.CHECKPOINT_PERIOD

    do_train(
        model,
        data_loader,
        optimizer,
        scheduler,
        checkpointer,
        device,
        checkpoint_period,
        arguments,
    )

    return model
Example #30
Source File: train_net.py    From HRNet-MaskRCNN-Benchmark with MIT License 4 votes vote down vote up 
def train(cfg, local_rank, distributed):
    model = build_detection_model(cfg)
    device = torch.device(cfg.MODEL.DEVICE)
    model.to(device)
    if cfg.SOLVER.ENABLE_FP16:
        model.half()

    optimizer = make_optimizer(cfg, model)
    scheduler = make_lr_scheduler(cfg, optimizer)

    if distributed:
        model = torch.nn.parallel.DistributedDataParallel(
            model, device_ids=[local_rank], output_device=local_rank,
            # this should be removed if we update BatchNorm stats
            # broadcast_buffers=False,
        )
    arguments = {}
    arguments["iteration"] = 0

    output_dir = cfg.OUTPUT_DIR

    save_to_disk = get_rank() == 0
    checkpointer = DetectronCheckpointer(
        cfg, model, optimizer, scheduler, output_dir, save_to_disk
    )
    print(cfg.MODEL.WEIGHT)
    extra_checkpoint_data = checkpointer.load(cfg.MODEL.WEIGHT)
    arguments.update(extra_checkpoint_data)

    data_loader = make_data_loader(
        cfg,
        is_train=True,
        is_distributed=distributed,
        start_iter=arguments["iteration"],
    )

    checkpoint_period = cfg.SOLVER.CHECKPOINT_PERIOD

    do_train(
        model,
        data_loader,
        optimizer,
        scheduler,
        checkpointer,
        device,
        checkpoint_period,
        arguments,
        cfg
    )

    return model