Python detectron.utils.c2.UnscopeName() Examples

The following are 30 code examples of detectron.utils.c2.UnscopeName(). 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 detectron.utils.c2 , or try the search function .
Example #1
Source File: test_restore_checkpoint.py    From CBNet with Apache License 2.0 6 votes vote down vote up 
def get_params(model):
    blobs = {}  # gpu_0 blobs with unscoped_name as key
    all_blobs = {}  # all blobs with scoped name as key
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    return blobs, all_blobs
Example #2
Source File: test_restore_checkpoint.py    From KL-Loss with Apache License 2.0 6 votes vote down vote up 
def get_params(model):
    blobs = {}  # gpu_0 blobs with unscoped_name as key
    all_blobs = {}  # all blobs with scoped name as key
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    return blobs, all_blobs
Example #3
Source File: test_restore_checkpoint.py    From Detectron-DA-Faster-RCNN with Apache License 2.0 6 votes vote down vote up 
def get_params(model):
    blobs = {}  # gpu_0 blobs with unscoped_name as key
    all_blobs = {}  # all blobs with scoped name as key
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    return blobs, all_blobs
Example #4
Source File: test_restore_checkpoint.py    From Clustered-Object-Detection-in-Aerial-Image with Apache License 2.0 6 votes vote down vote up 
def get_params(model):
    blobs = {}  # gpu_0 blobs with unscoped_name as key
    all_blobs = {}  # all blobs with scoped name as key
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    return blobs, all_blobs
Example #5
Source File: test_restore_checkpoint.py    From Detectron with Apache License 2.0 6 votes vote down vote up 
def get_params(model):
    blobs = {}  # gpu_0 blobs with unscoped_name as key
    all_blobs = {}  # all blobs with scoped name as key
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    return blobs, all_blobs
Example #6
Source File: test_restore_checkpoint.py    From Detectron-Cascade-RCNN with Apache License 2.0 6 votes vote down vote up 
def get_params(model):
    blobs = {}  # gpu_0 blobs with unscoped_name as key
    all_blobs = {}  # all blobs with scoped name as key
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if 'gpu_0' in scoped_name:
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
        all_blobs[scoped_name] = workspace.FetchBlob(scoped_name)
    return blobs, all_blobs
Example #7
Source File: net.py    From Detectron with Apache License 2.0 5 votes vote down vote up 
def save_model_to_weights_file(weights_file, model):
    """Stash model weights in a dictionary and pickle them to a file. We map
    GPU device scoped names to unscoped names (e.g., 'gpu_0/conv1_w' ->
    'conv1_w').
    """
    logger.info(
        'Saving parameters and momentum to {}'.format(
            os.path.abspath(weights_file)))
    blobs = {}
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save momentum
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save preserved blobs
    for scoped_name in workspace.Blobs():
        if scoped_name.startswith('__preserve__/'):
            unscoped_name = c2_utils.UnscopeName(scoped_name)
            if unscoped_name not in blobs:
                logger.debug(
                    ' {:s} -> {:s} (preserved)'.format(
                        scoped_name, unscoped_name))
                blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    cfg_yaml = envu.yaml_dump(cfg)
    save_object(dict(blobs=blobs, cfg=cfg_yaml), weights_file)
Example #8
Source File: test_restore_checkpoint.py    From CBNet with Apache License 2.0 5 votes vote down vote up 
def test_restore_checkpoint():
    # Create Model
    model = model_builder.create(cfg.MODEL.TYPE, train=True)
    add_momentum_init_ops(model)
    init_weights(model)
    # Fill input blobs
    roidb = combined_roidb_for_training(
        cfg.TRAIN.DATASETS, cfg.TRAIN.PROPOSAL_FILES
    )
    model_builder.add_training_inputs(model, roidb=roidb)
    workspace.CreateNet(model.net)
    # Bookkeeping for checkpoint creation
    iter_num = 0
    checkpoints = {}
    output_dir = get_output_dir(cfg.TRAIN.DATASETS, training=True)
    chk_file_path = os.path.join(output_dir, 'model_iter{}.pkl'.format(iter_num))
    checkpoints[iter_num] = chk_file_path
    # Save model weights
    nu.save_model_to_weights_file(checkpoints[iter_num], model)
    orig_gpu_0_params, orig_all_params = get_params(model)
    # Change the model weights
    init_weights(model)
    # Reload the weights in the model
    nu.initialize_gpu_from_weights_file(model, chk_file_path, gpu_id=0)
    nu.broadcast_parameters(model)
    shutil.rmtree(cfg.OUTPUT_DIR)
    _, restored_all_params = get_params(model)
    # Check if all params are loaded correctly
    for scoped_name, blob in orig_all_params.items():
        np.testing.assert_array_equal(blob, restored_all_params[scoped_name])
    # Check if broadcast_parameters works
    for scoped_name, blob in restored_all_params.items():
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        np.testing.assert_array_equal(blob, orig_gpu_0_params[unscoped_name])
Example #9
Source File: detector.py    From CBNet with Apache License 2.0 5 votes vote down vote up 
def AddLosses(self, losses):
        if not isinstance(losses, list):
            losses = [losses]
        # Conversion to str allows losses to include BlobReferences
        losses = [c2_utils.UnscopeName(str(l)) for l in losses]
        self.losses = list(set(self.losses + losses))
Example #10
Source File: net.py    From CBNet with Apache License 2.0 5 votes vote down vote up 
def save_model_to_weights_file(weights_file, model):
    """Stash model weights in a dictionary and pickle them to a file. We map
    GPU device scoped names to unscoped names (e.g., 'gpu_0/conv1_w' ->
    'conv1_w').
    """
    logger.info(
        'Saving parameters and momentum to {}'.format(
            os.path.abspath(weights_file)))
    blobs = {}
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save momentum
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save preserved blobs
    for scoped_name in workspace.Blobs():
        if scoped_name.startswith('__preserve__/'):
            unscoped_name = c2_utils.UnscopeName(scoped_name)
            if unscoped_name not in blobs:
                logger.debug(
                    ' {:s} -> {:s} (preserved)'.format(
                        scoped_name, unscoped_name))
                blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    cfg_yaml = yaml.dump(cfg)
    save_object(dict(blobs=blobs, cfg=cfg_yaml), weights_file)
Example #11
Source File: test_restore_checkpoint.py    From Detectron-DA-Faster-RCNN with Apache License 2.0 5 votes vote down vote up 
def test_restore_checkpoint():
    # Create Model
    model = model_builder.create(cfg.MODEL.TYPE, train=True)
    add_momentum_init_ops(model)
    init_weights(model)
    # Fill input blobs
    roidb = combined_roidb_for_training(
        cfg.TRAIN.DATASETS, cfg.TRAIN.PROPOSAL_FILES
    )
    model_builder.add_training_inputs(model, roidb=roidb)
    workspace.CreateNet(model.net)
    # Bookkeeping for checkpoint creation
    iter_num = 0
    checkpoints = {}
    output_dir = get_output_dir(cfg.TRAIN.DATASETS, training=True)
    chk_file_path = os.path.join(output_dir, 'model_iter{}.pkl'.format(iter_num))
    checkpoints[iter_num] = chk_file_path
    # Save model weights
    nu.save_model_to_weights_file(checkpoints[iter_num], model)
    orig_gpu_0_params, orig_all_params = get_params(model)
    # Change the model weights
    init_weights(model)
    # Reload the weights in the model
    nu.initialize_gpu_from_weights_file(model, chk_file_path, gpu_id=0)
    nu.broadcast_parameters(model)
    shutil.rmtree(cfg.OUTPUT_DIR)
    _, restored_all_params = get_params(model)
    # Check if all params are loaded correctly
    for scoped_name, blob in orig_all_params.items():
        np.testing.assert_array_equal(blob, restored_all_params[scoped_name])
    # Check if broadcast_parameters works
    for scoped_name, blob in restored_all_params.items():
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        np.testing.assert_array_equal(blob, orig_gpu_0_params[unscoped_name])
Example #12
Source File: detector.py    From Detectron-DA-Faster-RCNN with Apache License 2.0 5 votes vote down vote up 
def AddLosses(self, losses):
        if not isinstance(losses, list):
            losses = [losses]
        # Conversion to str allows losses to include BlobReferences
        losses = [c2_utils.UnscopeName(str(l)) for l in losses]
        self.losses = list(set(self.losses + losses))
Example #13
Source File: net.py    From Detectron-DA-Faster-RCNN with Apache License 2.0 5 votes vote down vote up 
def save_model_to_weights_file(weights_file, model):
    """Stash model weights in a dictionary and pickle them to a file. We map
    GPU device scoped names to unscoped names (e.g., 'gpu_0/conv1_w' ->
    'conv1_w').
    """
    logger.info(
        'Saving parameters and momentum to {}'.format(
            os.path.abspath(weights_file)))
    blobs = {}
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save momentum
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save preserved blobs
    for scoped_name in workspace.Blobs():
        if scoped_name.startswith('__preserve__/'):
            unscoped_name = c2_utils.UnscopeName(scoped_name)
            if unscoped_name not in blobs:
                logger.debug(
                    ' {:s} -> {:s} (preserved)'.format(
                        scoped_name, unscoped_name))
                blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    cfg_yaml = envu.yaml_dump(cfg)
    save_object(dict(blobs=blobs, cfg=cfg_yaml), weights_file)
Example #14
Source File: test_restore_checkpoint.py    From Detectron with Apache License 2.0 5 votes vote down vote up 
def test_restore_checkpoint():
    # Create Model
    model = model_builder.create(cfg.MODEL.TYPE, train=True)
    add_momentum_init_ops(model)
    init_weights(model)
    # Fill input blobs
    roidb = combined_roidb_for_training(
        cfg.TRAIN.DATASETS, cfg.TRAIN.PROPOSAL_FILES
    )
    model_builder.add_training_inputs(model, roidb=roidb)
    workspace.CreateNet(model.net)
    # Bookkeeping for checkpoint creation
    iter_num = 0
    checkpoints = {}
    output_dir = get_output_dir(cfg.TRAIN.DATASETS, training=True)
    chk_file_path = os.path.join(output_dir, 'model_iter{}.pkl'.format(iter_num))
    checkpoints[iter_num] = chk_file_path
    # Save model weights
    nu.save_model_to_weights_file(checkpoints[iter_num], model)
    orig_gpu_0_params, orig_all_params = get_params(model)
    # Change the model weights
    init_weights(model)
    # Reload the weights in the model
    nu.initialize_gpu_from_weights_file(model, chk_file_path, gpu_id=0)
    nu.broadcast_parameters(model)
    shutil.rmtree(cfg.OUTPUT_DIR)
    _, restored_all_params = get_params(model)
    # Check if all params are loaded correctly
    for scoped_name, blob in orig_all_params.items():
        np.testing.assert_array_equal(blob, restored_all_params[scoped_name])
    # Check if broadcast_parameters works
    for scoped_name, blob in restored_all_params.items():
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        np.testing.assert_array_equal(blob, orig_gpu_0_params[unscoped_name])
Example #15
Source File: detector.py    From Detectron with Apache License 2.0 5 votes vote down vote up 
def AddLosses(self, losses):
        if not isinstance(losses, list):
            losses = [losses]
        # Conversion to str allows losses to include BlobReferences
        losses = [c2_utils.UnscopeName(str(l)) for l in losses]
        self.losses = list(set(self.losses + losses))
Example #16
Source File: net.py    From KL-Loss with Apache License 2.0 5 votes vote down vote up 
def save_model_to_weights_file(weights_file, model):
    """Stash model weights in a dictionary and pickle them to a file. We map
    GPU device scoped names to unscoped names (e.g., 'gpu_0/conv1_w' ->
    'conv1_w').
    """
    logger.info(
        'Saving parameters and momentum to {}'.format(
            os.path.abspath(weights_file)))
    blobs = {}
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save momentum
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save preserved blobs
    for scoped_name in workspace.Blobs():
        if scoped_name.startswith('__preserve__/'):
            unscoped_name = c2_utils.UnscopeName(scoped_name)
            if unscoped_name not in blobs:
                logger.debug(
                    ' {:s} -> {:s} (preserved)'.format(
                        scoped_name, unscoped_name))
                blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    cfg_yaml = envu.yaml_dump(cfg)
    save_object(dict(blobs=blobs, cfg=cfg_yaml), weights_file)
Example #17
Source File: test_restore_checkpoint.py    From Detectron-Cascade-RCNN with Apache License 2.0 5 votes vote down vote up 
def test_restore_checkpoint():
    # Create Model
    model = model_builder.create(cfg.MODEL.TYPE, train=True)
    add_momentum_init_ops(model)
    init_weights(model)
    # Fill input blobs
    roidb = combined_roidb_for_training(
        cfg.TRAIN.DATASETS, cfg.TRAIN.PROPOSAL_FILES
    )
    model_builder.add_training_inputs(model, roidb=roidb)
    workspace.CreateNet(model.net)
    # Bookkeeping for checkpoint creation
    iter_num = 0
    checkpoints = {}
    output_dir = get_output_dir(cfg.TRAIN.DATASETS, training=True)
    chk_file_path = os.path.join(output_dir, 'model_iter{}.pkl'.format(iter_num))
    checkpoints[iter_num] = chk_file_path
    # Save model weights
    nu.save_model_to_weights_file(checkpoints[iter_num], model)
    orig_gpu_0_params, orig_all_params = get_params(model)
    # Change the model weights
    init_weights(model)
    # Reload the weights in the model
    nu.initialize_gpu_from_weights_file(model, chk_file_path, gpu_id=0)
    nu.broadcast_parameters(model)
    shutil.rmtree(cfg.OUTPUT_DIR)
    _, restored_all_params = get_params(model)
    # Check if all params are loaded correctly
    for scoped_name, blob in orig_all_params.items():
        np.testing.assert_array_equal(blob, restored_all_params[scoped_name])
    # Check if broadcast_parameters works
    for scoped_name, blob in restored_all_params.items():
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        np.testing.assert_array_equal(blob, orig_gpu_0_params[unscoped_name])
Example #18
Source File: detector.py    From Detectron-Cascade-RCNN with Apache License 2.0 5 votes vote down vote up 
def AddLosses(self, losses):
        if not isinstance(losses, list):
            losses = [losses]
        # Conversion to str allows losses to include BlobReferences
        losses = [c2_utils.UnscopeName(str(l)) for l in losses]
        self.losses = list(set(self.losses + losses))
Example #19
Source File: net.py    From Detectron-Cascade-RCNN with Apache License 2.0 5 votes vote down vote up 
def save_model_to_weights_file(weights_file, model):
    """Stash model weights in a dictionary and pickle them to a file. We map
    GPU device scoped names to unscoped names (e.g., 'gpu_0/conv1_w' ->
    'conv1_w').
    """
    logger.info(
        'Saving parameters and momentum to {}'.format(
            os.path.abspath(weights_file)))
    blobs = {}
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save momentum
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save preserved blobs
    for scoped_name in workspace.Blobs():
        if scoped_name.startswith('__preserve__/'):
            unscoped_name = c2_utils.UnscopeName(scoped_name)
            if unscoped_name not in blobs:
                logger.debug(
                    ' {:s} -> {:s} (preserved)'.format(
                        scoped_name, unscoped_name))
                blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    cfg_yaml = yaml.dump(cfg)
    save_object(dict(blobs=blobs, cfg=cfg_yaml), weights_file)
Example #20
Source File: test_restore_checkpoint.py    From Clustered-Object-Detection-in-Aerial-Image with Apache License 2.0 5 votes vote down vote up 
def test_restore_checkpoint():
    # Create Model
    model = model_builder.create(cfg.MODEL.TYPE, train=True)
    add_momentum_init_ops(model)
    init_weights(model)
    # Fill input blobs
    roidb = combined_roidb_for_training(
        cfg.TRAIN.DATASETS, cfg.TRAIN.PROPOSAL_FILES
    )
    model_builder.add_training_inputs(model, roidb=roidb)
    workspace.CreateNet(model.net)
    # Bookkeeping for checkpoint creation
    iter_num = 0
    checkpoints = {}
    output_dir = get_output_dir(cfg.TRAIN.DATASETS, training=True)
    chk_file_path = os.path.join(output_dir, 'model_iter{}.pkl'.format(iter_num))
    checkpoints[iter_num] = chk_file_path
    # Save model weights
    nu.save_model_to_weights_file(checkpoints[iter_num], model)
    orig_gpu_0_params, orig_all_params = get_params(model)
    # Change the model weights
    init_weights(model)
    # Reload the weights in the model
    nu.initialize_gpu_from_weights_file(model, chk_file_path, gpu_id=0)
    nu.broadcast_parameters(model)
    shutil.rmtree(cfg.OUTPUT_DIR)
    _, restored_all_params = get_params(model)
    # Check if all params are loaded correctly
    for scoped_name, blob in orig_all_params.items():
        np.testing.assert_array_equal(blob, restored_all_params[scoped_name])
    # Check if broadcast_parameters works
    for scoped_name, blob in restored_all_params.items():
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        np.testing.assert_array_equal(blob, orig_gpu_0_params[unscoped_name])
Example #21
Source File: detector.py    From Clustered-Object-Detection-in-Aerial-Image with Apache License 2.0 5 votes vote down vote up 
def AddLosses(self, losses):
        if not isinstance(losses, list):
            losses = [losses]
        # Conversion to str allows losses to include BlobReferences
        losses = [c2_utils.UnscopeName(str(l)) for l in losses]
        self.losses = list(set(self.losses + losses))
Example #22
Source File: net.py    From Clustered-Object-Detection-in-Aerial-Image with Apache License 2.0 5 votes vote down vote up 
def save_model_to_weights_file(weights_file, model):
    """Stash model weights in a dictionary and pickle them to a file. We map
    GPU device scoped names to unscoped names (e.g., 'gpu_0/conv1_w' ->
    'conv1_w').
    """
    logger.info(
        'Saving parameters and momentum to {}'.format(
            os.path.abspath(weights_file)))
    blobs = {}
    # Save all parameters
    for param in model.params:
        scoped_name = str(param)
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save momentum
    for param in model.TrainableParams():
        scoped_name = str(param) + '_momentum'
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        if unscoped_name not in blobs:
            logger.debug(' {:s} -> {:s}'.format(scoped_name, unscoped_name))
            blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    # Save preserved blobs
    for scoped_name in workspace.Blobs():
        if scoped_name.startswith('__preserve__/'):
            unscoped_name = c2_utils.UnscopeName(scoped_name)
            if unscoped_name not in blobs:
                logger.debug(
                    ' {:s} -> {:s} (preserved)'.format(
                        scoped_name, unscoped_name))
                blobs[unscoped_name] = workspace.FetchBlob(scoped_name)
    cfg_yaml = yaml.dump(cfg)
    save_object(dict(blobs=blobs, cfg=cfg_yaml), weights_file)
Example #23
Source File: test_restore_checkpoint.py    From KL-Loss with Apache License 2.0 5 votes vote down vote up 
def test_restore_checkpoint():
    # Create Model
    model = model_builder.create(cfg.MODEL.TYPE, train=True)
    add_momentum_init_ops(model)
    init_weights(model)
    # Fill input blobs
    roidb = combined_roidb_for_training(
        cfg.TRAIN.DATASETS, cfg.TRAIN.PROPOSAL_FILES
    )
    model_builder.add_training_inputs(model, roidb=roidb)
    workspace.CreateNet(model.net)
    # Bookkeeping for checkpoint creation
    iter_num = 0
    checkpoints = {}
    output_dir = get_output_dir(cfg.TRAIN.DATASETS, training=True)
    chk_file_path = os.path.join(output_dir, 'model_iter{}.pkl'.format(iter_num))
    checkpoints[iter_num] = chk_file_path
    # Save model weights
    nu.save_model_to_weights_file(checkpoints[iter_num], model)
    orig_gpu_0_params, orig_all_params = get_params(model)
    # Change the model weights
    init_weights(model)
    # Reload the weights in the model
    nu.initialize_gpu_from_weights_file(model, chk_file_path, gpu_id=0)
    nu.broadcast_parameters(model)
    shutil.rmtree(cfg.OUTPUT_DIR)
    _, restored_all_params = get_params(model)
    # Check if all params are loaded correctly
    for scoped_name, blob in orig_all_params.items():
        np.testing.assert_array_equal(blob, restored_all_params[scoped_name])
    # Check if broadcast_parameters works
    for scoped_name, blob in restored_all_params.items():
        unscoped_name = c2_utils.UnscopeName(scoped_name)
        np.testing.assert_array_equal(blob, orig_gpu_0_params[unscoped_name])
Example #24
Source File: detector.py    From KL-Loss with Apache License 2.0 5 votes vote down vote up 
def AddLosses(self, losses):
        if not isinstance(losses, list):
            losses = [losses]
        # Conversion to str allows losses to include BlobReferences
        losses = [c2_utils.UnscopeName(str(l)) for l in losses]
        self.losses = list(set(self.losses + losses))
Example #25
Source File: net.py    From Detectron with Apache License 2.0 4 votes vote down vote up 
def print_net(model, namescope='gpu_0'):
    """Print the model network."""
    logger.info('Printing model: {}'.format(model.net.Name()))
    op_list = model.net.Proto().op
    for op in op_list:
        input_name = op.input
        # For simplicity: only print the first output
        # Not recommended if there are split layers
        output_name = str(op.output[0])
        op_type = op.type
        op_name = op.name

        if namescope is None or output_name.startswith(namescope):
            # Only print the forward pass network
            if output_name.find('grad') >= 0 or output_name.find('__m') >= 0:
                continue

            try:
                # Under some conditions (e.g., dynamic memory optimization)
                # it is possible that the network frees some blobs when they are
                # no longer needed. Handle this case...
                output_shape = workspace.FetchBlob(output_name).shape
            except BaseException:
                output_shape = '<unknown>'

            first_blob = True
            op_label = op_type + (op_name if op_name == '' else ':' + op_name)
            suffix = ' ------- (op: {})'.format(op_label)
            for j in range(len(input_name)):
                if input_name[j] in model.params:
                    continue
                input_blob = workspace.FetchBlob(input_name[j])
                if isinstance(input_blob, np.ndarray):
                    input_shape = input_blob.shape
                    logger.info('{:28s}: {:20s} => {:28s}: {:20s}{}'.format(
                        c2_utils.UnscopeName(str(input_name[j])),
                        '{}'.format(input_shape),
                        c2_utils.UnscopeName(str(output_name)),
                        '{}'.format(output_shape),
                        suffix))
                    if first_blob:
                        first_blob = False
                        suffix = ' ------|'
    logger.info('End of model: {}'.format(model.net.Name()))
Example #26
Source File: net.py    From Detectron-Cascade-RCNN with Apache License 2.0 4 votes vote down vote up 
def print_net(model, namescope='gpu_0'):
    """Print the model network."""
    logger.info('Printing model: {}'.format(model.net.Name()))
    op_list = model.net.Proto().op
    for op in op_list:
        input_name = op.input
        # For simplicity: only print the first output
        # Not recommended if there are split layers
        output_name = str(op.output[0])
        op_type = op.type
        op_name = op.name

        if namescope is None or output_name.startswith(namescope):
            # Only print the forward pass network
            if output_name.find('grad') >= 0 or output_name.find('__m') >= 0:
                continue

            try:
                # Under some conditions (e.g., dynamic memory optimization)
                # it is possible that the network frees some blobs when they are
                # no longer needed. Handle this case...
                output_shape = workspace.FetchBlob(output_name).shape
            except BaseException:
                output_shape = '<unknown>'

            first_blob = True
            op_label = op_type + (op_name if op_name == '' else ':' + op_name)
            suffix = ' ------- (op: {})'.format(op_label)
            for j in range(len(input_name)):
                if input_name[j] in model.params:
                    continue
                input_blob = workspace.FetchBlob(input_name[j])
                if isinstance(input_blob, np.ndarray):
                    input_shape = input_blob.shape
                    logger.info('{:28s}: {:20s} => {:28s}: {:20s}{}'.format(
                        c2_utils.UnscopeName(str(input_name[j])),
                        '{}'.format(input_shape),
                        c2_utils.UnscopeName(str(output_name)),
                        '{}'.format(output_shape),
                        suffix))
                    if first_blob:
                        first_blob = False
                        suffix = ' ------|'
    logger.info('End of model: {}'.format(model.net.Name()))
Example #27
Source File: net.py    From Detectron-DA-Faster-RCNN with Apache License 2.0 4 votes vote down vote up 
def print_net(model, namescope='gpu_0'):
    """Print the model network."""
    logger.info('Printing model: {}'.format(model.net.Name()))
    op_list = model.net.Proto().op
    for op in op_list:
        input_name = op.input
        # For simplicity: only print the first output
        # Not recommended if there are split layers
        try:
            output_name = str(op.output[0])
        except BaseException:
            output_name = '<nothing>'
        op_type = op.type
        op_name = op.name

        if namescope is None or output_name.startswith(namescope):
            # Only print the forward pass network
            if output_name.find('grad') >= 0 or output_name.find('__m') >= 0:
                continue

            try:
                # Under some conditions (e.g., dynamic memory optimization)
                # it is possible that the network frees some blobs when they are
                # no longer needed. Handle this case...
                output_shape = workspace.FetchBlob(output_name).shape
            except BaseException:
                output_shape = '<unknown>'

            first_blob = True
            op_label = op_type + (op_name if op_name == '' else ':' + op_name)
            suffix = ' ------- (op: {})'.format(op_label)
            for j in range(len(input_name)):
                if input_name[j] in model.params:
                    continue
                input_blob = workspace.FetchBlob(input_name[j])
                if isinstance(input_blob, np.ndarray):
                    input_shape = input_blob.shape
                    logger.info('{:28s}: {:20s} => {:28s}: {:20s}{}'.format(
                        c2_utils.UnscopeName(str(input_name[j])),
                        '{}'.format(input_shape),
                        c2_utils.UnscopeName(str(output_name)),
                        '{}'.format(output_shape),
                        suffix))
                    if first_blob:
                        first_blob = False
                        suffix = ' ------|'
    logger.info('End of model: {}'.format(model.net.Name()))
Example #28
Source File: net.py    From Clustered-Object-Detection-in-Aerial-Image with Apache License 2.0 4 votes vote down vote up 
def print_net(model, namescope='gpu_0'):
    """Print the model network."""
    logger.info('Printing model: {}'.format(model.net.Name()))
    op_list = model.net.Proto().op
    for op in op_list:
        input_name = op.input
        # For simplicity: only print the first output
        # Not recommended if there are split layers
        output_name = str(op.output[0])
        op_type = op.type
        op_name = op.name

        if namescope is None or output_name.startswith(namescope):
            # Only print the forward pass network
            if output_name.find('grad') >= 0 or output_name.find('__m') >= 0:
                continue

            try:
                # Under some conditions (e.g., dynamic memory optimization)
                # it is possible that the network frees some blobs when they are
                # no longer needed. Handle this case...
                output_shape = workspace.FetchBlob(output_name).shape
            except BaseException:
                output_shape = '<unknown>'

            first_blob = True
            op_label = op_type + (op_name if op_name == '' else ':' + op_name)
            suffix = ' ------- (op: {})'.format(op_label)
            for j in range(len(input_name)):
                if input_name[j] in model.params:
                    continue
                input_blob = workspace.FetchBlob(input_name[j])
                if isinstance(input_blob, np.ndarray):
                    input_shape = input_blob.shape
                    logger.info('{:28s}: {:20s} => {:28s}: {:20s}{}'.format(
                        c2_utils.UnscopeName(str(input_name[j])),
                        '{}'.format(input_shape),
                        c2_utils.UnscopeName(str(output_name)),
                        '{}'.format(output_shape),
                        suffix))
                    if first_blob:
                        first_blob = False
                        suffix = ' ------|'
    logger.info('End of model: {}'.format(model.net.Name()))
Example #29
Source File: net.py    From CBNet with Apache License 2.0 4 votes vote down vote up 
def print_net(model, namescope='gpu_0'):
    """Print the model network."""
    logger.info('Printing model: {}'.format(model.net.Name()))
    op_list = model.net.Proto().op
    for op in op_list:
        input_name = op.input
        # For simplicity: only print the first output
        # Not recommended if there are split layers
        output_name = str(op.output[0])
        op_type = op.type
        op_name = op.name

        if namescope is None or output_name.startswith(namescope):
            # Only print the forward pass network
            if output_name.find('grad') >= 0 or output_name.find('__m') >= 0:
                continue

            try:
                # Under some conditions (e.g., dynamic memory optimization)
                # it is possible that the network frees some blobs when they are
                # no longer needed. Handle this case...
                output_shape = workspace.FetchBlob(output_name).shape
            except BaseException:
                output_shape = '<unknown>'

            first_blob = True
            op_label = op_type + (op_name if op_name == '' else ':' + op_name)
            suffix = ' ------- (op: {})'.format(op_label)
            for j in range(len(input_name)):
                if input_name[j] in model.params:
                    continue
                input_blob = workspace.FetchBlob(input_name[j])
                if isinstance(input_blob, np.ndarray):
                    input_shape = input_blob.shape
                    logger.info('{:28s}: {:20s} => {:28s}: {:20s}{}'.format(
                        c2_utils.UnscopeName(str(input_name[j])),
                        '{}'.format(input_shape),
                        c2_utils.UnscopeName(str(output_name)),
                        '{}'.format(output_shape),
                        suffix))
                    if first_blob:
                        first_blob = False
                        suffix = ' ------|'
    logger.info('End of model: {}'.format(model.net.Name()))
Example #30
Source File: net.py    From KL-Loss with Apache License 2.0 4 votes vote down vote up 
def print_net(model, namescope='gpu_0'):
    """Print the model network."""
    logger.info('Printing model: {}'.format(model.net.Name()))
    op_list = model.net.Proto().op
    for op in op_list:
        input_name = op.input
        # For simplicity: only print the first output
        # Not recommended if there are split layers
        output_name = str(op.output[0])
        op_type = op.type
        op_name = op.name

        if namescope is None or output_name.startswith(namescope):
            # Only print the forward pass network
            if output_name.find('grad') >= 0 or output_name.find('__m') >= 0:
                continue

            try:
                # Under some conditions (e.g., dynamic memory optimization)
                # it is possible that the network frees some blobs when they are
                # no longer needed. Handle this case...
                output_shape = workspace.FetchBlob(output_name).shape
            except BaseException:
                output_shape = '<unknown>'

            first_blob = True
            op_label = op_type + (op_name if op_name == '' else ':' + op_name)
            suffix = ' ------- (op: {})'.format(op_label)
            for j in range(len(input_name)):
                if input_name[j] in model.params:
                    continue
                input_blob = workspace.FetchBlob(input_name[j])
                if isinstance(input_blob, np.ndarray):
                    input_shape = input_blob.shape
                    logger.info('{:28s}: {:20s} => {:28s}: {:20s}{}'.format(
                        c2_utils.UnscopeName(str(input_name[j])),
                        '{}'.format(input_shape),
                        c2_utils.UnscopeName(str(output_name)),
                        '{}'.format(output_shape),
                        suffix))
                    if first_blob:
                        first_blob = False
                        suffix = ' ------|'
    logger.info('End of model: {}'.format(model.net.Name()))