batch_size=mpii.get_length(VALID_MODE),
shuffle=False)
logger.debug('Pre-loading MPII validation data...')
[x_val], [p_val, afmat_val, head_val] = mpii_val[0]
"""Define a loader for PennAction test samples. """
penn_te = BatchLoader(penn_seq, ['frame'], ['pennaction'],
TEST_MODE,
batch_size=1,
shuffle=False)
"""Evaluate on 2D action recognition (PennAction)."""
s = eval_singleclip_generator(models[1], penn_te, logdir=logdir)
logger.debug('Best score on PennAction (single-clip): ')
logger.debug(str(s))
s = eval_multiclip_dataset(models[1],
penn_seq,
subsampling=pennaction_dataconf.fixed_subsampling,
logdir=logdir)
logger.debug('Best score on PennAction (multi-clip): ')
logger.debug(str(s))
# MPII EVALUATION
pose_pred = np.zeros_like(p_val)
mode = VALID_MODE
"""Evaluate on 2D pose estimation (MPII)."""
s, y_pred = eval_singleperson_pckh(models[0], x_val, p_val[:, :, 0:2],
afmat_val, head_val)
logger.debug('Best score on MPII: ')
logger.debug(str(s))
pose_pred = get_pred_data(mpii, pose_pred=pose_pred, pred=y_pred, mode=mode)
num_joints,
num_blocks,
pose_dim=2,
pose_net_version='v1',
full_trainable=False)
"""Load pre-trained model."""
weights_path = get_file(weights_file,
TF_WEIGHTS_PATH,
md5_hash=md5_hash,
cache_subdir='models')
model.load_weights(weights_path)
"""Load PennAction dataset."""
penn_seq = PennAction('datasets/PennAction',
pennaction_dataconf,
poselayout=pa16j2d,
topology='sequences',
use_gt_bbox=use_bbox,
clip_size=num_frames)
penn_te = BatchLoader(penn_seq, ['frame'], ['pennaction'],
TEST_MODE,
batch_size=1,
shuffle=False)
printcn(OKGREEN, 'Evaluation on PennAction multi-clip using predicted bboxes')
eval_multiclip_dataset(
model,
penn_seq,
bboxes_file='datasets/PennAction/penn_pred_bboxes_16f.json',
logdir=logdir)
num_actions,
input_shape,
num_frames,
num_joints,
num_blocks,
pose_dim=2,
pose_net_version='v1',
full_trainable=False)
"""Load pre-trained model."""
weights_path = get_file(weights_file,
TF_WEIGHTS_PATH,
md5_hash=md5_hash,
cache_subdir='models')
model.load_weights(weights_path)
"""Load PennAction dataset."""
penn_seq = PennAction('E:\Bachelorarbeit-SS20\datasets\PennAction',
pennaction_dataconf,
poselayout=pa16j2d,
topology='sequences',
use_gt_bbox=use_bbox,
clip_size=num_frames,
pred_bboxes_file='penn_pred_bboxes_16f.json')
penn_te = BatchLoader(penn_seq, ['frame'], ['pennaction'],
TEST_MODE,
batch_size=1,
shuffle=False)
printcn(OKGREEN, 'Evaluation on PennAction multi-clip using predicted bboxes')
eval_multiclip_dataset(model, penn_seq, subsampling=2, logdir=logdir)