pose-tensorflow
pose-tensorflow copied to clipboard
How do you run test.py?
When I run test.py I get the following error:
[jalal@goku pose-tensorflow]$ TF_CUDNN_USE_AUTOTUNE=0 python test.py
/scratch/sjn/anaconda/lib/python3.6/importlib/_bootstrap.py:219: RuntimeWarning: compiletime version 3.5 of module 'tensorflow.python.framework.fast_tensor_util' does not match runtime version 3.6
return f(*args, **kwds)
/scratch/sjn/anaconda/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
from ._conv import register_converters as _register_converters
INFO:root:Config:
{'batch_size': 1,
'crop': False,
'crop_pad': 0,
'dataset': '/path/to/dataset.mat',
'dataset_type': 'mpii',
'display_iters': 20,
'fg_fraction': 0.25,
'global_scale': 0.8452830189,
'init_weights': '../../pretrained/resnet_v1_101.ckpt',
'intermediate_supervision': True,
'intermediate_supervision_layer': 12,
'location_refinement': True,
'locref_huber_loss': True,
'locref_loss_weight': 0.05,
'locref_stdev': 7.2801,
'log_dir': 'log',
'max_input_size': 850,
'mean_pixel': [123.68, 116.779, 103.939],
'mirror': True,
'multi_step': [[0.005, 10000],
[0.02, 430000],
[0.002, 730000],
[0.001, 1030000]],
'net_type': 'resnet_101',
'optimizer': 'sgd',
'pairwise_huber_loss': True,
'pairwise_loss_weight': 1.0,
'pairwise_predict': False,
'pairwise_stats_collect': False,
'pairwise_stats_fn': 'pairwise_stats.mat',
'pos_dist_thresh': 17,
'regularize': False,
'save_iters': 60000,
'scale_jitter_lo': 0.85,
'scale_jitter_up': 1.15,
'scoremap_dir': 'test',
'shuffle': True,
'snapshot_prefix': 'snapshot',
'sparse_graph': [],
'stride': 8.0,
'tensorflow_pairwise_order': True,
'use_gt_segm': False,
'video': False,
'video_batch': False,
'weigh_negatives': False,
'weigh_only_present_joints': False,
'weigh_part_predictions': False,
'weight_decay': 0.0001}
Traceback (most recent call last):
File "test.py", line 75, in <module>
test_net(not args.novis, args.cache)
File "test.py", line 20, in test_net
dataset = create_dataset(cfg)
File "/scratch2/body_pose/pose-tensorflow/dataset/factory.py", line 8, in create
data = MPII(cfg)
File "/scratch2/body_pose/pose-tensorflow/dataset/mpii.py", line 9, in __init__
super().__init__(cfg)
File "/scratch2/body_pose/pose-tensorflow/dataset/pose_dataset.py", line 89, in __init__
self.data = self.load_dataset() if cfg.dataset else []
File "/scratch2/body_pose/pose-tensorflow/dataset/pose_dataset.py", line 104, in load_dataset
mlab = sio.loadmat(file_name)
File "/scratch/sjn/anaconda/lib/python3.6/site-packages/scipy/io/matlab/mio.py", line 141, in loadmat
MR, file_opened = mat_reader_factory(file_name, appendmat, **kwargs)
File "/scratch/sjn/anaconda/lib/python3.6/site-packages/scipy/io/matlab/mio.py", line 64, in mat_reader_factory
byte_stream, file_opened = _open_file(file_name, appendmat)
TypeError: 'NoneType' object is not iterable
[jalal@goku pose-tensorflow]$ grep pose_cfg config.py
def load_config(filename = "models/mpii/train/pose_cfg.yaml"):
#def load_config(filename = "pose_cfg.yaml"):
[jalal@goku pose-tensorflow]$
according to https://stackoverflow.com/a/3887385/2414957 data in None.
So I ran it another time (I guess after I downloaded the resnet101 weights and modified the config.py to take the following pose_cfg.yaml available in models/mpii/train/pose_cfg.yaml
def load_config(filename = "models/mpii/train/pose_cfg.yaml"):
#def load_config(filename = "pose_cfg.yaml"):
if 'POSE_PARAM_PATH' in os.environ:
filename = os.environ['POSE_PARAM_PATH'] + '/' + filename
return cfg_from_file(filename)
and then I got the following error:
[jalal@goku pose-tensorflow]$ TF_CUDNN_USE_AUTOTUNE=0 python test.py
/scratch/sjn/anaconda/lib/python3.6/importlib/_bootstrap.py:219: RuntimeWarning: compiletime version 3.5 of module 'tensorflow.python.framework.fast_tensor_util' does not match runtime version 3.6
return f(*args, **kwds)
/scratch/sjn/anaconda/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
from ._conv import register_converters as _register_converters
INFO:root:Config:
{'batch_size': 1,
'crop': False,
'crop_pad': 0,
'dataset': '/scratch2/body_pose/pose-tensorflow/dataset/cropped/dataset.mat',
'dataset_type': 'mpii',
'display_iters': 20,
'fg_fraction': 0.25,
'global_scale': 0.8452830189,
'init_weights': '../../pretrained/resnet_v1_101.ckpt',
'intermediate_supervision': True,
'intermediate_supervision_layer': 12,
'location_refinement': True,
'locref_huber_loss': True,
'locref_loss_weight': 0.05,
'locref_stdev': 7.2801,
'log_dir': 'log',
'max_input_size': 850,
'mean_pixel': [123.68, 116.779, 103.939],
'mirror': True,
'multi_step': [[0.005, 10000],
[0.02, 430000],
[0.002, 730000],
[0.001, 1030000]],
'net_type': 'resnet_101',
'optimizer': 'sgd',
'pairwise_huber_loss': True,
'pairwise_loss_weight': 1.0,
'pairwise_predict': False,
'pairwise_stats_collect': False,
'pairwise_stats_fn': 'pairwise_stats.mat',
'pos_dist_thresh': 17,
'regularize': False,
'save_iters': 60000,
'scale_jitter_lo': 0.85,
'scale_jitter_up': 1.15,
'scoremap_dir': 'test',
'shuffle': True,
'snapshot_prefix': 'snapshot',
'sparse_graph': [],
'stride': 8.0,
'tensorflow_pairwise_order': True,
'use_gt_segm': False,
'video': False,
'video_batch': False,
'weigh_negatives': False,
'weigh_only_present_joints': False,
'weigh_part_predictions': False,
'weight_decay': 0.0001}
Traceback (most recent call last):
File "test.py", line 75, in <module>
test_net(not args.novis, args.cache)
File "test.py", line 21, in test_net
dataset.set_shuffle(False)
File "/scratch2/body_pose/pose-tensorflow/dataset/pose_dataset.py", line 147, in set_shuffle
assert not self.cfg.mirror
AssertionError
Can you please provide a little documentation on how to run the test.py? And what is the cfg it looks for? Seems the one I fed in is not matched.
For testing you need to use different pose_cfg.yaml file, which should look like this one: https://github.com/eldar/pose-tensorflow/blob/master/models/mpii/test/pose_cfg.yaml The test code is not really supported, what does work here is the demo code, which you can use as reference.
have you download the pretrained model?can you share it with me. because the download site cannot open.thank you very much!