yaml_cfg load error

[fisakhan]

I get the following error:

Traceback (most recent call last):
File "test_nyu_metric.py", line 22, in
merge_cfg_from_file(test_args)
File "/home/Projects/consistent_depth/VNL_Monocular_Depth_Prediction/lib/core/config.py", line 133, in merge_cfg_from_file
yaml_cfg = AttrDict(yaml.load(f))
TypeError: load() missing 1 required positional argument: 'Loader'

Following this, I replaced load with safe_load and then got the following error:

Traceback (most recent call last):
File "test_nyu_metric.py", line 22, in
merge_cfg_from_file(test_args)
File "/home/Projects/consistent_depth/VNL_Monocular_Depth_Prediction/lib/core/config.py", line 133, in merge_cfg_from_file
yaml_cfg = AttrDict(yaml.safe_load(f))
File "/home/Projects/consistent_depth/VNL_Monocular_Depth_Prediction/lib/utils/collections.py", line 8, in init
super(AttrDict, self).init(*args, **kwargs)
TypeError: 'NoneType' object is not iterable

[fisakhan]

When I solve the above problem then I get the following problem:

----------------- Options ---------------
batchsize: 2
cfg_file: lib/configs/resnext101_32x4d_nyudv2_class [default: lib/configs/resnext_32x4d_nyudv2_c1]
dataroot: ./datasets/NYUDV2 [default: None]
dataset: nyudv2
epoch: 30
load_ckpt: ./nyu_rawdata.pth [default: None]
phase: test
phase_anno: test
results_dir: ./evaluation
resume: False
start_epoch: 0
start_step: 0
thread: 4
use_tfboard: False
----------------- End -------------------
INFO nyudv2_dataset.py: 38: Loaded NYUDV2 data!
INFO load_dataset.py: 31: nyudv2 is created.
INFO test_nyu_metric.py: 26: test_data_size: 654
loading model...
cfg.MODEL.ENCODER: resnext101_32x4d_body_stride16
bottom_up_model: lateral_net.lateral_resnext101_32x4d_body_stride16
Traceback (most recent call last):
File "test_nyu_metric.py", line 30, in
model = MetricDepthModel()
File "/home/Projects/consistent_depth/VNL_Monocular_Depth_Prediction/lib/models/metric_depth_model.py", line 16, in init
self.depth_model = DepthModel()
File "/home/Projects/consistent_depth/VNL_Monocular_Depth_Prediction/lib/models/metric_depth_model.py", line 123, in init
self.decoder_modules = lateral_net.fcn_topdown(cfg.MODEL.ENCODER)
File "/home/Projects/consistent_depth/VNL_Monocular_Depth_Prediction/lib/models/lateral_net.py", line 190, in init
self.init_modules(self.init_type)
File "/home/Projects/consistent_depth/VNL_Monocular_Depth_Prediction/lib/models/lateral_net.py", line 193, in init_modules
self.init_weights(init_type)
File "/home/Projects/consistent_depth/VNL_Monocular_Depth_Prediction/lib/models/lateral_net.py", line 211, in init_weights
child_m.apply(init_func)
File "/home/.virtualenvs/consistent_depth/lib/python3.6/site-packages/torch/nn/modules/module.py", line 445, in apply
module.apply(fn)
File "/home/.virtualenvs/consistent_depth/lib/python3.6/site-packages/torch/nn/modules/module.py", line 446, in apply
fn(self)
File "/home/Projects/consistent_depth/VNL_Monocular_Depth_Prediction/lib/models/lateral_net.py", line 207, in init_func
nn.init.normal(m.weight.data, 1.0, 0.0)
File "/home/.virtualenvs/consistent_depth/lib/python3.6/site-packages/torch/nn/init.py", line 140, in normal
return no_grad_normal(tensor, mean, std)
File "/home/.virtualenvs/consistent_depth/lib/python3.6/site-packages/torch/nn/init.py", line 19, in no_grad_normal
return tensor.normal(mean, std)
RuntimeError: normal expects std > 0.0, but found std=0

I guess, it has something to do with __C.MODEL.INIT_TYPE = 'xavier'