TypeError: __init__() missing 2 required positional arguments: 'img_mean' and 'coord' about icnet-tensorflow HOT 2 CLOSED

Hey @npv0, Im sorry for that I have updated the wrong code. I have fixed it and it can work with evaluation now. Thanks.

from icnet-tensorflow.

Thanks for the fix @hellochick . Issue got resolved.

I have made a few changes locally to resolve few issues after checking out the latest [update

](6cd4fbf)

1. Changed Line 71 for param_name, data in data_dict[op_name].iteritems(): to for param_name, data in data_dict[op_name].items():

Traceback (most recent call last):
File "evaluate.py", line 155, in
main()
File "evaluate.py", line 126, in main
net.load(model_train30k, sess)
File "E:\Projects\DeepLearning\tf\ICNet\ICNet-tensorflow\network.py", line 71, in load
for param_name, data in data_dict[op_name].iteritems():
AttributeError: 'dict' object has no attribute 'iteritems'

2. I was getting issue as the cityscapes file names are difference in eval_list.txt for me.

npv0@LAPTOP-8FO36PI9 MINGW64 /e/Projects/DeepLearning/tf/ICNet/ICNet-tensorflow (master)
$ python3 evaluate.py --model=train
Namespace(flipped_eval=False, measure_time=False, model='train', save_dir='./output/')
Traceback (most recent call last):
File "evaluate.py", line 155, in
main()
File "evaluate.py", line 84, in main
coord)
File "E:\Projects\DeepLearning\tf\ICNet\ICNet-tensorflow\image_reader.py", line 107, in init
self.image_list, self.label_list = read_labeled_image_list(self.data_dir, self.data_list)
File "E:\Projects\DeepLearning\tf\ICNet\ICNet-tensorflow\image_reader.py", line 62, in read_labeled_image_list
raise ValueError('Failed to find file: ' + mask)
ValueError: Failed to find file: ./data/cityscapes_dataset/cityscape\gtFine/val/lindau/lindau_000000_000019_gtFine_labelTrainIds.png

I have changed eval_list.txt file (renamed all gtFine_labelTrainIds to gtFine_labelIds) for eg: lindau_000000_000019_gtFine_labelTrainIds.png
to lindau_000000_000019_gtFine_labelIds.png

Renaming resolved the error, but not sure whether the change is correct or not

from icnet-tensorflow.