search

RobotLocomotion / pytorch-dense-correspondence

Code for "Dense Object Nets: Learning Dense Visual Object Descriptors By and For Robotic Manipulation"
https://arxiv.org/pdf/1806.08756.pdf
Other
559 stars 133 forks source link

Problem with training_tutorial.ipynb #200

Closed kevinkengne closed 5 years ago

kevinkengne commented 5 years ago

dense_correspondence/dataset/simple_datasets_test.ipynb runs fine. However, I get the following error when I run dense_correspondence/training/training_tutorial.ipynb:

TypeError Traceback (most recent call last)

in () 5 print "training descriptor of dimension %d" %(d) 6 train = DenseCorrespondenceTraining(dataset=dataset, config=train_config) ----> 7 train.run() 8 print "finished training descriptor of dimension %d" %(d) /home/kevink/code/dense_correspondence/training/training.pyc in run(self, loss_current_iteration, use_pretrained) 242 if not use_pretrained: 243 # create new network and optimizer --> 244 self._dcn = self.build_network() 245 self._optimizer = self._construct_optimizer(self._dcn.parameters()) 246 else: /home/kevink/code/dense_correspondence/training/training.pyc in build_network(self) 129 130 return DenseCorrespondenceNetwork.from_config(self._config['dense_correspondence_network'], --> 131 load_stored_params=False) 132 133 def _construct_optimizer(self, parameters): /home/kevink/code/dense_correspondence/network/dense_correspondence_network.pyc in from_config(config, load_stored_params, model_param_file) 412 config["backbone"]["resnet_name"] = "Resnet34_8s" 413 --> 414 fcn = DenseCorrespondenceNetwork.get_fcn(config) 415 416 if 'normalize' in config: /home/kevink/code/dense_correspondence/network/dense_correspondence_network.pyc in get_fcn(config) 373 if config["backbone"]["model_class"] == "Resnet": 374 resnet_model = config["backbone"]["resnet_name"] --> 375 fcn = getattr(resnet_dilated, resnet_model)(num_classes=config['descriptor_dimension']) 376 377 elif config["backbone"]["model_class"] == "Unet": /home/kevink/code/external/pytorch-segmentation-detection/pytorch_segmentation_detection/models/resnet_dilated.py in __init__(self, num_classes) 245 pretrained=True, 246 output_stride=8, --> 247 remove_avg_pool_layer=True) 248 249 # Randomly initialize the 1x1 Conv scoring layer /usr/local/lib/python2.7/dist-packages/torchvision/models/resnet.pyc in resnet34(pretrained, progress, **kwargs) 235 """ 236 return _resnet('resnet34', BasicBlock, [3, 4, 6, 3], pretrained, progress, --> 237 **kwargs) 238 239 /usr/local/lib/python2.7/dist-packages/torchvision/models/resnet.pyc in _resnet(arch, block, layers, pretrained, progress, **kwargs) 208 209 def _resnet(arch, block, layers, pretrained, progress, **kwargs): --> 210 model = ResNet(block, layers, **kwargs) 211 if pretrained: 212 state_dict = load_state_dict_from_url(model_urls[arch], TypeError: __init__() got an unexpected keyword argument 'fully_conv'
kevinkengne commented 5 years ago

I figured out what the problem was. Thank you

chrisole commented 4 years ago

Hi bro, I got the same error. Could u plz tell me how to solve it?