Error while finetuning with ResNet101 (KeyError: "filename 'storages' not found")

[hamzarawal]

I was finetuning my dataset with ResNet101. Testing works fine with the pre-trained weights, but when I start training, it gives the following error:

Preparing training data... Extracting image sizes... (It may take long time) Done!! done before filtering, there are 10022 images... after filtering, there are 10022 images... 10022 roidb entries Loading pretrained weights from data/pretrained_model/resnet101_caffe.pth Traceback (most recent call last): File "trainval_net.py", line 249, in fasterRCNN.create_architecture() File "/home/itu/hamza/thesis/faster-rcnn.pytorch/lib/model/faster_rcnn/faster_rcnn.py", line 135, in create_architecture self._init_modules() File "/home/itu/hamza/thesis/faster-rcnn.pytorch/lib/model/faster_rcnn/resnet.py", line 244, in _init_modules state_dict = torch.load(self.model_path) File "/home/itu/virtual_envs/frcnn/lib/python3.6/site-packages/torch/serialization.py", line 303, in load return _load(f, map_location, pickle_module) File "/home/itu/virtual_envs/frcnn/lib/python3.6/site-packages/torch/serialization.py", line 454, in _load return legacy_load(f) File "/home/itu/virtual_envs/frcnn/lib/python3.6/site-packages/torch/serialization.py", line 383, in legacy_load tar.extract('storages', path=tmpdir) File "/home/itu/virtual_envs/frcnn/lib/python3.6/tarfile.py", line 2041, in extract tarinfo = self.getmember(member) File "/home/itu/virtual_envs/frcnn/lib/python3.6/tarfile.py", line 1752, in getmember raise KeyError("filename %r not found" % name) KeyError: "filename 'storages' not found"

[Zpadger]

You need to modify the class name of the 48th line in'lib/datasets/pascal_voc.py', and modify it according to your data set category name