search

facebookresearch / DensePose

A real-time approach for mapping all human pixels of 2D RGB images to a 3D surface-based model of the body
http://densepose.org
Other
6.98k stars 1.3k forks source link

How to solve the 'EOFError' problem about retrain densepose #161

Open Easyfeng222 opened 5 years ago

Easyfeng222 commented 5 years ago

INFO train.py: 171: Loading dataset: ('dense_coco_2014_train', 'dense_coco_2014_valminusminival') loading annotations into memory... Done (t=11.17s) creating index... index created! .INFO json_dataset.py: 299: Loading proposals from: /tmp/detectron-download-cache/DensePose-RPN-train_fpn_resnet50.pkl Traceback (most recent call last): File "tools/train_net.py", line 119, in main() File "tools/train_net.py", line 101, in main checkpoints = detectron.utils.train.train_model() File "/home/feng/densepose/detectron/utils/train.py", line 50, in train_model setup_model_for_training(model, weights_file, output_dir) File "/home/feng/densepose/detectron/utils/train.py", line 149, in setup_model_for_training add_model_training_inputs(model) File "/home/feng/densepose/detectron/utils/train.py", line 173, in add_model_training_inputs cfg.TRAIN.DATASETS, cfg.TRAIN.PROPOSAL_FILES File "/home/feng/densepose/detectron/datasets/roidb.py", line 53, in combined_roidb_for_training roidbs = [get_roidb(*args) for args in zip(dataset_names, proposal_files)] File "/home/feng/densepose/detectron/datasets/roidb.py", line 38, in get_roidb crowd_filter_thresh=cfg.TRAIN.CROWD_FILTER_THRESH File "/home/feng/densepose/detectron/datasets/json_dataset.py", line 110, in get_roidb crowd_filter_thresh File "/home/feng/densepose/detectron/datasets/json_dataset.py", line 301, in _add_proposals_from_file proposals = pickle.load(f) EOFError

torpor29 commented 5 years ago

Hi, I met the same problem, did you figure out how to solve it?

pavithraezhil commented 5 years ago

Hi, I still have the same problem even after changing the weights to ResNet50. Please let me know if there is a fix.

Thank you.