search

lzhbrian / deepfashion2-kps-agg-finetune

1st place solution (Team StylingAI Inc. & PKU AIIC) for CVPR 2020 DeepFashion2 Clothes Landmark Detection Track. Aggregation and Finetuning for Clothes Landmark Detection
https://arxiv.org/abs/2005.00419
MIT License
90 stars 13 forks source link

I meet some problems when i run it on win10 #1

Open lwhkop opened 4 years ago

lwhkop commented 4 years ago

Traceback (most recent call last): File "tools/train.py", line 248, in main() File "tools/train.py", line 137, in main normalize, File "E:\DeepFashion2_Dataset\HRNet-Human-Pose-Estimation\tools..\lib\dataset\deepfashion2.py", line 172, in init self.db = self._get_db() File "E:\DeepFashion2_Dataset\HRNet-Human-Pose-Estimation\tools..\lib\dataset\deepfashion2.py", line 197, in _get_db gt_db = self._load_coco_person_detection_results() File "E:\DeepFashion2_Dataset\HRNet-Human-Pose-Estimation\tools..\lib\dataset\deepfashion2.py", line 319, in _load_coco_person_detection_results det_res = all_boxes[n_img] KeyError: 0

how to solve it?

lzhbrian commented 4 years ago

Hi, the code for dataset preparation has not released yet. Seems like your code is having some problem with the dataset. If you can't wait, you can prepare the dataset as mentioned in the paper or readme.

lwhkop commented 4 years ago

Hi, the code for dataset preparation has not released yet. Seems like your code is having some problem with the dataset. If you can't wait, you can prepare the dataset as mentioned in the paper or readme.

I prepared my dataset as mentioned in https://github.com/switchablenorms/DeepFashion2/issues/13. But I found some differents with your validation_coco.json. Seems you just use validation['annotations'] as your validation_coco.json? And I couldn't find where the key 'score' exist. Would you mind telling me the format of your validation_coco.json?