PaddlePaddle / PaddleDetection

Object Detection toolkit based on PaddlePaddle. It supports object detection, instance segmentation, multiple object tracking and real-time multi-person keypoint detection.
Apache License 2.0
12.81k stars 2.89k forks source link

SSD模型代码ssd_vgg16_300_240e_voc.yml 训练代码无法运行 #8921

Open Zing-desire opened 7 months ago

Zing-desire commented 7 months ago

问题确认 Search before asking

Bug组件 Bug Component

Training

Bug描述 Describe the Bug

python -m paddle.distributed.launch tools/train.py -c configs/ssd/ssd_vgg16_300_240e_voc.yml

报错 Traceback (most recent call last): File "tools/train.py", line 209, in main() File "tools/train.py", line 205, in main run(FLAGS, cfg) File "tools/train.py", line 145, in run trainer = Trainer(cfg, mode='train') File "D:\zengying\code\PaddleDetection\ppdet\engine\trainer.py", line 100, in init self.loader = create('{}Reader'.format(capital_mode))( File "D:\zengying\code\PaddleDetection\ppdet\data\reader.py", line 168, in call self.dataset.parse_dataset() File "D:\zengying\code\PaddleDetection\ppdet\data\source\voc.py", line 111, in parse_dataset img_file, xml_file = [os.path.join(image_dir, x) \ ValueError: not enough values to unpack (expected 2, got 1) LAUNCH INFO 2024-04-14 23:55:19,245 Exit code 1

复现环境 Environment

最新的ppdet

Bug描述确认 Bug description confirmation

是否愿意提交PR? Are you willing to submit a PR?

lyuwenyu commented 7 months ago

请确保数据的路径配置正确