Mmdetection: IndexError: list index out of range

Created on 23 Jun 2020 · 10Comments · Source: open-mmlab/mmdetection

I use my own dataset to train, when the first epoch finished, i got the error as follows:
: 0.0005, loss_rpn_bbox: 0.0036, loss_cls: 0.0183, acc: 99.2578, loss_bbox: 0.0498, loss: 0.0723
2020-06-24 00:20:53,044 - mmdet - INFO - Saving checkpoint at 1 epochs
[>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 158/158, 9.1 task/s, elapsed: 17s, ETA: 0sTraceback (most recent call last):
File "tools/train.py", line 153, in
main()
File "tools/train.py", line 149, in main
meta=meta)
File "/home/d1999/ZY/mmdetection/mmdet/apis/train.py", line 128, in train_detector
runner.run(data_loaders, cfg.workflow, cfg.total_epochs)
File "/home/d1999/anaconda3/envs/mmlab/lib/python3.7/site-packages/mmcv/runner/epoch_based_runner.py", line 122, in run
epoch_runner(data_loaders[i], *kwargs)
File "/home/d1999/anaconda3/envs/mmlab/lib/python3.7/site-packages/mmcv/runner/epoch_based_runner.py", line 46, in train
self.call_hook('after_train_epoch')
File "/home/d1999/anaconda3/envs/mmlab/lib/python3.7/site-packages/mmcv/runner/base_runner.py", line 282, in call_hook
getattr(hook, fn_name)(self)
File "/home/d1999/ZY/mmdetection/mmdet/core/evaluation/eval_hooks.py", line 28, in after_train_epoch
self.evaluate(runner, results)
File "/home/d1999/ZY/mmdetection/mmdet/core/evaluation/eval_hooks.py", line 32, in evaluate
results, logger=runner.logger, *self.eval_kwargs)
File "/home/d1999/ZY/mmdetection/mmdet/datasets/voc.py", line 47, in evaluate
logger=logger)
File "/home/d1999/ZY/mmdetection/mmdet/core/evaluation/mean_ap.py", line 388, in eval_map
mean_ap, eval_results, dataset, area_ranges, logger=logger)
File "/home/d1999/ZY/mmdetection/mmdet/core/evaluation/mean_ap.py", line 451, in print_map_summary
label_names[j], num_gts[i, j], results[j]['num_dets'],
IndexError: list index out of range
Is there an solutions? thanks!

Source

zhangyu96

Most helpful comment

我使用的配置文件是"configs/pascal_voc/faster_rcnn_r50_fpn_1x_voc0712.py",该文件中的第五行model = dict(roi_head=dict(bbox_head=dict(num_classes=20))),num_classes这块也要同步改成class的个数.

NiuCY on 1 Sep 2020

😄4 👍1

All 10 comments

How many classes are you using in your dataset?

azibit on 23 Jun 2020

3，do not include background

---Original---
From: "AZEEZ IDRIS"<[email protected]>
Date: Wed, Jun 24, 2020 04:24 AM
To: "open-mmlab/mmdetection"<[email protected]>;
Cc: "zhangyu96"<[email protected]>;"Author"<[email protected]>;
Subject: Re: [open-mmlab/mmdetection] IndexError: list index out of range (#3113)

How many classes are you using in your dataset?

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub, or unsubscribe.

zhangyu96 on 24 Jun 2020

check the config neck setting :class num