使用Grounding dino demo的时候，一直卡住，有进度条但是一直没有结果

[aixiaodewugege]

我使用实例代码： python demo/image_demo.py demo/demo.jpg configs/grounding_dino/grounding_dino_swin-t_pretrain_obj365_goldg_cap4m.py --weights /mnt/sh_flex_storage/home/wushuche/mnist-fastapi-aio-triton/mmdetection/checkpoint/groundingdino_swint_ogc_mmdet-822d7e9d.pth --texts 'bench . car .'

显示如下： 进度条一直在晃动，但不给出结果

Loads checkpoint by local backend from path: /mnt/sh_flex_storage/home/wushuche/mnist-fastapi-aio-triton/mmdetection/checkpoint/groundingdino_swint_ogc_mmdet-822d7e9d.pth The model and loaded state dict do not match exactly

unexpected key in source state_dict: language_model.language_backbone.body.model.embeddings.position_ids

missing keys in source state_dict: dn_query_generator.label_embedding.weight

/mnt/sh_flex_storage/home/wushuche/mnist-fastapi-aio-triton/mmdetection/mmdet/apis/det_inferencer.py:130: UserWarning: dataset_meta or class names are not saved in the checkpoint's meta data, use COCO classes by default. warnings.warn( 12/05 17:30:42 - mmengine - WARNING - Failed to search registry with scope "mmdet" in the "function" registry tree. As a workaround, the current "function" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmdet" is a correct scope, or whether the registry is initialized. /mnt/sh_flex_storage/home/wushuche/anaconda3/envs/fastapi_triton/lib/python3.9/site-packages/mmengine/visualization/visualizer.py:196: UserWarning: Failed to add <class 'mmengine.visualization.vis_backend.LocalVisBackend'>, please provide the save_dir argument. warnings.warn(f'Failed to add {vis_backend.class}, ' Inference

[hhaAndroid]

估计是推理比较慢吧，我跑了没啥问题耶

[aixiaodewugege]

估计是推理比较慢吧，我跑了没啥问题耶

找到问题了，第一次运行的时候需要下载nltk里面的东西，没有网络的话就会卡住

[a773783082]

你好，我也遇到了这个问题，但是我下载好后放到/root/nltk_data下面，之前用还是好的，最近用又卡住了，看error好像是还在下载nltk

[nltk_data] Error loading punkt: <urlopen error [Errno 101] Network is
[nltk_data]     unreachable>
Inference ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━   

[gjd2017]

文件下载好后，将glip.py中：

try:

#     import nltk
#     nltk.download('punkt', download_dir='~/nltk_data')
#     nltk.download('averaged_perceptron_tagger', download_dir='~/nltk_data')
# except ImportError:
#     raise RuntimeError('nltk is not installed, please install it by: '
#                        'pip install nltk.')

注释掉 然后+ import nltk

这样就不会每次都要下载了。。。

[rookie1999]

，没有网络的话就会卡住

我也遇到这个问题了，你解决了吗