mxin262 / Bridging-Text-Spotting

(CVPR 2024) Bridging the Gap Between End-to-End and Two-Step Text Spotting.
https://arxiv.org/pdf/2404.04624.pdf
Other
48 stars 1 forks source link

[COCOEvaluator] Did not receive valid predictions. #7

Open kleinblue1 opened 5 months ago

kleinblue1 commented 5 months ago

请教一下,用了这个命令做evaluation:python tools/train_net.py --config-file configs/Bridge/TotalText/R_50_poly.yaml --eval-only MODEL.WEIGHTS totaltext_testr.pth,但是出来报错是[COCOEvaluator] Did not receive valid predictions. 先前已经按照要求弄好了测试文件 1

kleinblue1 commented 5 months ago

不好意思打扰了,问题已经解决了

Gavin-debug commented 5 months ago

@kleinblue1 同学你是用自己的的gpu跑的吗,我配环境配了两天还是有很多bug,所以想问一下能否请教你一下,看看你的最佳实践是怎么样的,预计不会超过半小时,我的邮箱是guodWU@163.com,如果方便我将不胜感激🥹。