EMS-YOLO
EMS-YOLO copied to clipboard
give_g1_data.py def build_dataset的sample回傳是空的list...
请问你解决这个问题了吗?处理coco和gen1数据集的loadImageAndLabels根本就一点都不一样哇,save_file也是空的
個人推薦別用了 要改很多東西
请问您成功训练事件数据集GEN1了吗可以借鉴下是怎么修改的吗?目前报错是这样的,如何生成的这个txt文件呢?源码里面好像是空的 Traceback (most recent call last): File "train_g1.py", line 639, in <module> main(opt) File "train_g1.py", line 536, in main train(opt.hyp, opt, device, callbacks) File "train_g1.py", line 213, in train train_loader, dataset = create_dataloader(train_path, opt.sample_size,opt.T,opt.image_shape, 'train', batch_size // WORLD_SIZE, gs, single_cls, File "/root/EMS-YOLO-main/g1-resnet/utils/datasets_g1T.py", line 122, in create_dataloader dataset = LoadImagesAndLabels(path, sample_size,T,image_shape, mode,batch=batch_size) File "/root/EMS-YOLO-main/g1-resnet/utils/datasets_g1T.py", line 410, in init with open(save_file_name) as t: FileNotFoundError: [Errno 2] No such file or directory: '/root/autodl-tmp/GEN_part_5/train_a.txt'
赵锦秀 @.***
------------------ 原始邮件 ------------------ 发件人: "BICLab/EMS-YOLO" @.>; 发送时间: 2024年4月10日(星期三) 下午5:37 @.>; @.@.>; 主题: Re: [BICLab/EMS-YOLO] give_g1_data.py def build_dataset的sample回傳是空的list... (Issue #14)
個人推薦別用了 要改很多東西
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>
同问这个问题,首先我们是要下载那个200多个GB的GEN1数据集吗?我对实验设置有些疑惑,我们真的是要完全使用那200GB的数据来训练?
That’s right
Yuguang Yang @.***>於 2024年6月13日 週四,下午2:06寫道:
同问这个问题,首先我们是要下载那个200多个GB的GEN1数据集吗?
— Reply to this email directly, view it on GitHub https://github.com/BICLab/EMS-YOLO/issues/14#issuecomment-2164521472, or unsubscribe https://github.com/notifications/unsubscribe-auth/ANIDFCRIFOF6K74CI46O2SDZHEZHLAVCNFSM6AAAAABFERYCGOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNRUGUZDCNBXGI . You are receiving this because you authored the thread.Message ID: @.***>
Thanks for your coming email