Gumpest / YOLOv5-Multibackbone-Compression

YOLOv5 Series Multi-backbone(TPH-YOLOv5, Ghostnet, ShuffleNetv2, Mobilenetv3Small, EfficientNetLite, PP-LCNet, SwinTransformer YOLO), Module(CBAM, DCN), Pruning (EagleEye, Network Slimming), Quantization (MQBench) and Deployment (TensorRT, ncnn) Compression Tool Box.
989 stars 201 forks source link

数据集labels,我用的yolov5的格式,为啥识别不了。 #60

Open bjw1107 opened 2 years ago

bjw1107 commented 2 years ago

模型用的yolov5xP2CBAM-Swin-BiFPN-SPP.yaml

Sa-UpWorld commented 2 years ago

模型用的yolov5xP2CBAM-Swin-BiFPN-SPP.yaml

你的Swin模型可以跑起来不报错吗

bjw1107 commented 2 years ago

可以,现在问题解决了

发自我的iPhone

------------------ 原始邮件 ------------------ 发件人: Sa-UpWorld @.> 发送时间: 2022年3月11日 11:10 收件人: Gumpest/YOLOv5-Multibackbone-Compression @.> 抄送: bjw1107 @.>, Author @.> 主题: Re: [Gumpest/YOLOv5-Multibackbone-Compression] 数据集labels,我用的yolov5的格式,为啥识别不了。 (Issue #60)

模型用的yolov5xP2CBAM-Swin-BiFPN-SPP.yaml

你的Swin模型可以跑起来不报错吗

— Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you authored the thread.Message ID: @.***>

Sa-UpWorld commented 2 years ago

兄弟,请问你的环境是什么,哪一款显卡训练的

bjw1107 commented 2 years ago

租了块titan xp , PyTorch  1.8.1 Python  3.8 Cuda  11.1

------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2022年3月11日(星期五) 晚上7:27 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [Gumpest/YOLOv5-Multibackbone-Compression] 数据集labels,我用的yolov5的格式,为啥识别不了。 (Issue #60)

兄弟,请问你的环境是什么,哪一款显卡训练的

— Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you authored the thread.Message ID: @.***>

PeterSmith1 commented 2 years ago

租了块titan xp , PyTorch  1.8.1 Python  3.8 Cuda  11.1 ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2022年3月11日(星期五) 晚上7:27 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [Gumpest/YOLOv5-Multibackbone-Compression] 数据集labels,我用的yolov5的格式,为啥识别不了。 (Issue #60) 兄弟,请问你的环境是什么,哪一款显卡训练的 — Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you authored the thread.Message ID: @.***>

我能问一下,你采用这个跑下来精度是多少,你有作者的预训练文件不

Myrault1998 commented 1 year ago

请问你是如何解决无法识别labels问题的,我下载了Visdrone数据集,用的EagleEye剪枝,AssertionError: train: No labels in xxxxxxxxxxxxxxxxx. Can not train without labels. See https://github.com/ultralytics/yolov5/wiki/Train-Custom-Data