deepsort模型ckpt.t7怎么重新训练？

[deeplearningers]

我要训练自己的数据，ckpt.t7需要重新训练吗？怎么训练啊？ i want to train my own datasets,does ckpt.t7 model need to retrain again? how to train it?

[Sarouch]

Hello, Same issue here, anyone has the answer please ? Thank you.

[yuninn]

I ran into the same problem. Do I need to retrain the weights? If I need to retrain, how do I label the data?

[JAYCHOU2020]

I ran into the same problem. Do I need to retrain the weights? If I need to retrain, how do I label the data?

help i also want to know!!!!!!!!thanks

[yuninn]

You can refer to this example，https://blog.csdn.net/qq_35975447/article/details/114412978

------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年7月28日(星期三) 中午11:37 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192)

I ran into the same problem. Do I need to retrain the weights? If I need to retrain, how do I label the data?

help i also want to know!!!!!!!!thanks

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

[Sarouch]

You can refer to this example，https://blog.csdn.net/qq_35975447/article/details/114412978 … ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年7月28日(星期三) 中午11:37 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192) I ran into the same problem. Do I need to retrain the weights? If I need to retrain, how do I label the data? help i also want to know!!!!!!!!thanks — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

Hello @yuninn , unfortunatelly I can't access to the hole web site, may be because I live in France, could you please copy the tutorial to me and send it to [email protected] ?

[yuninn]

My email was rejected by your mailbox

------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年8月6日(星期五) 凌晨3:02 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192)

You can refer to this example，https://blog.csdn.net/qq_35975447/article/details/114412978 … ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年7月28日(星期三) 中午11:37 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192) I ran into the same problem. Do I need to retrain the weights? If I need to retrain, how do I label the data? help i also want to know!!!!!!!!thanks — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

Hello @yuninn , unfortunatelly I can't access to the hole web site, may be because I live in France, could you please copy the tutorial to me and send it to @.*** ?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

[JAYCHOU2020]

My email was rejected by your mailbox … ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年8月6日(星期五) 凌晨3:02 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192) You can refer to this example，https://blog.csdn.net/qq_35975447/article/details/114412978 … ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年7月28日(星期三) 中午11:37 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192) I ran into the same problem. Do I need to retrain the weights? If I need to retrain, how do I label the data? help i also want to know!!!!!!!!thanks — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Hello @yuninn , unfortunatelly I can't access to the hole web site, may be because I live in France, could you please copy the tutorial to me and send it to @.*** ? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. when i train deepsort maybe my target is too small, how to fix it please?thanks

[JAYCHOU2020]

My email was rejected by your mailbox … ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年8月6日(星期五) 凌晨3:02 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192) You can refer to this example，https://blog.csdn.net/qq_35975447/article/details/114412978 … ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年7月28日(星期三) 中午11:37 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192) I ran into the same problem. Do I need to retrain the weights? If I need to retrain, how do I label the data? help i also want to know!!!!!!!!thanks — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Hello @yuninn , unfortunatelly I can't access to the hole web site, may be because I live in France, could you please copy the tutorial to me and send it to @.*** ? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

when i modify torchvision.transforms.RandomCrop((128,64)) to torchvision.transforms.RandomCrop((30,20)) i meet this mistake

[yuninn]

transform_train = torchvision.transforms.Compose([     torchvision.transforms.Resize((128, 64)),     torchvision.transforms.RandomCrop((128, 64), padding=4),     torchvision.transforms.RandomHorizontalFlip(),     torchvision.transforms.ToTensor(),     torchvision.transforms.Normalize(         [0.485, 0.456, 0.406], [0.229, 0.224, 0.225]) ])

train.py  resize your data to(128,64)

------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年8月6日(星期五) 下午4:21 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192)

My email was rejected by your mailbox … ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年8月6日(星期五) 凌晨3:02 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192) You can refer to this example，https://blog.csdn.net/qq_35975447/article/details/114412978 … ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年7月28日(星期三) 中午11:37 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192) I ran into the same problem. Do I need to retrain the weights? If I need to retrain, how do I label the data? help i also want to know!!!!!!!!thanks — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Hello @yuninn , unfortunatelly I can't access to the hole web site, may be because I live in France, could you please copy the tutorial to me and send it to @.*** ? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

when i modify torchvision.transforms.RandomCrop((128,64)) to torchvision.transforms.RandomCrop((30,20)) i meet this mistake

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

[Sarouch]

@yuninn Try this please : [email protected] Thank you

[JAYCHOU2020]

transform_train = torchvision.transforms.Compose([     torchvision.transforms.Resize((128, 64)),     torchvision.transforms.RandomCrop((128, 64), padding=4),     torchvision.transforms.RandomHorizontalFlip(),     torchvision.transforms.ToTensor(),     torchvision.transforms.Normalize(         [0.485, 0.456, 0.406], [0.229, 0.224, 0.225]) ]) train.py  resize your data to(128,64) … ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年8月6日(星期五) 下午4:21 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192) My email was rejected by your mailbox … ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年8月6日(星期五) 凌晨3:02 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192) You can refer to this example，https://blog.csdn.net/qq_35975447/article/details/114412978 … ------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2021年7月28日(星期三) 中午11:37 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [ZQPei/deep_sort_pytorch] deepsort模型ckpt.t7怎么重新训练？ (#192) I ran into the same problem. Do I need to retrain the weights? If I need to retrain, how do I label the data? help i also want to know!!!!!!!!thanks — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Hello @yuninn , unfortunatelly I can't access to the hole web site, may be because I live in France, could you please copy the tutorial to me and send it to @.*** ? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. when i modify torchvision.transforms.RandomCrop((128,64)) to torchvision.transforms.RandomCrop((30,20)) i meet this mistake — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

thank u very much !!! I'll try it