How to run the code with batchsize >1

[ZhuYingJessica]

trafficstars

I'll be grateful if any one can give me some suggestions about how to run the code with batchsize greater than 1

[chrisway613]

in train.py, u can freely define the batch size in train().

[skyundead]

dataloader = data_.DataLoader(dataset,
batch_size=1,
shuffle=True,
# pin_memory=True, num_workers=opt.num_workers)

[chrisway613]

But I found if we set batch size > 1, it will raise an exception error..

[skyundead]

但是我发现如果我们将批处理大小设置为> 1，则会引发异常错误。

haha，me too。if batch_size = 2，it can‘t work。

[joqiaoy]

Have you solved this problem? I encountered this problem while training the model.

[ZhuYingJessica]

Sorry，i just changed another source code

---- 回复的原邮件 ---- | 发件人 | @.> | | 日期 | 2022年03月24日 16:43 | | 收件人 | @.> | | 抄送至 | @.@.> | | 主题 | Re: [chenyuntc/simple-faster-rcnn-pytorch] How to run the code with batchsize >1 (#181) |

Have you solved this problem? I encountered this problem while training the model.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

[Ayu000]

Is the cause of this problem the image size is not the same?

[ZhuYingJessica]

No，it seems that this code is not supportive for batchsize>1

---- 回复的原邮件 ---- | 发件人 | @.> | | 日期 | 2022年05月02日 10:45 | | 收件人 | @.> | | 抄送至 | @.@.> | | 主题 | Re: [chenyuntc/simple-faster-rcnn-pytorch] How to run the code with batchsize >1 (#181) |

Is the cause of this problem the image size is not the same?

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>