Adan

icon company CAS, Information Office of Shanghai icon location Beijing & Zhejiang , China icon location Your eyes, your world! {The right answer is usually more than one, not all things that can be calculated can be counted}

Results 13 comments of Adan

How can I use MultiGPU?

You use this method to achieve，eg：os.environ["CUDA_VISIBLE_DEVICES"] = "0"

How can I use MultiGPU?

你好，简单的使用这个调用是没法使用多GPU的，你要在数据加载的前面编写这段代码，我用一个GPU训练过没问题的，并且这个代码的train文件有你要自己修改数据划分部分和训练加载部分。 ------------------ 原始邮件 ------------------ 发件人: "任聪"; 发送时间: 2019年8月18日(星期天) 晚上6:27 收件人: "Adamdad/keras-YOLOv3-mobilenet"; 抄送: "山野龆龀";"Comment"; 主题: Re: [Adamdad/keras-YOLOv3-mobilenet] How can I use MultiGPU? (#18) I have use the "os.environ["CUDA_VISIBLE_DEVICES"] = "0,1,2,3", but...

auth ,i use kmeans.py get my anchor met valueRrror: File "mtrand.pyx", line 1115, in mtrand.RandomState.choice ValueError: a must be greater than 0

ths we kill qus

Training errors

谢谢。 Eric3911 [email protected] 电子名片新出VIP模板啦，快来体验>> 扫一扫, 快速添加名片到手机 -------------------------------------------------------------------------------- ------------------ 原始邮件 ------------------ 发件人: Karbon Chen 发送时间: 2019-01-19 14:43:36 收件人:Adamdad/keras-YOLOv3-mobilenet 抄送:Eric3911 ,Author 主题: Re: [Adamdad/keras-YOLOv3-mobilenet] Training errors (#22) 'logs/carMobilenet/001_Mobilenet_finetune/' may not exist, so...

when change the size of input image ,what need to change on train?? the anchor size??

if you train myself datas ,you only exchang to train.py 25line input_image ,this size must multiple of 32, dataset setup tell me ,we help you

License

Yes, welcome for your work is very！

Export onnx model error

ths > This below error occur when I run export.py. Could you check it? Thank you very much! > > TorchScript export failure: > Could not export Python function call...

Export onnx model error

This is a python package version issue

requirements

Mish is for acceleration and independent compilation. According to your question, I know you should check your version of Python

你好，请问我如何能得到000044test.jpg这样一张图

你点进去下载就行！