haofeixu / aanet

[CVPR'20] AANet: Adaptive Aggregation Network for Efficient Stereo Matching
Apache License 2.0
534 stars 102 forks source link

ValueError: cannot reshape array of size 6139 into shape (540,960) #65

Closed old-popsicle closed 3 years ago

old-popsicle commented 3 years ago

I encountered the following problems while running the program:

ValueError: cannot reshape array of size 6139 into shape (540,960)

QQ图片20210527152638

Excuse me, what went wrong with this?

old-popsicle commented 3 years ago

@haofeixu I would appreciate if you can tell me how to solve the problem

haofeixu commented 3 years ago

Hi @old-popsicle , could you please provide more info regarding the error, for example, the running script, thanks.

old-popsicle commented 3 years ago

This is how I run the program    “bash scripts/aanettrain.sh”     Because of the GPU, I put the batch size is set to 1, and an error occurs   ValueError: cannot reshape array of size 6139 into shape (540,960) I appreciate your willingness to help me solve this problem 

------------------ 原始邮件 ------------------ 发件人: "Haofei @.>; 发送时间: 2021年5月28日(星期五) 下午5:03 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [haofeixu/aanet] ValueError: cannot reshape array of size 6139 into shape (540,960) (#65)

Hi @old-popsicle , could you please provide more info regarding the error, for example, the running script, thanks.

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

haofeixu commented 3 years ago

It seems that the error occurs when loading the ground truth disparity of Scene Flow dataset, maybe you can check whether your dataset is broken or the data is not loaded properly (e.g., the file path is not correct).

old-popsicle commented 3 years ago

My friend encountered this problem when running psmnet before, but when he put the batch size is set to 1, the problem is solved. However, the problem I encountered happened to be in batch Size is set to 1. Maybe, as you said, there's something wrong with my data set. I appreciate your willingness to solve my problem. Thank you.

------------------ 原始邮件 ------------------ 发件人: "Haofei @.>; 发送时间: 2021年5月28日(星期五) 下午5:27 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [haofeixu/aanet] ValueError: cannot reshape array of size 6139 into shape (540,960) (#65)

It seems that the error occurs when loading the ground truth disparity of Scene Flow dataset, maybe you can check whether your dataset is broken or the data is not loaded properly (e.g., the file path is not correct).

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

old-popsicle commented 3 years ago

Hi,when I run predict.sh.the output image is a low density image.Could you help me solve it? I would appreciate it if you could help me solve it.

------------------ 原始邮件 ------------------ 发件人: "刘振国" @.>; 发送时间: 2021年5月28日(星期五) 晚上7:42 @*.**@*.>; @.>; 主题: 回复: [haofeixu/aanet] ValueError: cannot reshape array of size 6139 into shape (540,960) (#65)

My friend encountered this problem when running psmnet before, but when he put the batch size is set to 1, the problem is solved. However, the problem I encountered happened to be in batch Size is set to 1. Maybe, as you said, there's something wrong with my data set. I appreciate your willingness to solve my problem. Thank you.

------------------ 原始邮件 ------------------ 发件人: "Haofei @.>; 发送时间: 2021年5月28日(星期五) 下午5:27 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [haofeixu/aanet] ValueError: cannot reshape array of size 6139 into shape (540,960) (#65)

It seems that the error occurs when loading the ground truth disparity of Scene Flow dataset, maybe you can check whether your dataset is broken or the data is not loaded properly (e.g., the file path is not correct).

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