remotebiosensing / rppg

Benchmark Framework for fair evaluation of rPPG
Apache License 2.0
260 stars 32 forks source link

some question about interpreting result.csv file in the path of [rppg/result/csv/result.csv] #53

Closed aytekin827 closed 1 year ago

aytekin827 commented 1 year ago

Hello.

I have some question for in the file at [rppg/result/csv/result.csv]

  1. what does each column mean in 'result.csv' file?

    [example] row : 'BigSmall_PURE_PURE_72_30_10' used model : BigSmall training dataset : PURE test dataset : PURE Image Sise : 72 EVAL_TIME_LENGTH : 30 ??? : 10

    • what does the 10 mean in this row?
    • and the other is right?
  2. All of the model combinations are trained and tested by your teams?

  3. How can I read 'calc_comp.csv'file? I found all the file in this repository but cannot found any description about this.

Thank you and your team for this great job!!

paperchae commented 1 year ago

Hey!

Thanks for your interest.

The last column supposed to mean the epochs for traing.

Sorry for the confusion LOL

Hope you have a great day!

John chae

On Wed, Aug 9, 2023 at 6:11 PM aytekin827 @.***> wrote:

Hello.

I have some question for in the file at [rppg/result/csv/result.csv]

  1. what does each column mean in 'result.csv' file?

[example] row : 'BigSmall_PURE_PURE_72_30_10' used model : BigSmall training dataset : PURE test dataset : PURE Image Sise : 72 EVAL_TIME_LENGTH : 30 ??? : 10

  • what does the 10 mean in this row?
  • and the other is right?

    1.

    All of the model combinations are trained and tested by your teams? 2.

    How can I read 'calc_comp.csv'file? I found all the file in this repository but cannot found any description about this.

thank you and your team for this great job!!

and

— Reply to this email directly, view it on GitHub https://github.com/remotebiosensing/rppg/issues/53, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALV7SLHN3DSITQOFFY3NXA3XUNH5VANCNFSM6AAAAAA3JV47UI . You are receiving this because you are subscribed to this thread.Message ID: @.***>

SpicyYeol commented 1 year ago

Hello

  1. what does each column mean in 'result.csv' file?

[example] row : 'BigSmall_PURE_PURE_72_30_10' used model : BigSmall training dataset : PURE test dataset : PURE Image Sise : 72 EVAL_TIME_LENGTH : 30 ??? : 10

=>30 refers to the model's output length, and 10 refers to the hour evaluation time window length. We are not using result.csv now. Perhaps it will converge to only one file, calc_comp.csv.

1.

All of the model combinations are trained and tested by your teams?

=> Not yet. We have tested all non-DNN methods, along with deepphys, mtts, efficientphys, bigsmall (not multimodal, only heart rate), and physnet 2.

How can I read 'calc_comp.csv'file? I found all the file in this repository but cannot found any description about this.

=> Now we are preparing for the alpha release, so we don't have any documentation for this project

Methods_trainset_testset_hr timewindow

2023년 8월 9일 (수) 오후 6:11, aytekin827 @.***>님이 작성:

Hello.

I have some question for in the file at [rppg/result/csv/result.csv]

  1. what does each column mean in 'result.csv' file?

[example] row : 'BigSmall_PURE_PURE_72_30_10' used model : BigSmall training dataset : PURE test dataset : PURE Image Sise : 72 EVAL_TIME_LENGTH : 30 ??? : 10

  • what does the 10 mean in this row?
  • and the other is right?

    1.

    All of the model combinations are trained and tested by your teams? 2.

    How can I read 'calc_comp.csv'file? I found all the file in this repository but cannot found any description about this.

thank you and your team for this great job!!

and

— Reply to this email directly, view it on GitHub https://github.com/remotebiosensing/rppg/issues/53, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGQAS75SGVF3ZFFHAOLF6KTXUNH5VANCNFSM6AAAAAA3JV47UI . You are receiving this because you are subscribed to this thread.Message ID: @.***>

SpicyYeol commented 1 year ago

I apologize, the answer to the first question was correct jongeui's one.

2023년 8월 9일 (수) 오후 6:59, Kim Dae Yeol @.***>님이 작성:

Hello

  1. what does each column mean in 'result.csv' file?

[example] row : 'BigSmall_PURE_PURE_72_30_10' used model : BigSmall training dataset : PURE test dataset : PURE Image Sise : 72 EVAL_TIME_LENGTH : 30 ??? : 10

  • what does the 10 mean in this row?
  • and the other is right?

=>30 refers to the model's output length, and 10 refers to the hour evaluation time window length. We are not using result.csv now. Perhaps it will converge to only one file, calc_comp.csv.

1.

All of the model combinations are trained and tested by your teams?

=> Not yet. We have tested all non-DNN methods, along with deepphys, mtts, efficientphys, bigsmall (not multimodal, only heart rate), and physnet 2.

How can I read 'calc_comp.csv'file? I found all the file in this repository but cannot found any description about this.

=> Now we are preparing for the alpha release, so we don't have any documentation for this project

Methods_trainset_testset_hr timewindow

2023년 8월 9일 (수) 오후 6:11, aytekin827 @.***>님이 작성:

Hello.

I have some question for in the file at [rppg/result/csv/result.csv]

  1. what does each column mean in 'result.csv' file?

[example] row : 'BigSmall_PURE_PURE_72_30_10' used model : BigSmall training dataset : PURE test dataset : PURE Image Sise : 72 EVAL_TIME_LENGTH : 30 ??? : 10

  • what does the 10 mean in this row?
  • and the other is right?

    1.

    All of the model combinations are trained and tested by your teams? 2.

    How can I read 'calc_comp.csv'file? I found all the file in this repository but cannot found any description about this.

thank you and your team for this great job!!

and

— Reply to this email directly, view it on GitHub https://github.com/remotebiosensing/rppg/issues/53, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGQAS75SGVF3ZFFHAOLF6KTXUNH5VANCNFSM6AAAAAA3JV47UI . You are receiving this because you are subscribed to this thread.Message ID: @.***>

aytekin827 commented 1 year ago

thanks very much for help!