Wrong landmarks of full region

[WangXuan2401]

hi ~ thanks for your great work！ I am doing the NoW evaluation with full face region to decrease the error value. However, after getting BFM model of full region, the error evaluated in NoW was higher than the mesh without neck.

full region - median: 1.447974, mean: 2.707136, std: 4.441936

I think it is due to the wrong 7 landmarks, as the mesh shows correctly in MeshLab. But the landmarks in img were wrong

What can I do to get right landmarks of full region ? I really need your help

Thanks for your attention

[WangXuan2401]

I think it may be causded by mean_shape , but I am not sure and cannot solve it.

[lingleong981130]

Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem?

[WangXuan2401]

Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem?

I‘m very sorry that we have not solved the problem. Do u have any ideas?

[lingleong981130]

Hi, actually I get the correct output before from the code, but I not sure where I save the code ady...and after that I download again this repository...and its not working ady....

[lingleong981130]

Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem?

I‘m very sorry that we have not solved the problem. Do u have any ideas?

Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address

[WangXuan2401]

Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem?

I‘m very sorry that we have not solved the problem. Do u have any ideas?

Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address

wow ! It is excited to hear you have solved the problem ! My email address is [email protected] . And I am very appreciated for your kindness and great work!

[lingleong981130]

Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem?

I‘m very sorry that we have not solved the problem. Do u have any ideas?

Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address

wow ! It is excited to hear you have solved the problem ! My email address is [email protected] . And I am very appreciated for your kindness and great work!

Hi, I had send you via email just now...TQ

[WangXuan2401]

Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem?

I‘m very sorry that we have not solved the problem. Do u have any ideas?

Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address

wow ! It is excited to hear you have solved the problem ! My email address is [email protected] . And I am very appreciated for your kindness and great work!

Hi, I had send you via email just now...TQ

Thank u very much ! I have received your email.

[lingleong981130]

Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem?

I‘m very sorry that we have not solved the problem. Do u have any ideas?

Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address

wow ! It is excited to hear you have solved the problem ! My email address is [email protected] . And I am very appreciated for your kindness and great work!

Hi, I had send you via email just now...TQ

Thank u very much ! I have received your email. Hi, Do you found any errors in getting the results when you train your own model from the code?

[WangXuan2401]

@lingleong981130

hi~ I'm not sure exactly what errors you mentioned. When I trained with a sub-dataset, about 15000 photos, I only met the problem of "tensorflow" version, which is used in the ./util/detect_68lm.py and ./util/skin_mask.py

[Oliver-YX]

I am very appreciated for your great work! I also meet this promblem when trainning the full region BFM, could you please send me your file for a refrence? My email address is [email protected] .

[lingleong981130]

I am very appreciated for your great work! I also meet this promblem when trainning the full region BFM, could you please send me your file for a refrence? My email address is [email protected] .

Hi, I had send the folder to your email.

[Oliver-YX]

It's really helpful！Thanks for your kind reply～

---Original--- From: @.> Date: Wed, May 10, 2023 20:55 PM To: @.>; Cc: @.@.>; Subject: Re: [sicxu/Deep3DFaceRecon_pytorch] Wrong landmarks of full region(Issue #126)

I am very appreciated for your great work! I also meet this promblem when trainning the full region BFM, could you please send me your file for a refrence? My email address is @.*** .

Hi, I had send the folder to your email.

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

[hieuney73]

Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem?

I‘m very sorry that we have not solved the problem. Do u have any ideas?

Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address

Hello mister, can you send me the file, my email is: [email protected] . Thank you for you kindness!

[lingleong981130]

Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem?

I‘m very sorry that we have not solved the problem. Do u have any ideas?

Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address

Hello mister, can you send me the file, my email is: [email protected] . Thank you for you kindness!

Hi, I had sent you the folder just now. Anything you may email me.

[hieuney73]

Dear, Thank you for your quick respond. Im really appriciate it! I just recently get into the ml dl, computer vision field and i have been looking into this project for a week for my school project and it was kinda hard for a newbie like me. I want to ask that is it ok if i continue to get in touch with you through email for further problem as i go on in this project ( if that cost you no inconvenience of course ). Thank you for your consideration! Have a nice day, Hieu

Vào Th 5, 25 thg 5, 2023 lúc 18:18 lingleong981130 @.***> đã viết:

Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem?

I‘m very sorry that we have not solved the problem. Do u have any ideas?

Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address

Hello mister, can you send me the file, my email is: @.*** . Thank you for you kindness!

Hi, I had sent you the folder just now. Anything you may email me.

— Reply to this email directly, view it on GitHub https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1562727499, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYEL4HNSJ2RBJRWODRTROVDXH45XDANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.***>

[lingleong981130]

Dear, Thank you for your quick respond. Im really appriciate it! I just recently get into the ml dl, computer vision field and i have been looking into this project for a week for my school project and it was kinda hard for a newbie like me. I want to ask that is it ok if i continue to get in touch with you through email for further problem as i go on in this project ( if that cost you no inconvenience of course ). Thank you for your consideration! Have a nice day, Hieu Vào Th 5, 25 thg 5, 2023 lúc 18:18 lingleong981130 @.> đã viết: … Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem? I‘m very sorry that we have not solved the problem. Do u have any ideas? Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address Hello mister, can you send me the file, my email is: @. . Thank you for you kindness! Hi, I had sent you the folder just now. Anything you may email me. — Reply to this email directly, view it on GitHub <#126 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYEL4HNSJ2RBJRWODRTROVDXH45XDANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.***>

Yes, we can get in touch as sometimes I need to find someone which can discuss together. You are welcome.

[hieuney73]

Hello, I just got the problem when I try to do the data_preparation.py step. It only works when I use an image folder of around 100 images or less. I have also successfully trained about 100 images. But when I try with a bigger images folder of 1000 or so, the error returns as in the picture below. Can you show me how to get through it please, by the way I'm using the CelebA-HQ dataset. Thank you in advance! Ps: By using your code, i have been able to collect the right result on the test.py. Best regard, Hieu. [image: image.png] Vào Th 5, 25 thg 5, 2023 vào lúc 19:52 lingleong981130 < @.***> đã viết:

Dear, Thank you for your quick respond. Im really appriciate it! I just recently get into the ml dl, computer vision field and i have been looking into this project for a week for my school project and it was kinda hard for a newbie like me. I want to ask that is it ok if i continue to get in touch with you through email for further problem as i go on in this project ( if that cost you no inconvenience of course ). Thank you for your consideration! Have a nice day, Hieu Vào Th 5, 25 thg 5, 2023 lúc 18:18 lingleong981130 @.

> đã viết: … <#m_-5972733552536722377_> Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem? I‘m very sorry that we have not solved the problem. Do u have any ideas? Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address Hello mister, can you send me the file, my email is: @. . Thank you for you kindness! Hi, I had sent you the folder just now. Anything you may email me. — Reply to this email directly, view it on GitHub <#126 (comment) https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1562727499>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYEL4HNSJ2RBJRWODRTROVDXH45XDANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.***>

Yes, we can get in touch as sometimes I need to find someone which can discuss together. You are welcome.

— Reply to this email directly, view it on GitHub https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1562854526, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYEL4HNGPDSZC2ATJXXNPLDXH5IX7ANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.***>

[hieuney73]

I think the picture in the above email I have sent is bugged. Here is the picture of the returned error. Thank you!

Vào Th 3, 30 thg 5, 2023 vào lúc 11:32 huy hieu nguyen < @.***> đã viết:

Hello, I just got the problem when I try to do the data_preparation.py step. It only works when I use an image folder of around 100 images or less. I have also successfully trained about 100 images. But when I try with a bigger images folder of 1000 or so, the error returns as in the picture below. Can you show me how to get through it please, by the way I'm using the CelebA-HQ dataset. Thank you in advance! Ps: By using your code, i have been able to collect the right result on the test.py. Best regard, Hieu. [image: image.png] Vào Th 5, 25 thg 5, 2023 vào lúc 19:52 lingleong981130 < @.***> đã viết:

Dear, Thank you for your quick respond. Im really appriciate it! I just recently get into the ml dl, computer vision field and i have been looking into this project for a week for my school project and it was kinda hard for a newbie like me. I want to ask that is it ok if i continue to get in touch with you through email for further problem as i go on in this project ( if that cost you no inconvenience of course ). Thank you for your consideration! Have a nice day, Hieu Vào Th 5, 25 thg 5, 2023 lúc 18:18 lingleong981130 @.

> đã viết: … <#m_-7544762941631894051_m_-5972733552536722377_> Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem? I‘m very sorry that we have not solved the problem. Do u have any ideas? Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address Hello mister, can you send me the file, my email is: @. . Thank you for you kindness! Hi, I had sent you the folder just now. Anything you may email me. — Reply to this email directly, view it on GitHub <#126 (comment) https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1562727499>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYEL4HNSJ2RBJRWODRTROVDXH45XDANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.***>

Yes, we can get in touch as sometimes I need to find someone which can discuss together. You are welcome.

— Reply to this email directly, view it on GitHub https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1562854526, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYEL4HNGPDSZC2ATJXXNPLDXH5IX7ANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.***>

[lingleong981130]

I think the picture in the above email I have sent is bugged. Here is the picture of the returned error. Thank you! Vào Th 3, 30 thg 5, 2023 vào lúc 11:32 huy hieu nguyen < @.> đã viết: … Hello, I just got the problem when I try to do the data_preparation.py step. It only works when I use an image folder of around 100 images or less. I have also successfully trained about 100 images. But when I try with a bigger images folder of 1000 or so, the error returns as in the picture below. Can you show me how to get through it please, by the way I'm using the CelebA-HQ dataset. Thank you in advance! Ps: By using your code, i have been able to collect the right result on the test.py. Best regard, Hieu. [image: image.png] Vào Th 5, 25 thg 5, 2023 vào lúc 19:52 lingleong981130 < @.> đã viết: > Dear, Thank you for your quick respond. Im really appriciate it! I just > recently get into the ml dl, computer vision field and i have been looking > into this project for a week for my school project and it was kinda hard > for a newbie like me. I want to ask that is it ok if i continue to get in > touch with you through email for further problem as i go on in this project > ( if that cost you no inconvenience of course ). Thank you for your > consideration! Have a nice day, Hieu Vào Th 5, 25 thg 5, 2023 lúc 18:18 > lingleong981130 @. > > > đã viết: … <#m_-7544762941631894051_m_-5972733552536722377_> Hi, I get > the same output results for all the datasets from examples that provided by > author, may I know do you face the same problem? I‘m very sorry that we > have not solved the problem. Do u have any ideas? Hi, I found back my old > codes, it able to make a new train and the output is correct, if you need > it, I can send you the file via email, just give me your email address > Hello mister, can you send me the file, my email is: @. . Thank you for > you kindness! Hi, I had sent you the folder just now. Anything you may > email me. — Reply to this email directly, view it on GitHub <#126 > (comment) > <#126 (comment)>>, > or unsubscribe > https://github.com/notifications/unsubscribe-auth/AYEL4HNSJ2RBJRWODRTROVDXH45XDANCNFSM6AAAAAAUZFFYVE > . You are receiving this because you commented.Message ID: @.> > > Yes, we can get in touch as sometimes I need to find someone which can > discuss together. You are welcome. > > — > Reply to this email directly, view it on GitHub > <#126 (comment)>, > or unsubscribe > https://github.com/notifications/unsubscribe-auth/AYEL4HNGPDSZC2ATJXXNPLDXH5IX7ANCNFSM6AAAAAAUZFFYVE > . > You are receiving this because you commented.Message ID: > @.> >

sorry, I cant see the image here

[hieuney73]

I'm sorry for the confusion. I have included it. File "data_preparation.py", line 45, in data_prepare([os.path.join(opt.data_root,folder) for folder in opt.img_folder],opt.mode) File "data_preparation.py", line 26, in data_prepare detect_68p(img_folder,lm_sess,input_op,output_op) # detect landmarks for images File "C:\Users\Administrator\Downloads\deep3d\deep3d\util\detect_lm68.py", line 81, in detect_68p input_img, scale, bbox = align_for_lm(img, five_points) # align for 68 landmark detection File "C:\Users\Administrator\Downloads\deep3d\deep3d\util\preprocess.py", line 134, in align_for_lm five_points = np.array(five_points).reshape([1, 10]) ValueError: cannot reshape array of size 20 into shape (1,10)

Vào Th 3, 30 thg 5, 2023 vào lúc 20:08 lingleong981130 < @.***> đã viết:

I think the picture in the above email I have sent is bugged. Here is the picture of the returned error. Thank you! Vào Th 3, 30 thg 5, 2023 vào lúc 11:32 huy hieu nguyen < @.

> đã viết: … <#m_-3093600782904286399_> Hello, I just got the problem when I try to do the data_preparation.py step. It only works when I use an image folder of around 100 images or less. I have also successfully trained about 100 images. But when I try with a bigger images folder of 1000 or so, the error returns as in the picture below. Can you show me how to get through it please, by the way I'm using the CelebA-HQ dataset. Thank you in advance! Ps: By using your code, i have been able to collect the right result on the test.py. Best regard, Hieu. [image: image.png] Vào Th 5, 25 thg 5, 2023 vào lúc 19:52 lingleong981130 < @.> đã viết: > Dear, Thank you for your quick respond. Im really appriciate it! I just > recently get into the ml dl, computer vision field and i have been looking > into this project for a week for my school project and it was kinda hard > for a newbie like me. I want to ask that is it ok if i continue to get in > touch with you through email for further problem as i go on in this project > ( if that cost you no inconvenience of course ). Thank you for your > consideration! Have a nice day, Hieu Vào Th 5, 25 thg 5, 2023 lúc 18:18 > lingleong981130 @. > > > đã viết: … <#m_-7544762941631894051_m_-5972733552536722377_> Hi, I get > the same output results for all the datasets from examples that provided by > author, may I know do you face the same problem? I‘m very sorry that we > have not solved the problem. Do u have any ideas? Hi, I found back my old > codes, it able to make a new train and the output is correct, if you need > it, I can send you the file via email, just give me your email address > Hello mister, can you send me the file, my email is: @. . Thank you for

you kindness! Hi, I had sent you the folder just now. Anything you may > email me. — Reply to this email directly, view it on GitHub <#126 https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126 > (comment) <#126 (comment) https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1562727499>>, or unsubscribe > https://github.com/notifications/unsubscribe-auth/AYEL4HNSJ2RBJRWODRTROVDXH45XDANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.> > > Yes, we can get in touch as sometimes I need to find someone which can > discuss together. You are welcome. > > — > Reply to this email directly, view it on GitHub > <#126 (comment) https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1562854526>, or unsubscribe > https://github.com/notifications/unsubscribe-auth/AYEL4HNGPDSZC2ATJXXNPLDXH5IX7ANCNFSM6AAAAAAUZFFYVE https://github.com/notifications/unsubscribe-auth/AYEL4HNGPDSZC2ATJXXNPLDXH5IX7ANCNFSM6AAAAAAUZFFYVE . > You are receiving this because you commented.Message ID: > @.> >

sorry, I cant see the image here

— Reply to this email directly, view it on GitHub https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1568406201, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYEL4HJCRIBEBAULMDGOS5LXIXWL7ANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.***>

[hieuney73]

I have been able to fix the problem and successfully trained. But I'm a little confused about the loss function and the model evaluation metrics. Can you give me a brief insight about it? Thank you in advance!

Vào Th 3, 30 thg 5, 2023 vào lúc 20:27 huy hieu nguyen < @.***> đã viết:

I'm sorry for the confusion. I have included it. File "data_preparation.py", line 45, in data_prepare([os.path.join(opt.data_root,folder) for folder in opt.img_folder],opt.mode) File "data_preparation.py", line 26, in data_prepare detect_68p(img_folder,lm_sess,input_op,output_op) # detect landmarks for images File "C:\Users\Administrator\Downloads\deep3d\deep3d\util\detect_lm68.py", line 81, in detect_68p input_img, scale, bbox = align_for_lm(img, five_points) # align for 68 landmark detection File "C:\Users\Administrator\Downloads\deep3d\deep3d\util\preprocess.py", line 134, in align_for_lm five_points = np.array(five_points).reshape([1, 10]) ValueError: cannot reshape array of size 20 into shape (1,10)

Vào Th 3, 30 thg 5, 2023 vào lúc 20:08 lingleong981130 < @.***> đã viết:

I think the picture in the above email I have sent is bugged. Here is the picture of the returned error. Thank you! Vào Th 3, 30 thg 5, 2023 vào lúc 11:32 huy hieu nguyen < @.

> đã viết: … <#m_1123774765714044496_m_-3093600782904286399_> Hello, I just got the problem when I try to do the data_preparation.py step. It only works when I use an image folder of around 100 images or less. I have also successfully trained about 100 images. But when I try with a bigger images folder of 1000 or so, the error returns as in the picture below. Can you show me how to get through it please, by the way I'm using the CelebA-HQ dataset. Thank you in advance! Ps: By using your code, i have been able to collect the right result on the test.py. Best regard, Hieu. [image: image.png] Vào Th 5, 25 thg 5, 2023 vào lúc 19:52 lingleong981130 < @.> đã viết: > Dear, Thank you for your quick respond. Im really appriciate it! I just > recently get into the ml dl, computer vision field and i have been looking > into this project for a week for my school project and it was kinda hard > for a newbie like me. I want to ask that is it ok if i continue to get in > touch with you through email for further problem as i go on in this project > ( if that cost you no inconvenience of course ). Thank you for your > consideration! Have a nice day, Hieu Vào Th 5, 25 thg 5, 2023 lúc 18:18 > lingleong981130 @. > > > đã viết: … <#m_-7544762941631894051_m_-5972733552536722377_> Hi, I get > the same output results for all the datasets from examples that provided by > author, may I know do you face the same problem? I‘m very sorry that we > have not solved the problem. Do u have any ideas? Hi, I found back my old > codes, it able to make a new train and the output is correct, if you need > it, I can send you the file via email, just give me your email address > Hello mister, can you send me the file, my email is: @. . Thank you for

you kindness! Hi, I had sent you the folder just now. Anything you may > email me. — Reply to this email directly, view it on GitHub <#126 https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126 > (comment) > <#126 (comment) https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1562727499>>, or unsubscribe > https://github.com/notifications/unsubscribe-auth/AYEL4HNSJ2RBJRWODRTROVDXH45XDANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.> > > Yes, we can get in touch as sometimes I need to find someone which can > discuss together. You are welcome. > > — > Reply to this email directly, view it on GitHub > <#126 (comment) https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1562854526>, or unsubscribe > https://github.com/notifications/unsubscribe-auth/AYEL4HNGPDSZC2ATJXXNPLDXH5IX7ANCNFSM6AAAAAAUZFFYVE https://github.com/notifications/unsubscribe-auth/AYEL4HNGPDSZC2ATJXXNPLDXH5IX7ANCNFSM6AAAAAAUZFFYVE . > You are receiving this because you commented.Message ID: > @.> >

sorry, I cant see the image here

— Reply to this email directly, view it on GitHub https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1568406201, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYEL4HJCRIBEBAULMDGOS5LXIXWL7ANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.***>

[hieuney73]

Hello, i have just finished the training and testing process. But the only problem left is that i don't see any information or code about how to calculate the RMSE metrics, and it is crucial for me to finish the report. I want to ask that if you have any solution to this problem? Thank you!

Vào Th 3, 30 thg 5, 2023 vào lúc 20:08 lingleong981130 < @.***> đã viết:

I think the picture in the above email I have sent is bugged. Here is the picture of the returned error. Thank you! Vào Th 3, 30 thg 5, 2023 vào lúc 11:32 huy hieu nguyen < @.

> đã viết: … <#m_-2757606410261688998_> Hello, I just got the problem when I try to do the data_preparation.py step. It only works when I use an image folder of around 100 images or less. I have also successfully trained about 100 images. But when I try with a bigger images folder of 1000 or so, the error returns as in the picture below. Can you show me how to get through it please, by the way I'm using the CelebA-HQ dataset. Thank you in advance! Ps: By using your code, i have been able to collect the right result on the test.py. Best regard, Hieu. [image: image.png] Vào Th 5, 25 thg 5, 2023 vào lúc 19:52 lingleong981130 < @.> đã viết: > Dear, Thank you for your quick respond. Im really appriciate it! I just > recently get into the ml dl, computer vision field and i have been looking > into this project for a week for my school project and it was kinda hard > for a newbie like me. I want to ask that is it ok if i continue to get in > touch with you through email for further problem as i go on in this project > ( if that cost you no inconvenience of course ). Thank you for your > consideration! Have a nice day, Hieu Vào Th 5, 25 thg 5, 2023 lúc 18:18 > lingleong981130 @. > > > đã viết: … <#m_-7544762941631894051_m_-5972733552536722377_> Hi, I get > the same output results for all the datasets from examples that provided by > author, may I know do you face the same problem? I‘m very sorry that we > have not solved the problem. Do u have any ideas? Hi, I found back my old > codes, it able to make a new train and the output is correct, if you need > it, I can send you the file via email, just give me your email address > Hello mister, can you send me the file, my email is: @. . Thank you for

you kindness! Hi, I had sent you the folder just now. Anything you may > email me. — Reply to this email directly, view it on GitHub <#126 https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126 > (comment) <#126 (comment) https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1562727499>>, or unsubscribe > https://github.com/notifications/unsubscribe-auth/AYEL4HNSJ2RBJRWODRTROVDXH45XDANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.> > > Yes, we can get in touch as sometimes I need to find someone which can > discuss together. You are welcome. > > — > Reply to this email directly, view it on GitHub > <#126 (comment) https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1562854526>, or unsubscribe > https://github.com/notifications/unsubscribe-auth/AYEL4HNGPDSZC2ATJXXNPLDXH5IX7ANCNFSM6AAAAAAUZFFYVE https://github.com/notifications/unsubscribe-auth/AYEL4HNGPDSZC2ATJXXNPLDXH5IX7ANCNFSM6AAAAAAUZFFYVE . > You are receiving this because you commented.Message ID: > @.> >

sorry, I cant see the image here

— Reply to this email directly, view it on GitHub https://github.com/sicxu/Deep3DFaceRecon_pytorch/issues/126#issuecomment-1568406201, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYEL4HJCRIBEBAULMDGOS5LXIXWL7ANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.***>

[lingleong981130]

Hello, i have just finished the training and testing process. But the only problem left is that i don't see any information or code about how to calculate the RMSE metrics, and it is crucial for me to finish the report. I want to ask that if you have any solution to this problem? Thank you! Vào Th 3, 30 thg 5, 2023 vào lúc 20:08 lingleong981130 < @.> đã viết: … I think the picture in the above email I have sent is bugged. Here is the picture of the returned error. Thank you! Vào Th 3, 30 thg 5, 2023 vào lúc 11:32 huy hieu nguyen < @. > đã viết: … <#m_-2757606410261688998_> Hello, I just got the problem when I try to do the data_preparation.py step. It only works when I use an image folder of around 100 images or less. I have also successfully trained about 100 images. But when I try with a bigger images folder of 1000 or so, the error returns as in the picture below. Can you show me how to get through it please, by the way I'm using the CelebA-HQ dataset. Thank you in advance! Ps: By using your code, i have been able to collect the right result on the test.py. Best regard, Hieu. [image: image.png] Vào Th 5, 25 thg 5, 2023 vào lúc 19:52 lingleong981130 < @.> đã viết: > Dear, Thank you for your quick respond. Im really appriciate it! I just > recently get into the ml dl, computer vision field and i have been looking > into this project for a week for my school project and it was kinda hard > for a newbie like me. I want to ask that is it ok if i continue to get in > touch with you through email for further problem as i go on in this project > ( if that cost you no inconvenience of course ). Thank you for your > consideration! Have a nice day, Hieu Vào Th 5, 25 thg 5, 2023 lúc 18:18 > lingleong981130 @. > > > đã viết: … <#m_-7544762941631894051_m_-5972733552536722377_> Hi, I get > the same output results for all the datasets from examples that provided by > author, may I know do you face the same problem? I‘m very sorry that we > have not solved the problem. Do u have any ideas? Hi, I found back my old > codes, it able to make a new train and the output is correct, if you need > it, I can send you the file via email, just give me your email address > Hello mister, can you send me the file, my email is: @. . Thank you for > you kindness! Hi, I had sent you the folder just now. Anything you may > email me. — Reply to this email directly, view it on GitHub <#126 <#126> > (comment) > <#126 (comment) <#126 (comment)>>>, > or unsubscribe > https://github.com/notifications/unsubscribe-auth/AYEL4HNSJ2RBJRWODRTROVDXH45XDANCNFSM6AAAAAAUZFFYVE > . You are receiving this because you commented.Message ID: @.> > > Yes, we can get in touch as sometimes I need to find someone which can > discuss together. You are welcome. > > — > Reply to this email directly, view it on GitHub > <#126 (comment) <#126 (comment)>>, > or unsubscribe > https://github.com/notifications/unsubscribe-auth/AYEL4HNGPDSZC2ATJXXNPLDXH5IX7ANCNFSM6AAAAAAUZFFYVE https://github.com/notifications/unsubscribe-auth/AYEL4HNGPDSZC2ATJXXNPLDXH5IX7ANCNFSM6AAAAAAUZFFYVE > . > You are receiving this because you commented.Message ID: > @.> > sorry, I cant see the image here — Reply to this email directly, view it on GitHub <#126 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYEL4HJCRIBEBAULMDGOS5LXIXWL7ANCNFSM6AAAAAAUZFFYVE . You are receiving this because you commented.Message ID: @.**>

I had replied you via your email just now.

[wengjincheng]

Hi, I get the same output results for all the datasets from examples that provided by author, may I know do you face the same problem?

I‘m very sorry that we have not solved the problem. Do u have any ideas?

Hi, I found back my old codes, it able to make a new train and the output is correct, if you need it, I can send you the file via email, just give me your email address

Hi, I am also want this code to fix my problem, very appreciate that you can send me via my email: [email protected]

[julia-cherry]

Hello, can you send me the file, my email is: [email protected] . Thank you for you kindness!

[lingleong981130]

Hello, can you send me the file, my email is: [email protected] . Thank you for you kindness! Hi, which file you mentioned? the whole files???