Clustered-Object-Detection-in-Aerial-Image
Clustered-Object-Detection-in-Aerial-Image copied to clipboard
Generating cluster region ground truth
I want to train model in server, so I can not run:
run "./detectron/ops/add_cluster_annotation.m" to generate cluster ground truth and add it to original object annotation files. run "./detectron/ops/visdrone2cocoformat.m" to convert VisDrone format annotation to COCO format.
When I run this I have permission denied.
I guess this is code implemented in Matlab. May by there is some python implementation?
you can run it locally and transfer the ground truth file to server.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 3:59 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Subscribed [email protected] Subject: [External] [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
I want to train model in server, so I can not run:
run "./detectron/ops/add_cluster_annotation.m" to generate cluster ground truth and add it to original object annotation files. run "./detectron/ops/visdrone2cocoformat.m" to convert VisDrone format annotation to COCO format.
When I run this I have permission denied.
I guess this is code implemented in Matlab. May by there is some python implementation?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTR7R54GN77T3FZWB2DS37IYVANCNFSM4WU4F3BQ.
Ok, thank you, I thought about the same solution. Could you send me some material about how to generate clustered ground truth?27.01.2021, 17:49, "fyangneil" [email protected]: you can run it locally and transfer the ground truth file to server.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 3:59 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Subscribed [email protected] Subject: [External] [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
I want to train model in server, so I can not run:
run "./detectron/ops/add_cluster_annotation.m" to generate cluster ground truth and add it to original object annotation files. run "./detectron/ops/visdrone2cocoformat.m" to convert VisDrone format annotation to COCO format.
When I run this I have permission denied.
I guess this is code implemented in Matlab. May by there is some python implementation?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTR7R54GN77T3FZWB2DS37IYVANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
run meanshift clustering algorithm on location (x,y) of ground truth bounding box.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 9:55 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Fan Yang [email protected]; Comment [email protected] Subject: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
Ok, thank you, I thought about the same solution. Could you send me some material about how to generate clustered ground truth?27.01.2021, 17:49, "fyangneil" [email protected]: you can run it locally and transfer the ground truth file to server.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 3:59 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Subscribed [email protected] Subject: [External] [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
I want to train model in server, so I can not run:
run "./detectron/ops/add_cluster_annotation.m" to generate cluster ground truth and add it to original object annotation files. run "./detectron/ops/visdrone2cocoformat.m" to convert VisDrone format annotation to COCO format.
When I run this I have permission denied.
I guess this is code implemented in Matlab. May by there is some python implementation?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTR7R54GN77T3FZWB2DS37IYVANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14#issuecomment-768340727, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTTSJSM73FVWI5KDXCTS4ASPTANCNFSM4WU4F3BQ.
then use boxes belong to the same cluster to yield a larger box as cluster box
From: Fan Yang [email protected] Sent: Wednesday, January 27, 2021 9:58 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Subject: Re: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
run meanshift clustering algorithm on location (x,y) of ground truth bounding box.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 9:55 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Fan Yang [email protected]; Comment [email protected] Subject: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
Ok, thank you, I thought about the same solution. Could you send me some material about how to generate clustered ground truth?27.01.2021, 17:49, "fyangneil" [email protected]: you can run it locally and transfer the ground truth file to server.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 3:59 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Subscribed [email protected] Subject: [External] [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
I want to train model in server, so I can not run:
run "./detectron/ops/add_cluster_annotation.m" to generate cluster ground truth and add it to original object annotation files. run "./detectron/ops/visdrone2cocoformat.m" to convert VisDrone format annotation to COCO format.
When I run this I have permission denied.
I guess this is code implemented in Matlab. May by there is some python implementation?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTR7R54GN77T3FZWB2DS37IYVANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14#issuecomment-768340727, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTTSJSM73FVWI5KDXCTS4ASPTANCNFSM4WU4F3BQ.
I have one more question. In detectron it is need to convert data to cocoformat. I see that in visdrone2cocoformat.m it is with val data, but in add_cluster_annotation.m data is saved in old format.27.01.2021, 17:59, "fyangneil" [email protected]: then use boxes belong to the same cluster to yield a larger box as cluster box
From: Fan Yang [email protected] Sent: Wednesday, January 27, 2021 9:58 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Subject: Re: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
run meanshift clustering algorithm on location (x,y) of ground truth bounding box.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 9:55 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Fan Yang [email protected]; Comment [email protected] Subject: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
Ok, thank you, I thought about the same solution. Could you send me some material about how to generate clustered ground truth?27.01.2021, 17:49, "fyangneil" [email protected]: you can run it locally and transfer the ground truth file to server.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 3:59 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Subscribed [email protected] Subject: [External] [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
I want to train model in server, so I can not run:
run "./detectron/ops/add_cluster_annotation.m" to generate cluster ground truth and add it to original object annotation files. run "./detectron/ops/visdrone2cocoformat.m" to convert VisDrone format annotation to COCO format.
When I run this I have permission denied.
I guess this is code implemented in Matlab. May by there is some python implementation?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTR7R54GN77T3FZWB2DS37IYVANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14#issuecomment-768340727, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTTSJSM73FVWI5KDXCTS4ASPTANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
And why is clustered bboxes written in one file with simple bboxes? I would like to know how I should set data with clustered bboxes that I could train.28.01.2021, 10:37, "Якуб Шарафутдинов" [email protected]:I have one more question. In detectron it is need to convert data to cocoformat. I see that in visdrone2cocoformat.m it is with val data, but in add_cluster_annotation.m data is saved in old format.27.01.2021, 17:59, "fyangneil" [email protected]: then use boxes belong to the same cluster to yield a larger box as cluster box
From: Fan Yang [email protected] Sent: Wednesday, January 27, 2021 9:58 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Subject: Re: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
run meanshift clustering algorithm on location (x,y) of ground truth bounding box.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 9:55 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Fan Yang [email protected]; Comment [email protected] Subject: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
Ok, thank you, I thought about the same solution. Could you send me some material about how to generate clustered ground truth?27.01.2021, 17:49, "fyangneil" [email protected]: you can run it locally and transfer the ground truth file to server.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 3:59 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Subscribed [email protected] Subject: [External] [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
I want to train model in server, so I can not run:
run "./detectron/ops/add_cluster_annotation.m" to generate cluster ground truth and add it to original object annotation files. run "./detectron/ops/visdrone2cocoformat.m" to convert VisDrone format annotation to COCO format.
When I run this I have permission denied.
I guess this is code implemented in Matlab. May by there is some python implementation?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTR7R54GN77T3FZWB2DS37IYVANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14#issuecomment-768340727, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTTSJSM73FVWI5KDXCTS4ASPTANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
Do you have pretrained weights?28.01.2021, 11:45, "Якуб Шарафутдинов" [email protected]:And why is clustered bboxes written in one file with simple bboxes? I would like to know how I should set data with clustered bboxes that I could train.28.01.2021, 10:37, "Якуб Шарафутдинов" [email protected]:I have one more question. In detectron it is need to convert data to cocoformat. I see that in visdrone2cocoformat.m it is with val data, but in add_cluster_annotation.m data is saved in old format.27.01.2021, 17:59, "fyangneil" [email protected]: then use boxes belong to the same cluster to yield a larger box as cluster box
From: Fan Yang [email protected] Sent: Wednesday, January 27, 2021 9:58 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Subject: Re: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
run meanshift clustering algorithm on location (x,y) of ground truth bounding box.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 9:55 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Fan Yang [email protected]; Comment [email protected] Subject: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
Ok, thank you, I thought about the same solution. Could you send me some material about how to generate clustered ground truth?27.01.2021, 17:49, "fyangneil" [email protected]: you can run it locally and transfer the ground truth file to server.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 3:59 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Subscribed [email protected] Subject: [External] [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
I want to train model in server, so I can not run:
run "./detectron/ops/add_cluster_annotation.m" to generate cluster ground truth and add it to original object annotation files. run "./detectron/ops/visdrone2cocoformat.m" to convert VisDrone format annotation to COCO format.
When I run this I have permission denied.
I guess this is code implemented in Matlab. May by there is some python implementation?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTR7R54GN77T3FZWB2DS37IYVANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14#issuecomment-768340727, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTTSJSM73FVWI5KDXCTS4ASPTANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
the cluster box is not in a simple file, it is stored as a class of object with other normal object boxes. Sorry, since it is long time ago, I do not have any weight file in hand.
From: sh7jacobi [email protected] Sent: Thursday, January 28, 2021 8:31 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Fan Yang [email protected]; Comment [email protected] Subject: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
Do you have pretrained weights?28.01.2021, 11:45, "Якуб Шарафутдинов" [email protected]:And why is clustered bboxes written in one file with simple bboxes? I would like to know how I should set data with clustered bboxes that I could train.28.01.2021, 10:37, "Якуб Шарафутдинов" [email protected]:I have one more question. In detectron it is need to convert data to cocoformat. I see that in visdrone2cocoformat.m it is with val data, but in add_cluster_annotation.m data is saved in old format.27.01.2021, 17:59, "fyangneil" [email protected]: then use boxes belong to the same cluster to yield a larger box as cluster box
From: Fan Yang [email protected] Sent: Wednesday, January 27, 2021 9:58 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Subject: Re: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
run meanshift clustering algorithm on location (x,y) of ground truth bounding box.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 9:55 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Fan Yang [email protected]; Comment [email protected] Subject: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
Ok, thank you, I thought about the same solution. Could you send me some material about how to generate clustered ground truth?27.01.2021, 17:49, "fyangneil" [email protected]: you can run it locally and transfer the ground truth file to server.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 3:59 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Subscribed [email protected] Subject: [External] [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
I want to train model in server, so I can not run:
run "./detectron/ops/add_cluster_annotation.m" to generate cluster ground truth and add it to original object annotation files. run "./detectron/ops/visdrone2cocoformat.m" to convert VisDrone format annotation to COCO format.
When I run this I have permission denied.
I guess this is code implemented in Matlab. May by there is some python implementation?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTR7R54GN77T3FZWB2DS37IYVANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14#issuecomment-768340727, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTTSJSM73FVWI5KDXCTS4ASPTANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14#issuecomment-769053811, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTRUKYLNATXWLYSZJPTS4FRK3ANCNFSM4WU4F3BQ.
I have fail when I run train, I think that is problem with version of cudnn. What version of cudnn you used?28.01.2021, 18:28, "fyangneil" [email protected]: the cluster box is not in a simple file, it is stored as a class of object with other normal object boxes. Sorry, since it is long time ago, I do not have any weight file in hand.
From: sh7jacobi [email protected] Sent: Thursday, January 28, 2021 8:31 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Fan Yang [email protected]; Comment [email protected] Subject: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
Do you have pretrained weights?28.01.2021, 11:45, "Якуб Шарафутдинов" [email protected]:And why is clustered bboxes written in one file with simple bboxes? I would like to know how I should set data with clustered bboxes that I could train.28.01.2021, 10:37, "Якуб Шарафутдинов" [email protected]:I have one more question. In detectron it is need to convert data to cocoformat. I see that in visdrone2cocoformat.m it is with val data, but in add_cluster_annotation.m data is saved in old format.27.01.2021, 17:59, "fyangneil" [email protected]: then use boxes belong to the same cluster to yield a larger box as cluster box
From: Fan Yang [email protected] Sent: Wednesday, January 27, 2021 9:58 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Subject: Re: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
run meanshift clustering algorithm on location (x,y) of ground truth bounding box.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 9:55 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Fan Yang [email protected]; Comment [email protected] Subject: [External] Re: [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
Ok, thank you, I thought about the same solution. Could you send me some material about how to generate clustered ground truth?27.01.2021, 17:49, "fyangneil" [email protected]: you can run it locally and transfer the ground truth file to server.
From: sh7jacobi [email protected] Sent: Wednesday, January 27, 2021 3:59 AM To: fyangneil/Clustered-Object-Detection-in-Aerial-Image [email protected] Cc: Subscribed [email protected] Subject: [External] [fyangneil/Clustered-Object-Detection-in-Aerial-Image] Generating cluster region ground truth (#14)
I want to train model in server, so I can not run:
run "./detectron/ops/add_cluster_annotation.m" to generate cluster ground truth and add it to original object annotation files. run "./detectron/ops/visdrone2cocoformat.m" to convert VisDrone format annotation to COCO format.
When I run this I have permission denied.
I guess this is code implemented in Matlab. May by there is some python implementation?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTR7R54GN77T3FZWB2DS37IYVANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14#issuecomment-768340727, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTTSJSM73FVWI5KDXCTS4ASPTANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image/issues/14#issuecomment-769053811, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALAUWTRUKYLNATXWLYSZJPTS4FRK3ANCNFSM4WU4F3BQ.
—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.
Hello, I have an error on train on first epoch: ...INFO loader.py: 235: [50/64]INFO loader.py: 235: [51/64]INFO loader.py: 235: [52/64]INFO loader.py: 235: [53/64]INFO loader.py: 235: [53/64]INFO loader.py: 235: [53/64]INFO loader.py: 235: [53/64]INFO loader.py: 235: [55/64]INFO loader.py: 235: [56/64]INFO loader.py: 235: [57/64]INFO loader.py: 235: [57/64]INFO loader.py: 235: [58/64]INFO loader.py: 235: [59/64]INFO loader.py: 235: [61/64]INFO loader.py: 235: [61/64]INFO loader.py: 235: [61/64]INFO loader.py: 235: [61/64]START IterTic !INFO detector.py: 753: Changing learning rate 0.000000 -> 0.000833 at iter 0START RunNet !I0202 13:32:29.233763 2948 context_gpu.cu:321] GPU 0: 811 MBI0202 13:32:29.233817 2948 context_gpu.cu:325] Total: 811 MBterminate called after throwing an instance of 'caffe2::EnforceNotMet' what(): [enforce fail at conv_op_cudnn.cc:572] status == CUDNN_STATUS_SUCCESS. 8 vs 0. , Error at: /var/lib/jenkins/workspace/caffe2/operators/conv_op_cudnn.cc:572: CUDNN_STATUS_EXECUTION_FAILED Error from operator:input: "gpu_0/data" input: "gpu_0/conv1_w" output: "gpu_0/conv1" name: "" type: "Conv" arg { name: "kernel" i: 7 } arg { name: "exhaustive_search" i: 0 } arg { name: "pad" i: 3 } arg { name: "order" s: "NCHW" } arg { name: "stride" i: 2 } device_option { device_type: 1 cuda_gpu_id: 0 } engine: "CUDNN"*** Aborted at 1612272750 (unix time) try "date -d @1612272750" if you are using GNU date PC: @ 0x7fdb9b88e428 gsignal SIGABRT (@0xb47) received by PID 2887 (TID 0x7fd9b17fa700) from PID 2887; stack trace: *** @ 0x7fdb9bc34390 (unknown) @ 0x7fdb9b88e428 gsignal @ 0x7fdb9b89002a abort @ 0x7fdb9593c84d __gnu_cxx::__verbose_terminate_handler() @ 0x7fdb9593a6b6 (unknown) @ 0x7fdb9593a701 std::terminate() @ 0x7fdb95965d38 (unknown) @ 0x7fdb9bc2a6ba start_thread @ 0x7fdb9b96041d clone @ 0x0 (unknown)Aborted (core dumped) How I can fix it?
It fails on codeline: worspace.RunNet(model.net.Proto().name)02.02.2021, 16:45, "Якуб Шарафутдинов" [email protected]:Hello, I have an error on train on first epoch: ...INFO loader.py: 235: [50/64]INFO loader.py: 235: [51/64]INFO loader.py: 235: [52/64]INFO loader.py: 235: [53/64]INFO loader.py: 235: [53/64]INFO loader.py: 235: [53/64]INFO loader.py: 235: [53/64]INFO loader.py: 235: [55/64]INFO loader.py: 235: [56/64]INFO loader.py: 235: [57/64]INFO loader.py: 235: [57/64]INFO loader.py: 235: [58/64]INFO loader.py: 235: [59/64]INFO loader.py: 235: [61/64]INFO loader.py: 235: [61/64]INFO loader.py: 235: [61/64]INFO loader.py: 235: [61/64]START IterTic !INFO detector.py: 753: Changing learning rate 0.000000 -> 0.000833 at iter 0START RunNet !I0202 13:32:29.233763 2948 context_gpu.cu:321] GPU 0: 811 MBI0202 13:32:29.233817 2948 context_gpu.cu:325] Total: 811 MBterminate called after throwing an instance of 'caffe2::EnforceNotMet' what(): [enforce fail at conv_op_cudnn.cc:572] status == CUDNN_STATUS_SUCCESS. 8 vs 0. , Error at: /var/lib/jenkins/workspace/caffe2/operators/conv_op_cudnn.cc:572: CUDNN_STATUS_EXECUTION_FAILED Error from operator:input: "gpu_0/data" input: "gpu_0/conv1_w" output: "gpu_0/conv1" name: "" type: "Conv" arg { name: "kernel" i: 7 } arg { name: "exhaustive_search" i: 0 } arg { name: "pad" i: 3 } arg { name: "order" s: "NCHW" } arg { name: "stride" i: 2 } device_option { device_type: 1 cuda_gpu_id: 0 } engine: "CUDNN"*** Aborted at 1612272750 (unix time) try "date -d @1612272750" if you are using GNU date PC: @ 0x7fdb9b88e428 gsignal SIGABRT (@0xb47) received by PID 2887 (TID 0x7fd9b17fa700) from PID 2887; stack trace: *** @ 0x7fdb9bc34390 (unknown) @ 0x7fdb9b88e428 gsignal @ 0x7fdb9b89002a abort @ 0x7fdb9593c84d __gnu_cxx::__verbose_terminate_handler() @ 0x7fdb9593a6b6 (unknown) @ 0x7fdb9593a701 std::terminate() @ 0x7fdb95965d38 (unknown) @ 0x7fdb9bc2a6ba start_thread @ 0x7fdb9b96041d clone @ 0x0 (unknown)Aborted (core dumped) How I can fix it?