Open an99990 opened 2 years ago
@Yuliang-Liu do you have any idea on why i am not able to load the weight correctly and evaluate the model ?
@Yuliang-Liu do you have any idea on why i am not able to load the weight correctly and evaluate the model ?
hi, got a quic question, what's your cuda version and pytorch version. I had trouble with the compatible issues.
@an99990 i am facing same issue.. were you able to resolve it?
My issue was the labeling. Check and make sure they fit the coco format !
Get Outlook for Androidhttps://aka.ms/AAb9ysg
From: engrjav @.> Sent: Saturday, July 9, 2022 1:13:54 AM To: aim-uofa/AdelaiDet @.> Cc: an99990 @.>; Mention @.> Subject: Re: [aim-uofa/AdelaiDet] training blendmask on custom doesnt give any results (Issue #536)
@an99990https://www.google.com/url?q=https://github.com/an99990&source=gmail-imap&ust=1657948437000000&usg=AOvVaw0rcLl7E7k8NXICSIWcCKve i am facing same issue.. were you able to resolve it?
— Reply to this email directly, view it on GitHubhttps://www.google.com/url?q=https://github.com/aim-uofa/AdelaiDet/issues/536%23issuecomment-1179481637&source=gmail-imap&ust=1657948437000000&usg=AOvVaw34jKIYkqh_jj-Nz_M8GW-L, or unsubscribehttps://www.google.com/url?q=https://github.com/notifications/unsubscribe-auth/AUX7UXNOAGXOAS3YWRS2AQTVTEDBFANCNFSM5STCYBUA&source=gmail-imap&ust=1657948437000000&usg=AOvVaw1S0X91CGnDase4_zqx9Gtk. You are receiving this because you were mentioned.Message ID: @.***>
@an99990 but i have trained centernet with same annotations does this box inst or cond inst need special labelling?
I don't know about centernet but I know that once I got the label done by labelme. Boendmaks was able to detect stuff successfully. Try creating a small dataset and get labelme to generate labels and you'll see if that's the issue or no.
Get Outlook for Androidhttps://aka.ms/AAb9ysg
From: engrjav @.> Sent: Tuesday, July 12, 2022 4:06:25 AM To: aim-uofa/AdelaiDet @.> Cc: an99990 @.>; Mention @.> Subject: Re: [aim-uofa/AdelaiDet] training blendmask on custom doesnt give any results (Issue #536)
@an99990https://www.google.com/url?q=https://github.com/an99990&source=gmail-imap&ust=1658217987000000&usg=AOvVaw2f60J5wiYerwhgKkLdOhsf but i have trained centernet with same annotations does this box inst or cond inst need special labelling?
— Reply to this email directly, view it on GitHubhttps://www.google.com/url?q=https://github.com/aim-uofa/AdelaiDet/issues/536%23issuecomment-1181451805&source=gmail-imap&ust=1658217987000000&usg=AOvVaw0jHdpdemHRREQqgT1xQG_q, or unsubscribehttps://www.google.com/url?q=https://github.com/notifications/unsubscribe-auth/AUX7UXP4ZRWTHNULTITUSUTVTURQDANCNFSM5STCYBUA&source=gmail-imap&ust=1658217987000000&usg=AOvVaw0rK8kd54_GpWoyQphv1aQA. You are receiving this because you were mentioned.Message ID: @.***>
Thank you i will try that
Get Outlook for Androidhttps://aka.ms/AAb9ysg
From: an99990 @.> Sent: Tuesday, July 12, 2022 5:33:27 PM To: aim-uofa/AdelaiDet @.> Cc: engrjav @.>; Comment @.> Subject: Re: [aim-uofa/AdelaiDet] training blendmask on custom doesnt give any results (Issue #536)
I don't know about centernet but I know that once I got the label done by labelme. Boendmaks was able to detect stuff successfully. Try creating a small dataset and get labelme to generate labels and you'll see if that's the issue or no.
Get Outlook for Androidhttps://aka.ms/AAb9ysg
From: engrjav @.> Sent: Tuesday, July 12, 2022 4:06:25 AM To: aim-uofa/AdelaiDet @.> Cc: an99990 @.>; Mention @.> Subject: Re: [aim-uofa/AdelaiDet] training blendmask on custom doesnt give any results (Issue #536)
@an99990https://www.google.com/url?q=https://github.com/an99990&source=gmail-imap&ust=1658217987000000&usg=AOvVaw2f60J5wiYerwhgKkLdOhsf but i have trained centernet with same annotations does this box inst or cond inst need special labelling?
— Reply to this email directly, view it on GitHubhttps://www.google.com/url?q=https://github.com/aim-uofa/AdelaiDet/issues/536%23issuecomment-1181451805&source=gmail-imap&ust=1658217987000000&usg=AOvVaw0jHdpdemHRREQqgT1xQG_q, or unsubscribehttps://www.google.com/url?q=https://github.com/notifications/unsubscribe-auth/AUX7UXP4ZRWTHNULTITUSUTVTURQDANCNFSM5STCYBUA&source=gmail-imap&ust=1658217987000000&usg=AOvVaw0rK8kd54_GpWoyQphv1aQA. You are receiving this because you were mentioned.Message ID: @.***>
— Reply to this email directly, view it on GitHubhttps://github.com/aim-uofa/AdelaiDet/issues/536#issuecomment-1181704474, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AR6CY5L6QBNSG3FPWA6YGYTVTVQZPANCNFSM5STCYBUA. You are receiving this because you commented.Message ID: @.***>
So the training works but when i tested with an image i didnt get any segmented region or predictions. anyone has an idea on why ?
some log when i try to evaluate it :