KeyError: "No object named 'build_fcos_vovnet_fpn_backbone' found in 'BACKBONE' registry!" about centermask2

[ThelilinNB]

this is doing inference on one image using the detectron method. the code should be running in the appropriate directory

import detectron2 from detectron2.engine import DefaultPredictor from detectron2.config import get_cfg from vovnet_detectron2.vovnet.config import add_vovnet_config

Load a pre-trained VovNet-based model

cfg = get_cfg() print(1) add_vovnet_config(cfg) ----------->this is the part i was talking about print(2) cfg.merge_from_file("~~~~~~~~") # Specify your model's configuration file cfg.OUTPUT_DIR = "~~~~~~~" # Specify the path to your pre-trained checkpoint cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, "model_final.pth") cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.6

I have also solved this problem. Thank you very much for your help

[ThelilinNB]

Hello, have you solved this problem

[FruitPigFoot]

need help on this too

[ThelilinNB]

need help on this too

hi, do you have the same problem? Have you found a solution? Can we discuss it? Thank you

[FruitPigFoot]

i managed to solve it

[ThelilinNB]

i managed to solve it

can you tell me how you solved it? This problem has been bothering me for several days, but I still haven't found a solution. I am very eager to get your help, thank you bro

[FruitPigFoot]

in my instance i had to add code line : add_vovnet_config(cfg) and also had to import the add_vovnet_config. It has gotta be the same on centermask too. Or else can you upload the code where it's bothering you?

[ThelilinNB]

in my instance i had to add code line : add_vovnet_config(cfg) and also had to import the add_vovnet_config. It has gotta be the same on centermask too. Or else can you upload the code where it's bothering you?

Sorry, I just saw the information today. As for the add_vovnet_congig(cfg) you said, I can't understand where to add this part. Could you please upload the modified part of the code so that I can see it more intuitively

[FruitPigFoot]

this is doing inference on one image using the detectron method. the code should be running in the appropriate directory

import detectron2
from detectron2.engine import DefaultPredictor
from detectron2.config import get_cfg
from vovnet_detectron2.vovnet.config import add_vovnet_config

Load a pre-trained VovNet-based model

cfg = get_cfg()
print(1)
add_vovnet_config(cfg) ----------->this is the part i was talking about
print(2)
cfg.merge_from_file("~~~~~~~~") # Specify your model's configuration file
cfg.OUTPUT_DIR = "~~~~~~~" # Specify the path to your pre-trained checkpoint
cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, "model_final.pth")
cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.6