Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Cannot use my own pretrained model, Loss going NaN #836

Open
Santzu-27 opened this issue Apr 24, 2024 · 1 comment
Open

Cannot use my own pretrained model, Loss going NaN #836

Santzu-27 opened this issue Apr 24, 2024 · 1 comment

Comments

@Santzu-27
Copy link

When i try to use my own pretrained model it runs a error:
trainer.setTrainConfig(object_names_array=["insetos"], batch_size=6, num_experiments=270, train_from_pretrained_model="yolov3_insetos_mAP-0.54737_epoch-30.pt")

Generating anchor boxes for training images...
thr=0.25: 1.0000 best possible recall, 8.64 anchors past thr
n=9, img_size=416, metric_all=0.640/0.884-mean/best, past_thr=0.658-mean: 
pretrained weight loading failed. Defaulting to using random weight.
====================
Pretrained YOLOv3 model loaded to initialize weights
====================

But i "solved" the "pretrained weight loading failed. " error by putting the entire path way, but the results and losses keep giving to 0.0000 and Nan

`trainer.setTrainConfig(object_names_array=["insetos"], batch_size=6, num_experiments=270, train_from_pretrained_model="/home/oficinas40/2-BioIn/armadilhas/insetos/models/yolov3_insetos_mAP-0.54737_epoch-30.pt")

`


Generating anchor boxes for training images...
thr=0.25: 1.0000 best possible recall, 8.64 anchors past thr
n=9, img_size=416, metric_all=0.641/0.884-mean/best, past_thr=0.659-mean: 
====================
Pretrained YOLOv3 model loaded to initialize weights
====================
Epoch 1/90
----------
Train: 
29it [02:03,  4.25s/it]
    box loss-> nan, object loss-> nan, class loss-> 0.00000
Validation:
31it [03:55,  7.60s/it]
    recall: 0.000000 precision: 0.000000 [email protected]: 0.000000, [email protected]: 0.000000
@YXIDA
Copy link

YXIDA commented Jul 29, 2024

hello, finally the problem is solved no.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants