search

SysCV / transfiner

Mask Transfiner for High-Quality Instance Segmentation, CVPR 2022
https://www.vis.xyz/pub/transfiner
Apache License 2.0
528 stars 60 forks source link

No predictions from the model! #17

Closed Fhujinwu closed 2 years ago

Fhujinwu commented 2 years ago

When I was testing COCO2017, the following error occurred：

[04/02 16:19:28 d2.evaluation.coco_evaluation]: Preparing results for COCO format ... [04/02 16:19:28 d2.evaluation.coco_evaluation]: Saving results to ./output/inference/coco_instances_results.json [04/02 16:19:28 d2.evaluation.coco_evaluation]: Evaluating predictions with unofficial COCO API... /home/Bxl/hujinwu/experience/transfiner-main/detectron2/evaluation/coco_evaluation.py:329: DeprecationWarning: The 'warn' method is deprecated, use 'warning' instead self._logger.warn("No predictions from the model!") WARNING [04/02 16:19:28 d2.evaluation.coco_evaluation]: No predictions from the model! [04/02 16:19:28 d2.engine.defaults]: Evaluation results for coco_2017_val in csv format: [04/02 16:19:28 d2.evaluation.testing]: copypaste: Task: bbox [04/02 16:19:28 d2.evaluation.testing]: copypaste: AP,AP50,AP75,APs,APm,APl [04/02 16:19:28 d2.evaluation.testing]: copypaste: nan,nan,nan,nan,nan,nan

Hello. Please tell me why there are "No predictions from the model!“ thanks.

lkeab commented 2 years ago

which command leads to this error? is the validation annotation file in the correct place?

Fhujinwu commented 2 years ago
  1. The command I use is : python tools/train_net.py --config-file configs/COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_1x_4gpu_transfiner.yaml --num-gpus 1 --eval-only MODEL.WEIGHTS xxxxx/transfiner-main/output/model_final.pth 2.Verification set location is correct.
  2. coco_instances_results.json is only 2 Bytes I look forward to your suggestions.
lkeab commented 2 years ago

what is your used model_final.pth? does it match the setting of the config file, such as both r50?

Fhujinwu commented 2 years ago

Hi, model_final.pth is the result of my trained model. I tried the training model you provided and I also get the error above. The environment I am using is python 3.6 + torch1.6.0 + cu92 + torchvision0.7.0 + cu92. I look forward to your suggestions. thanks. could this be the cause of the error?

lkeab commented 2 years ago

what about the config file matching question? is the loss decreasing normally in your training log?

Fhujinwu commented 2 years ago

Thank you for your help. When I changed the environment to "torch==1.7.1 torchvision==0.8.2 torchaudio==0.7.2 cudatoolkit=9.2", it solved the problem.