Detectron2 Checkpoint not found

Question

I've been getting an error like this since last night. I trained 5 models, there was no problem. Then there was such a problem. I would be glad if you could help.

AssertionError                            Traceback (most recent call last)
<ipython-input-9-08522bc16525> in <module>()
     34 os.makedirs(cfg.OUTPUT_DIR, exist_ok=True)
     35 trainer = CocoTrainer(cfg)
---> 36 trainer.resume_or_load(resume=False)
     37 trainer.train()

2 frames
/usr/local/lib/python3.6/dist-packages/fvcore/common/checkpoint.py in load(self, path, checkpointables)
    118         if not os.path.isfile(path):
    119             path = self.path_manager.get_local_path(path)
--> 120             assert os.path.isfile(path), "Checkpoint {} not found!".format(path)
    121 
    122         checkpoint = self._load_file(path)

AssertionError: Checkpoint https://dl.fbaipublicfiles.com/detectron2/COCO-Detection/faster_rcnn_X_101_32x8d_FPN_3x/139173657/model_final_68b088.pkl not found!

Answer 1

For now, I found a manual solution. I downloaded the model from the link and saved it to my drive profile. Then I copied the model from the drive profile to the output file and made changes to the code below.

cfg.MODEL.WEIGHTS = "/content/output/model_final_68b088.pkl"

Maybe there is a better solution but I don't know. I think there is a problem in accessing the model file in the link.

Answer 2

After downloading the model from model zoo,

Adding

MODEL.WEIGHTS path/to/model

in command line.

No need to change the code or config file.

Answer 3

Just mount the pickle file from google drive to your app.

cfg.MODEL.WEIGHTS = "/content/drive/MyDrive/file_model_weight/model_final_68b088.pkl"  # Let training initialize from model zoo

it can't access the model pickle file from model zoo, hence we have to give it locally or we can directly mount it from google drive.