Hi, Im trying to train object detection on a custom dataset. I’m following [this guide](hailo_model_zoo/training/yolov8/README.rst at master · hailo-ai/hailo_model_zoo · GitHub to a tee, including running the training in the supplied docker container.)
My yolo detect train command:
yolo detect train data=/workspace/shared_with_docker/datasets/hbrv2/data.yaml model=yolov8s.pt name=retrain_yolov8s epochs=100 batch=16
Upon execution I recieve the following error:
Traceback (most recent call last):
File "/opt/conda/bin/yolo", line 33, in <module>
sys.exit(load_entry_point('ultralytics', 'console_scripts', 'yolo')())
File "/workspace/ultralytics/ultralytics/yolo/cfg/__init__.py", line 318, in entrypoint
getattr(model, mode)(**overrides) # default args from model
File "/workspace/ultralytics/ultralytics/yolo/engine/model.py", line 326, in train
self.trainer.train()
File "/workspace/ultralytics/ultralytics/yolo/engine/trainer.py", line 186, in train
self._do_train(RANK, world_size)
File "/workspace/ultralytics/ultralytics/yolo/engine/trainer.py", line 263, in _do_train
self._setup_train(rank, world_size)
File "/workspace/ultralytics/ultralytics/yolo/engine/trainer.py", line 246, in _setup_train
self.train_loader = self.get_dataloader(self.trainset, batch_size=batch_size, rank=rank, mode='train')
File "/workspace/ultralytics/ultralytics/yolo/v8/detect/train.py", line 43, in get_dataloader
build_dataloader(self.args, batch_size, img_path=dataset_path, stride=gs, rank=rank, mode=mode,
File "/workspace/ultralytics/ultralytics/yolo/data/build.py", line 71, in build_dataloader
dataset = YOLODataset(
File "/workspace/ultralytics/ultralytics/yolo/data/dataset.py", line 66, in __init__
super().__init__(img_path, imgsz, cache, augment, hyp, prefix, rect, batch_size, stride, pad, single_cls,
File "/workspace/ultralytics/ultralytics/yolo/data/base.py", line 48, in __init__
self.labels = self.get_labels()
File "/workspace/ultralytics/ultralytics/yolo/data/dataset.py", line 128, in get_labels
cache, exists = np.load(str(cache_path), allow_pickle=True).item(), True # load dict
File "/opt/conda/lib/python3.8/site-packages/numpy/lib/npyio.py", line 432, in load
return format.read_array(fid, allow_pickle=allow_pickle,
File "/opt/conda/lib/python3.8/site-packages/numpy/lib/format.py", line 792, in read_array
array = pickle.load(fp, **pickle_kwargs)
ModuleNotFoundError: No module named 'numpy._core'
I have tried re-building and re-deploying the container, aswell as manually reinstalling numpy inside the container. If anyone could point me in the right direction in order to troubleshoot this issue I would be grateful