Open kunsungwoo opened 1 month ago
Is codebase the latest version?
Yes, it is the latest version.
could you tell me the command you entered CLI? and What model did you use? (yolov10n/s...)
I used yolov10n and used it in colab.
from ultralytics import YOLOv10
model = YOLOv10('/content/drive/MyDrive/yolo1/model/epoch7.pt') results= model.train(data='/content/sample_data/test/data.yaml',epochs=20,imgsz=640,batch=16,patience=2,save_period=1,name='yolo2',lr0=0.001,lrf=0.001,exist_ok=True,pretrained=True ,translate=0.01,scale=0.01,mosaic=0.01,mixup=0.00,box=20,close_mosaic=5,shear=0.0,erasing=0,perspective=0.0,flipud=0,fliplr=0,iou=0.95,copy_paste=0,optimizer='AdamW',cos_lr=True)
I met the same problem!!! It seems like the one2one and one2many selection
![Uploading 企业微信截图_17188466767849.png…]()
the same error
您好,您发的邮件我已收到,我尽快给您回复。
找到问题了,prdeication是个字典,dict_keys(['one2many', 'one2one'])
For example, after training for 20 epochs, best.pt works normally in Python. However, if you download and use, for example, epoch17.pt instead of best.pt, an error will occur. An error occurs in epoch14.pt and epoch17.pt, but best.pt operates normally.
File "C:\yolov10\ultralytics\engine\model.py", line 441, in predict return self.predictor.predict_cli(source=source) if is_cli else self.predictor(source=source, stream=stream) File "C:\yolov10\ultralytics\engine\predictor.py", line 168, in call return list(self.stream_inference(source, model, *args, **kwargs)) # merge list of Result into one File "C:\anaconda\envs\yolov10\lib\site-packages\torch\utils_contextlib.py", line 35, in generator_context response = gen.send(None) File "C:\yolov10\ultralytics\engine\predictor.py", line 255, in stream_inference self.results = self.postprocess(preds, im, im0s) File "C:\yolov10\ultralytics\models\yolo\detect\predict.py", line 25, in postprocess preds = ops.non_max_suppression( File "C:\yolov10\ultralytics\utils\ops.py", line 216, in non_max_suppression bs = prediction.shape[0] # batch size AttributeError: 'dict' object has no attribute 'shape'