我是机器学习和编程的新手。 现在，我正在尝试使用自己的数据来开发YOLACT AI。 但是，当我运行train.py时，出现以下错误，无法学习。 我该怎么做才能克服此错误？`
(yolact) tmori@tmori-Lenovo-Legion-Y740-15IRHg:~/yolact$ python train.py --config=can_config --save_interval=2000
loading annotations into memory...
Done (t=0.00s)
creating index...
index created!
loading annotations into memory...
Done (t=0.00s)
creating index...
index created!
Initializing weights...
Begin training!
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
[ 0] 0 || B: 4.840 | C: 16.249 | M: 4.682 | S: 2.749 | T: 28.521 || ETA: 9:18:44 || timer: 3.352
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
[ 1] 10 || B: 4.535 | C: 9.228 | M: 4.379 | S: 1.867 | T: 20.008 || ETA: 3:25:24 || timer: 0.864
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
/home/tmori/yolact/utils/augmentations.py:309: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
mode = random.choice(self.sample_options)
Computing validation mAP (this may take a while)...
Traceback (most recent call last):
File "train.py", line 504, in <module>
train()
File "train.py", line 371, in train
compute_validation_map(epoch, iteration, yolact_net, val_dataset, log if args.log else None)
File "train.py", line 492, in compute_validation_map
val_info = eval_script.evaluate(yolact_net, dataset, train_mode=True)
File "/home/tmori/yolact/eval.py", line 956, in evaluate
prep_metrics(ap_data, preds, img, gt, gt_masks, h, w, num_crowd, dataset.ids[image_idx], detections)
File "/home/tmori/yolact/eval.py", line 427, in prep_metrics
detections.add_bbox(image_id, classes[i], boxes[i,:], box_scores[i])
File "/home/tmori/yolact/eval.py", line 315, in add_bbox
'category_id': get_coco_cat(int(category_id)),
File "/home/tmori/yolact/eval.py", line 293, in get_coco_cat
return coco_cats[transformed_cat_id]
KeyError: 0
我正在尝试开发一种能找到罐子并将其细分的AI。 首先，我仅用labelme注释了一个“ can”项，然后使用labelme2coco.py创建了一个COCO格式的json文件。 之后，我根据YOLACT的GitHub上的“自定义数据集”修改了config.py并运行了train.py。
我的开发环境如下。 作业系统：Ubuntu20.04LTS 水蟒：4.8.3 的Python：3.6.12 火炬：1.4.0 CUDA工具包：10.1 cuDNN：7.6.5
答案 0 :(得分：0)
annotations.json中的类ID应该从1开始而不是0。如果它们是从0开始，请在config.py中的标签映射“ my_custom_dataset”中尝试添加此
'label_map': { 0: 1, 1: 2, 2: 3... and so on}
在这种情况下，共有3个班级！
在相同脚本num_classes中的yolact_base_config中，也应比您的类数大1，例如，在这种情况下，它应为4。