运行train时报错

[ChrisZ-123]

default_dtype() and torch.set_default_device() as alternatives. (Triggered internally at ..\torch\csrc\tensor\python_tensor.cpp:453.)
_C._set_default_tensor_type(t)
Traceback (most recent call last):
File "c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\Train.py", line 111, in
train()
File "c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\Train.py", line 84, in train
for step,(img,target) in enumerate(data_loader):
File "C:\Users\Lenovo\AppData\Roaming\Python\Python310\site-packages\torch\utils\data\dataloader.py", line 630, in next
data = self._next_data()
File "C:\Users\Lenovo\AppData\Roaming\Python\Python310\site-packages\torch\utils\data\dataloader.py", line 674, in _next_data
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
File "C:\Users\Lenovo\AppData\Roaming\Python\Python310\site-packages\torch\utils\data_utils\fetch.py", line 51, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "C:\Users\Lenovo\AppData\Roaming\Python\Python310\site-packages\torch\utils\data_utils\fetch.py", line 51, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\voc0712.py", line 111, in getitem
im, gt, h, w = self.pull_item(index)
File "c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\voc0712.py", line 130, in pull_item
img, boxes, labels = self.transform(img, target[:, :4], target[:, 4])
File "c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\augmentations.py", line 416, in call
return self.augment(img, boxes, labels)
File "c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\augmentations.py", line 51, in call
img, boxes, labels = t(img, boxes, labels)
File "c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\augmentations.py", line 237, in call
mode = random.choice(self.sample_options)
File "numpy\random\mtrand.pyx", line 936, in numpy.random.mtrand.RandomState.choice
ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (6,) + inhomogeneous part.

[xiaozhang1017]

default_dtype() 和 torch.set_default_device() 作为替代方案。 （在 ..\torch\csrc\tensor\python_tensor.cpp:453 内部触发。） _C._set_default_tensor_type(t) 回溯（最近一次调用）： 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection -基于 CNN-master\Train.py”，第 111 行， train() 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\Train.py” ，第 84 行，在 枚举（data_loader）中的步骤（img，目标）中： 文件“C：\ Users \ Lenovo \ AppData \ Roaming \ Python \ Python310 \ site-packages \ torch \ utils \ data \ dataloader.py ”，第 630 行，下一个 data = self._next_data() 文件“C:\Users\Lenovo\AppData\Roaming\Python\Python310\site-packages\torch\utils\data\dataloader.py”，第 674 行，在_next_data data = self._dataset_fetcher.fetch(index) # 可能会引发 StopIteration File "C:\Users\Lenovo\AppData\Roaming\Python\Python310\site-packages\torch\utils\data_utils\fetch.py​​", line 51,在 fetch data = [self.dataset[idx] for idx in possible_batched_index] 文件“C:\Users\Lenovo\AppData\Roaming\Python\Python310\site-packages\torch\utils\data_utils\fetch.py​​”，第 51 行，在 data = [self.dataset[idx] for idx in possible_batched_index] 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\voc0712.py”，第 111 行，在getitem im, gt, h, w = self.pull_item(index) 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\voc0712.py”，第 130 行，在 pull_item 中 img, box, labels = self.transform(img, target[:, :4], target[:, 4]) 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN- master\augmentations.py”，第 416 行，调用中 返回 self.augment(img,boxes, labels) 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\augmentations. py”，第 51 行，调用img ,boxes, labels = t(img,boxes,labels) 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\augmentations.py ”，第 237 行，在调用 模式 = random.choice(self.sample_options) 文件“numpy\random\mtrand.pyx”，第 936 行，在 numpy.random.mtrand.RandomState.choice ValueError：用序列设置数组元素。请求的数组在 1 维之后具有不均匀的形状。检测到的形状为(6,)+不均匀部分。

@ChrisZ-123 我也有这个问题，你是否已经成功解决了？

[xiaozhang1017]

default_dtype() 和 torch.set_default_device() 作为替代方案。 （在 ..\torch\csrc\tensor\python_tensor.cpp:453 内部触发。） _C._set_default_tensor_type(t) 回溯（最近一次调用）： 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection -基于 CNN-master\Train.py”，第 111 行， train() 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\Train.py” ，第 84 行，在 枚举（data_loader）中的步骤（img，目标）中： 文件“C：\ Users \ Lenovo \ AppData \ Roaming \ Python \ Python310 \ site-packages \ torch \ utils \ data \ dataloader.py ”，第 630 行，下一个 data = self._next_data() 文件“C:\Users\Lenovo\AppData\Roaming\Python\Python310\site-packages\torch\utils\data\dataloader.py”，第 674 行，在_next_data data = self._dataset_fetcher.fetch(index) # 可能会引发 StopIteration File "C:\Users\Lenovo\AppData\Roaming\Python\Python310\site-packages\torch\utils\data_utils\fetch.py​​", line 51,在 fetch data = [self.dataset[idx] for idx in possible_batched_index] 文件“C:\Users\Lenovo\AppData\Roaming\Python\Python310\site-packages\torch\utils\data_utils\fetch.py​​”，第 51 行，在 data = [self.dataset[idx] for idx in possible_batched_index] 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\voc0712.py”，第 111 行，在getitem im, gt, h, w = self.pull_item(index) 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\voc0712.py”，第 130 行，在 pull_item 中 img, box, labels = self.transform(img, target[:, :4], target[:, 4]) 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN- master\augmentations.py”，第 416 行，调用中 返回 self.augment(img,boxes, labels) 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\augmentations. py”，第 51 行，调用img ,boxes, labels = t(img,boxes,labels) 文件“c:\Users\Lenovo\Desktop\Fatigue-Driven-Detection-Based-on-CNN-master\augmentations.py ”，第 237 行，在调用 模式 = random.choice(self.sample_options) 文件“numpy\random\mtrand.pyx”，第 936 行，在 numpy.random.mtrand.RandomState.choice ValueError：用序列设置数组元素。请求的数组在 1 维之后具有不均匀的形状。检测到的形状为(6,)+不均匀部分。

@ChrisZ-123 我也有这个问题，你是否已经成功解决了？

我发现报错是numpy版本问题。
解决办法：
卸载现有版本numpy，安装numpy 1.21.0(python 3.6)版本或者1.23.5(python 3.11)
pip install numpy==1.21.0或者pip install numpy==1.23.5

[Yunone1]

D:\Anaconda3\envs\tensorflow-gpu\python.exe D:\mybs\Fatigue-Driven-Detection-Based-on-CNN-master\Train.py
Traceback (most recent call last):
File "D:\mybs\Fatigue-Driven-Detection-Based-on-CNN-master\Train.py", line 111, in
train()
File "D:\mybs\Fatigue-Driven-Detection-Based-on-CNN-master\Train.py", line 61, in train
Config.MEANS))
File "D:\mybs\Fatigue-Driven-Detection-Based-on-CNN-master\voc0712.py", line 107, in init
for line in open(osp.join(rootpath, 'ImageSets', 'Main', name + '.txt')):
FileNotFoundError: [Errno 2] No such file or directory: './dataset/ImageSets\Main\trainval.txt'