Paddle [maybe a bug] when breaking from an iteration of a dataloader, and the python process terminates, then it may report a strange error.

b4lqfgs4  于 2022-12-31  发布在  Python
关注(0)|答案(5)|浏览(148)

The paddle experts can just ignore this issure. I just record the error here. It is not urgent. I will check it when I have time.

bug描述 Describe the Bug

for index, (images, labels) in enumerate(paddle_dataloader):
        if index > 5:
            break
        print(images.shape, labels.shape)

environment: Win11, pycharm, cpu, PaddlePaddle 2.3, python 3.7
I find the bug when I run the file C:\Users\Administrator\Desktop\contests\20220715_paddle_lwfx_7th\Going-deeper-with-Image-Transformers-using-PaddlePaddle\02_test_data.py

The error message is as follows:

Not using distributed mode
Namespace(aa='rand-m9-mstd0.5-inc1', batch_size=8, clip_grad=None, color_jitter=0.4, cooldown_epochs=10, cutmix=1.0, cutmix_minmax=None, data_path='C:\\Users\\Administrator\\Desktop\\contests\\20220715_paddle_lwfx_7th\\imagenet_dataset\\ILSVRC2012_img_val', data_set='IMNET', decay_epochs=30, decay_rate=0.1, device='cpu', dist_eval=False, dist_url='env://', distillation_alpha=0.5, distillation_tau=1.0, distillation_type='none', distributed=False, drop=0.0, drop_path=0.1, epochs=300, eval=False, finetune='', inat_category='name', input_size=224, lr=0.0005, lr_noise=None, lr_noise_pct=0.67, lr_noise_std=1.0, min_lr=1e-05, mixup=0.8, mixup_mode='batch', mixup_prob=1.0, mixup_switch_prob=0.5, model='cait_XXS24', model_ema=True, model_ema_decay=0.99996, model_ema_force_cpu=False, momentum=0.9, num_workers=0, opt='adamw', opt_betas=None, opt_eps=1e-08, output_dir='', patience_epochs=10, pin_mem=True, pretrained=False, recount=1, remode='pixel', repeated_aug=True, reprob=0.25, resplit=False, resume='', sched='cosine', seed=0, smoothing=0.1, start_epoch=0, teacher_model='regnety_160', teacher_path='', train_info_txt='C:\\Users\\Administrator\\Desktop\\contests\\20220715_paddle_lwfx_7th\\imagenet_dataset\\train_list_empty.txt', train_interpolation='bicubic', val_info_txt='C:\\Users\\Administrator\\Desktop\\contests\\20220715_paddle_lwfx_7th\\imagenet_dataset\\val_list.txt', warmup_epochs=5, warmup_lr=1e-06, weight_decay=0.05, world_size=1)
[12, 3, 224, 224] [12]
[12, 3, 224, 224] [12]
[12, 3, 224, 224] [12]
[12, 3, 224, 224] [12]
[12, 3, 224, 224] [12]
[12, 3, 224, 224] [12]
fail to perform transform [<paddle.vision.transforms.transforms.ToTensor object at 0x0000025CB918B9E8>] with error: We only support 'to_tensor()' in dynamic graph mode, please call 'paddle.disable_static()' to enter dynamic graph mode. and stack:
Traceback (most recent call last):
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\vision\transforms\transforms.py", line 113, in __call__
    data = f(data)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\vision\transforms\transforms.py", line 269, in __call__
    outputs.append(apply_func(inputs[i]))
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\vision\transforms\transforms.py", line 355, in _apply_image
    return F.to_tensor(img, self.data_format)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\vision\transforms\functional.py", line 82, in to_tensor
    return F_pil.to_tensor(pic, data_format)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\vision\transforms\functional_pil.py", line 88, in to_tensor
    img = paddle.to_tensor(np.array(pic, copy=False))
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\decorator.py", line 232, in fun
    return caller(func, *(extras + args), **kw)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\fluid\wrapped_decorator.py", line 25, in __impl__
    return wrapped_func(*args, **kwargs)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\fluid\framework.py", line 433, in __impl__
    ), "We only support '%s()' in dynamic graph mode, please call 'paddle.disable_static()' to enter dynamic graph mode." % func.__name__
AssertionError: We only support 'to_tensor()' in dynamic graph mode, please call 'paddle.disable_static()' to enter dynamic graph mode.

Exception in thread Thread-3:
Traceback (most recent call last):
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\threading.py", line 917, in _bootstrap_inner
    self.run()
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\threading.py", line 865, in run
    self._target(*self._args, **self._kwargs)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\fluid\dataloader\dataloader_iter.py", line 218, in _thread_loop
    self._thread_done_event)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\fluid\dataloader\fetcher.py", line 121, in fetch
    data.append(self.dataset[idx])
  File "C:\Users\Administrator\Desktop\contests\20220715_paddle_lwfx_7th\Going-deeper-with-Image-Transformers-using-PaddlePaddle\CaiT_paddle\datasets.py", line 126, in __getitem__
    image = self.transforms(image)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\vision\transforms\transforms.py", line 118, in __call__
    raise e
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\vision\transforms\transforms.py", line 113, in __call__
    data = f(data)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\vision\transforms\transforms.py", line 269, in __call__
    outputs.append(apply_func(inputs[i]))
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\vision\transforms\transforms.py", line 355, in _apply_image
    return F.to_tensor(img, self.data_format)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\vision\transforms\functional.py", line 82, in to_tensor
    return F_pil.to_tensor(pic, data_format)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\vision\transforms\functional_pil.py", line 88, in to_tensor
    img = paddle.to_tensor(np.array(pic, copy=False))
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\decorator.py", line 232, in fun
    return caller(func, *(extras + args), **kw)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\fluid\wrapped_decorator.py", line 25, in __impl__
    return wrapped_func(*args, **kwargs)
  File "C:\Users\Administrator\Anaconda3\envs\paddle-2.3-pytorch-1.8-python-3.7-env\lib\site-packages\paddle\fluid\framework.py", line 433, in __impl__
    ), "We only support '%s()' in dynamic graph mode, please call 'paddle.disable_static()' to enter dynamic graph mode." % func.__name__
AssertionError: We only support 'to_tensor()' in dynamic graph mode, please call 'paddle.disable_static()' to enter dynamic graph mode.

Process finished with exit code 0

Some situations that won't report the error:

1. using paddle dataloader, but the dataset is very small, so the dataloader is iterated without breaking from the "for" statement.

[12, 3, 224, 224] [12]
[12, 3, 224, 224] [12]
[6, 3, 224, 224] [6]

Process finished with exit code 0

2. use torch dataloader

3. when Process finished with exit code 0 doesn't occur, in my understanding, that is the python process doesn't terminate (I don't know how to express this situation)

In a word, I think when running a python file which includes the above code, if the program breaks from the iteration and then the python process terminates, then some strange error will occur. I'm not sure about it. I will check it in detail when I have time.

其他补充信息 Additional Supplementary Information

I'm not sure about the bug. I will check it in detail when I have time. You should pay attention to whether "Process finished with exit code 0" occurs. You need to understand what "Process finished with exit code 0" means.

dphi5xsq

dphi5xsq1#

您好,我们已经收到了您的问题,会安排技术人员尽快解答您的问题,请耐心等待。请您再次检查是否提供了清晰的问题描述、复现代码、环境&版本、报错信息等。同时,您也可以通过查看 官网API文档常见问题历史IssueAI社区 来寻求解答。祝您生活愉快~

Hi! We've received your issue and please be patient to get responded. We will arrange technicians to answer your questions as soon as possible. Please make sure that you have posted enough message to demo your request. You may also check out the APIFAQGithub Issue and AI community to get the answer.Have a nice day!

uz75evzq

uz75evzq2#

when running C:\Users\Administrator\Desktop\contests\contests_that_have_ended\20220509_paddle_lwfx\task3-Going-deeper-with-Image-Transformers\02_test_data.py , the result is as follows:

when running C:\Users\Administrator\Desktop\contests\20220715_paddle_lwfx_7th\Going-deeper-with-Image-Transformers-using-PaddlePaddle\02_test_data.py, the result is as follows:

mwecs4sa

mwecs4sa3#

Is the environment variable PYTHONUNBUFFERED the cause of the error ?? I'm not sure.

vbkedwbf

vbkedwbf5#

Try call paddle.disable_static() at the beginning of your code, it seems enter static graph mode in default.

相关问题