PyAV未安装错误，但已安装(GoogleColab)

Question

我试图使用以下方法导入ucf101数据集

dataset = torchvision.datasets.UCF101('/content/drive/My Drive/Colab Notebooks/ucf101', annotation_path='/content/drive/My Drive/Colab Notebooks/ucfTrainTestlist', frames_per_clip=16, step_between_clips=1, frame_rate=None, fold=1, train=True, transform=transforms.Compose([transforms.ToTensor(),]))

但它给了我这个错误

    ImportError                               Traceback (most recent call last)
<ipython-input-8-2c7f10121439> in <module>()
----> 1 dataset = torchvision.datasets.UCF101('/content/drive/My Drive/Colab Notebooks/ucf101', annotation_path='/content/drive/My Drive/Colab Notebooks/ucfTrainTestlist', frames_per_clip=16, step_between_clips=1, frame_rate=None, fold=1, train=True, transform=transforms.Compose([transforms.ToTensor(),]))

6 frames
/usr/local/lib/python3.6/dist-packages/torchvision/datasets/ucf101.py in __init__(self, root, annotation_path, frames_per_clip, step_between_clips, frame_rate, fold, train, transform, _precomputed_metadata, num_workers, _video_width, _video_height, _video_min_dimension, _audio_samples)
     70             _video_height=_video_height,
     71             _video_min_dimension=_video_min_dimension,
---> 72             _audio_samples=_audio_samples,
     73         )
     74         self.video_clips_metadata = video_clips.metadata

/usr/local/lib/python3.6/dist-packages/torchvision/datasets/video_utils.py in __init__(self, video_paths, clip_length_in_frames, frames_between_clips, frame_rate, _precomputed_metadata, num_workers, _video_width, _video_height, _video_min_dimension, _video_max_dimension, _audio_samples, _audio_channels)
    116 
    117         if _precomputed_metadata is None:
--> 118             self._compute_frame_pts()
    119         else:
    120             self._init_from_metadata(_precomputed_metadata)

/usr/local/lib/python3.6/dist-packages/torchvision/datasets/video_utils.py in _compute_frame_pts(self)
    140 
    141         with tqdm(total=len(dl)) as pbar:
--> 142             for batch in dl:
    143                 pbar.update(1)
    144                 clips, fps = list(zip(*batch))

/usr/local/lib/python3.6/dist-packages/torch/utils/data/dataloader.py in __next__(self)
    343 
    344     def __next__(self):
--> 345         data = self._next_data()
    346         self._num_yielded += 1
    347         if self._dataset_kind == _DatasetKind.Iterable and \

/usr/local/lib/python3.6/dist-packages/torch/utils/data/dataloader.py in _next_data(self)
    854             else:
    855                 del self._task_info[idx]
--> 856                 return self._process_data(data)
    857 
    858     def _try_put_index(self):

/usr/local/lib/python3.6/dist-packages/torch/utils/data/dataloader.py in _process_data(self, data)
    879         self._try_put_index()
    880         if isinstance(data, ExceptionWrapper):
--> 881             data.reraise()
    882         return data
    883 

/usr/local/lib/python3.6/dist-packages/torch/_utils.py in reraise(self)
    393             # (https://bugs.python.org/issue2651), so we work around it.
    394             msg = KeyErrorMessage(msg)
--> 395         raise self.exc_type(msg)

ImportError: Caught ImportError in DataLoader worker process 0.
Original Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/_utils/worker.py", line 178, in _worker_loop
    data = fetcher.fetch(index)
  File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/_utils/fetch.py", line 44, in <listcomp>
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/usr/local/lib/python3.6/dist-packages/torchvision/datasets/video_utils.py", line 61, in __getitem__
    return read_video_timestamps(self.x[idx])
  File "/usr/local/lib/python3.6/dist-packages/torchvision/io/video.py", line 318, in read_video_timestamps
    _check_av_available()
  File "/usr/local/lib/python3.6/dist-packages/torchvision/io/video.py", line 40, in _check_av_available
    raise av
ImportError: PyAV is not installed, and is necessary for the video operations in torchvision.
See https://github.com/mikeboers/PyAV#installation for instructions on how to
install PyAV on your system.

---------------------------------------------------------------------------
NOTE: If your import is failing due to a missing package, you can
manually install dependencies using either !pip or !apt.

To view examples of installing some common dependencies, click the
"Open Examples" button below.
---------------------------------------------------------------------------

问题是，我已经安装并导入了上面的一些单元格pyAV。

pip install av
import av
print (av.__version__)
>>8.0.1

Answer 1

在您安装av之后

!pip install av

您需要首先重新启动运行时。

MENU > Runtime > Restart runtime

然后再调用您的代码

import torchvision
dataset = torchvision.datasets.UCF101(...)

Answer 2

如果有人对kaggle上的这个问题感到困惑，重新设置会话将不会有帮助(至少在那里进行了测试)，因为它将删除安装PyAV的所有进度。

正确的解决方案是首先执行!pip install av，然后执行像import torchvision这样的任何导入。不需要会话重置。

我怀疑(但我太懒于检查)原因是PyAV的可用性是在torchvision验证的早期阶段进行的。