ImportError After Install

[Skylion007]

I tried running some of the sample scripts and got the following import error:

/phoenix/S7/akg87/deformable_robotics_videos/train_video_transformer.py:1 in <module>            │
│                                                                                                  │
│ ❱  1 from video_transformers import TimeDistributed, VideoModel                                  │
│    2 from video_transformers.backbones.timm import TimmBackbone                                  │
│    3 from video_transformers.data import VideoDataModule                                         │
│    4 from video_transformers.heads import LinearHead                                             │
│                                                                                                  │
│ /home/akg87/miniconda3/envs/TensoRF/lib/python3.8/site-packages/video_transformers/__init__.py:1 │
│ in <module>                                                                                      │
│                                                                                                  │
│ ❱ 1 from video_transformers.auto.backbone import AutoBackbone                                    │
│   2 from video_transformers.auto.head import AutoHead                                            │
│   3 from video_transformers.auto.neck import AutoNeck                                            │
│   4 from video_transformers.modeling import TimeDistributed, VideoModel                          │
│                                                                                                  │
│ /home/akg87/miniconda3/envs/TensoRF/lib/python3.8/site-packages/video_transformers/auto/backbone │
│ .py:4 in <module>                                                                                │
│                                                                                                  │
│    1 from typing import Dict, Union                                                              │
│    2                                                                                             │
│    3 from video_transformers.backbones.base import Backbone                                      │
│ ❱  4 from video_transformers.modeling import TimeDistributed                                     │
│    5                                                                                             │
│    6                                                                                             │
│    7 class AutoBackbone:                                                                         │
│                                                                                                  │
│ /home/akg87/miniconda3/envs/TensoRF/lib/python3.8/site-packages/video_transformers/modeling.py:1 │
│ 3 in <module>                                                                                    │
│                                                                                                  │
│    10 from torch import nn                                                                       │
│    11                                                                                            │
│    12 import video_transformers.backbones.base                                                   │
│ ❱  13 import video_transformers.predict                                                          │
│    14 from video_transformers.heads import LinearHead                                            │
│    15 from video_transformers.hfhub_wrapper.hub_mixin import push_to_hub                         │
│    16 from video_transformers.utils.torch import get_num_total_params, get_num_trainable_param   │
│                                                                                                  │
│ /home/akg87/miniconda3/envs/TensoRF/lib/python3.8/site-packages/video_transformers/predict.py:9  │
│ in <module>                                                                                      │
│                                                                                                  │
│     6 import torch.utils.data                                                                    │
│     7 from iopath.common.file_io import g_pathmgr                                                │
│     8                                                                                            │
│ ❱   9 from video_transformers.data import VideoPreprocessor                                      │
│    10 from video_transformers.pytorchvideo_wrapper.data.labeled_video_paths import LabeledVide   │
│    11                                                                                            │
│    12                                                                                            │
│                                                                                                  │
│ /home/akg87/miniconda3/envs/TensoRF/lib/python3.8/site-packages/video_transformers/data.py:17 in │
│ <module>                                                                                         │
│                                                                                                  │
│    14 from torch.utils.data import DataLoader                                                    │
│    15 from torchvision.transforms import CenterCrop, Compose, Lambda, RandomCrop, RandomHorizo   │
│    16                                                                                            │
│ ❱  17 from video_transformers.pytorchvideo_wrapper.data.labeled_video_paths import LabeledVide   │
│    18 from video_transformers.utils.extra import class_to_config                                 │
│    19                                                                                            │
│    20 logger = get_logger(__name__)

[fcakyon]

Hello @Skylion007, which sample script have you run? Can you post the full error traceback? What are your Python and torch versions?

[fcakyon]

Please try the latest release video-transformers==0.0.7.