Win10安装SlowFast 环境配置 及大Bug --“No module named torch._six “ 的修复

一、安装及配置

1、slowfast 源码下载
2、创建虚拟环境，一定要装python3.8，不然后面会报错。

conda create -n slowfast python=3.8

conda activate slowfast

3.安装pytorch等，pytorch版本一定不要用最新版，会报很多错，用老一点的更好。我一开始用了最新版，后面装着装着就一堆错，经过多次卸载安装再卸载再安装，终于调试出适合我的配方。没有别的办法，多尝试就行。

conda install pytorch==2.0.0 torchvision==0.15.0 torchaudio==2.0.1 pytorch-cuda=11.8 -c pytorch -c nvidia

4.安装别的包，其中要连github的需要连外网，此处教程较多，可以多搜索，多尝试。

pip install -U torch torchvision cython

pip install -U git+https://github.com/facebookresearch/fvcore.git

pip install git+https://github.com/philferriere/cocoapi.git#subdirectory=PythonAPI 

git clone https://github.com/facebookresearch/detectron2 detectron2_repo

pip install -e detectron2_repo

5.构建SlowFast，用pycharm 打开工程文件，在terminal 中输入以下代码，需关闭外网

python setup.py build develop 

6. 修改setup.py,主要是替换 PIL 和 sklearn,否则也会有报错

#!/usr/bin/env python3
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.

from setuptools import find_packages, setup

setup(
    name="slowfast",
    version="1.0",
    author="FAIR",
    url="unknown",
    description="SlowFast Video Understanding",
    install_requires=[
        "yacs>=0.1.6",
        "pyyaml>=5.1",
        "av",
        "matplotlib",
        "termcolor>=1.1",
        "simplejson",
        "tqdm",
        "psutil",
        "matplotlib",
        "detectron2",
        "opencv-python",
        "pandas",
        "torchvision>=0.4.2",
       # "PIL",
        "pillow",
       # "sklearn",
        "scikit-learn",
        "tensorboard",
        "fairscale",
    ],
    extras_require={"tensorboard_video_visualization": ["moviepy"]},
    packages=find_packages(exclude=("configs", "tests")),
)

7.继续安装包

pip install scikit-learn

pip install pillow

8. 在工程下新建两个文件夹 vinput 和voutput ，把测试视频放到vinput下
E:\abnormal action detect\slowfast\SlowFast-main\vinput
E:\abnormal action detect\slowfast\SlowFast-main\output

9.下载预训练权重
https://github.com/facebookresearch/SlowFast/blob/main/MODEL_ZOO.md

10.新建一个my.json文件，

 放到E:\abnormal action detect\slowfast\SlowFast-main\demo\my.json下，即你的工程目录下

{
    "bend/bow (at the waist)": 0,
    "crawl": 1,
    "crouch/kneel": 2,
    "dance": 3,
    "fall down": 4,
    "get up": 5,
    "jump/leap": 6,
    "lie/sleep": 7,
    "martial art": 8,
    "run/jog": 9,
    "sit": 10,
    "stand": 11,
    "swim": 12,
    "walk": 13,
    "answer phone": 14,
    "brush teeth": 15,
    "carry/hold (an object)": 16,
    "catch (an object)": 17,
    "chop": 18,
    "climb (e.g., a mountain)": 19,
    "clink glass": 20,
    "close (e.g., a door, a box)": 21,
    "cook": 22,
    "cut": 23,
    "dig": 24,
    "dress/put on clothing": 25,
    "drink": 26,
    "drive (e.g., a car, a truck)": 27,
    "eat": 28,
    "enter": 29,
    "exit": 30,
    "extract": 31,
    "fishing": 32,
    "hit (an object)": 33,
    "kick (an object)": 34,
    "lift/pick up": 35,
    "listen (e.g., to music)": 36,
    "open (e.g., a window, a car door)": 37,
    "paint": 38,
    "play board game": 39,
    "play musical instrument": 40,
    "play with pets": 41,
    "point to (an object)": 42,
    "press": 43,
    "pull (an object)": 44,
    "push (an object)": 45,
    "put down": 46,
    "read": 47,
    "ride (e.g., a bike, a car, a horse)": 48,
    "row boat": 49,
    "sail boat": 50,
    "shoot": 51,
    "shovel": 52,
    "smoke": 53,
    "stir": 54,
    "take a photo": 55,
    "text on/look at a cellphone": 56,
    "throw": 57,
    "touch (an object)": 58,
    "turn (e.g., a screwdriver)": 59,
    "watch (e.g., TV)": 60,
    "work on a computer": 61,
    "write": 62,
    "fight/hit (a person)": 63,
    "give/serve (an object) to (a person)": 64,
    "grab (a person)": 65,
    "hand clap": 66,
    "hand shake": 67,
    "hand wave": 68,
    "hug (a person)": 69,
    "kick (a person)": 70,
    "kiss (a person)": 71,
    "lift (a person)": 72,
    "listen to (a person)": 73,
    "play with kids": 74,
    "push (another person)": 75,
    "sing to (e.g., self, a person, a group)": 76,
    "take (an object) from (a person)": 77,
    "talk to (e.g., self, a person, a group)": 78,
    "watch (a person)": 79
}

11.修改yaml文件

E:\abnormal actiondetect\slowfast\SlowFastmain\demo

\AVA\SLOWFAST_32x2_R101_50_50.yaml

主要有以下几个地方：
（1）CHECKPOINT_FILE_PATH: 预训练权重路径
（2）LABEL_FILE_PATH:
（3）INPUT_VIDEO:
（4）OUTPUT_FILE：
（5）注释以下三行代码：

#TENSORBOARD:
#  MODEL_VIS:
#    TOPK: 2
（6）注释以下代码
 # WEBCAM: 0 

TRAIN:
  ENABLE: False
  DATASET: ava
  BATCH_SIZE: 16
  EVAL_PERIOD: 1
  CHECKPOINT_PERIOD: 1
  AUTO_RESUME: True
  #CHECKPOINT_FILE_PATH: ./SLOWFAST_32x2_R101_50_50.pkl  #path to pretrain model 预训练模型的路径
  CHECKPOINT_FILE_PATH: E:\abnormal action detect\slowfast\SlowFast-main\SLOWFAST_32x2_R101_50_50.pkl
  CHECKPOINT_TYPE: pytorch
DATA:
  NUM_FRAMES: 32
  SAMPLING_RATE: 2
  TRAIN_JITTER_SCALES: [256, 320]
  TRAIN_CROP_SIZE: 224
  TEST_CROP_SIZE: 256
  INPUT_CHANNEL_NUM: [3, 3]
DETECTION:
  ENABLE: True
  ALIGNED: False
AVA:
  BGR: False
  DETECTION_SCORE_THRESH: 0.8
  TEST_PREDICT_BOX_LISTS: ["person_box_67091280_iou90/ava_detection_val_boxes_and_labels.csv"]
SLOWFAST:
  ALPHA: 4
  BETA_INV: 8
  FUSION_CONV_CHANNEL_RATIO: 2
  FUSION_KERNEL_SZ: 5
RESNET:
  ZERO_INIT_FINAL_BN: True
  WIDTH_PER_GROUP: 64
  NUM_GROUPS: 1
  DEPTH: 101
  TRANS_FUNC: bottleneck_transform
  STRIDE_1X1: False
  NUM_BLOCK_TEMP_KERNEL: [[3, 3], [4, 4], [6, 6], [3, 3]]
  SPATIAL_DILATIONS: [[1, 1], [1, 1], [1, 1], [2, 2]]
  SPATIAL_STRIDES: [[1, 1], [2, 2], [2, 2], [1, 1]]
NONLOCAL:
  LOCATION: [[[], []], [[], []], [[6, 13, 20], []], [[], []]]
  GROUP: [[1, 1], [1, 1], [1, 1], [1, 1]]
  INSTANTIATION: dot_product
  POOL: [[[2, 2, 2], [2, 2, 2]], [[2, 2, 2], [2, 2, 2]], [[2, 2, 2], [2, 2, 2]], [[2, 2, 2], [2, 2, 2]]]
BN:
  USE_PRECISE_STATS: False
  NUM_BATCHES_PRECISE: 200
SOLVER:
  MOMENTUM: 0.9
  WEIGHT_DECAY: 1e-7
  OPTIMIZING_METHOD: sgd
MODEL:
  NUM_CLASSES: 80
  ARCH: slowfast
  MODEL_NAME: SlowFast
  LOSS_FUNC: bce
  DROPOUT_RATE: 0.5
  HEAD_ACT: sigmoid
TEST:
  ENABLE: False
  DATASET: ava
  BATCH_SIZE: 8
DATA_LOADER:
  NUM_WORKERS: 2
  PIN_MEMORY: True

NUM_GPUS: 1
NUM_SHARDS: 1
RNG_SEED: 0
OUTPUT_DIR: .
#TENSORBOARD:
#  MODEL_VIS:
#    TOPK: 2
DEMO:
  ENABLE: True


  LABEL_FILE_PATH: "E:\\abnormal action detect\\slowfast\\SlowFast-main\\demo\\my.json "  # Add local label file path here.
  INPUT_VIDEO: "E:\\abnormal action detect\\slowfast\\SlowFast-main\\vinput\\2.mp4"
  OUTPUT_FILE: "E:\\abnormal action detect\\slowfast\\SlowFast-main\\voutput\\2out.mp4"



  #WEBCAM: 0
  DETECTRON2_CFG: "COCO-Detection/faster_rcnn_R_50_FPN_3x.yaml"
  DETECTRON2_WEIGHTS: detectron2://COCO-Detection/faster_rcnn_R_50_FPN_3x/137849458/model_final_280758.pkl

12. 运行

python tools/run_net.py --cfg demo/AVA/SLOWFAST_32x2_R101_50_50.yaml

新手要注意，此处在pycharm 终端运行时，一定要将终端切换到虚拟环境下，具体切换方法见

PyCharm的终端（terminal）中进入指定conda虚拟环境_pycharm配置conda终端-CSDN博客

如果此时能跑通，那你将是幸运儿，不过大概率是跑不通的，会有很多莫名其妙的问题，好在大多数bug网上都能搜到解决办法，感谢辛苦写经验贴的技术达人们。

接下来我也分享几个我碰到的比较麻烦的bug:
二、Bug大修复

1、第一个大天坑 ：No module named 'torch._six'

 这个问题耗费了我大量时间，解决办法分为几步：

（1）一开始安装环境时，pytorch版本不能太高，并且要和torvision 匹配，匹配问题见前人分享的经验贴
PyTorch中torch、torchvision、torchaudio、torchtext版本对应关系_torch2.0.1对应的torchvision-CSDN博客

（2）但很快你会发现，就算版本匹配好后，运行还是出现No module named ‘torch._six ‘
解决办法，将build/lib/slowfast/datasets/multigrid_helper.py文件中的

from torch._six import int_classes as _int_classes修改为
from torch import int_classes as _int_classes （对我没用）
或者直接注释掉from torch._six import int_classes as _int_classes

（3）完成上一步后会悲催的发现，问题还是没有解决，原因是报错的不在本地工程文件下，而是虚拟环境下。在本地工程下进行第二步修改，并不能解决这个bug，比如我安装时就一直提示下面这个文件出错

D:\miniconda3\envs\slowfast\Lib\site-packages\
slowfast-1.0-py3.8.egg!\slowfast\datasets\multigrid_helper.py

这个文件不好找，并且用pycharm打开后，发现slowfast-1.0-py3.8.egg这个文件夹下的文件不能编辑，以为是权限问题，找了半天还是不能解决。

冥思苦想，又借助kimi, 想到了一个粗暴的解决办法。slowfast-1.0-py3.8.egg其实是一个压缩文件，用7-zip 打开后，将ultigrid_helper.py复制出来，然后用Pycharm打开，修改该文件下的from torch._six import int_classes as _int_classes语句，保存。此时一定要把pycharm关闭，然后将修改后ultigrid_helper.py文件拖进slowfast-1.0-py3.8.egg压缩包里，覆盖掉原文件，便可完成修改。
 

2、No module named ‘vision’
解决：这就是二个天坑，他的模块放置的位置不在他写的这个地址，需要手动去修改.

而且有几个地方

第一个地方，run_net.py下

第二个地方，
将vision.fair.slowfast.ava_evaluation替换为‘ava_evaluation’就可以了，这个地方也有在虚拟环境下的，不过此处可以在pycharm中直接修改

import slowfast.utils.distributed as du
from slowfast.utils.env import pathmgr
from vision.fair.slowfast.ava_evaluation import (
    object_detection_evaluation,
    standard_fields,
)

#修改为
import slowfast.utils.distributed as du
from slowfast.utils.env import pathmgr
from ava_evaluation import (
    object_detection_evaluation,
    standard_fields,
)

3. 一运行就出现AttributeError: 'NoneType' object has no attribute 'loader'
这个bug不影响运行，但是影响观感，参考pip install pyomo 损坏了 python3 virtualenv ·问题 #95 ·Pyomo/pyomo ·GitHub的

将虚拟环境下 D:\miniconda3\envs\slowfast\Lib\site-packages\vision-1.0.0-py3.7-nspkg.pth 

原本只有一行代码 在 import 代码后添加一个回车，变成两行后即解决，原因未知。具体见下图

import sys, types, os;
has_mfs = sys.version_info > (3, 5);p = os.path.join(sys._getframe(1).f_locals['sitedir'], *('vision',));importlib = has_mfs and __import__('importlib.util');has_mfs and __import__('importlib.machinery');m = has_mfs and sys.modules.setdefault('vision', importlib.util.module_from_spec(importlib.machinery.PathFinder.find_spec('vision', [os.path.dirname(p)])));m = m or sys.modules.setdefault('vision', types.ModuleType('vision'));mp = (m or []) and m.__dict__.setdefault('__path__',[]);(p not in mp) and mp.append(p)

3、总是提示以下信息

"The 'torchvision.transforms._functional_video' module is deprecated since 0.12 and will be removed in the future. "
"Please use the 'torchvision.transforms.functional' module instead."

解决方法
切换到D:\miniconda3\envs\slowfast\Lib\site-packages\torchvision\transforms\_transforms_video.py文件下的

将from . import _functional_video as F手动修改为
from . import functional as F

花了整整一天的时间终于搞定了第一个demo。其中遇见的bug不下20个，只能耐心一个个上网查寻解决办法，很多问题还是一知半解，不过能成功运行就是胜利，希望我的经验能给需要的人一点帮助，同时也方便自己后续查找。这也是我写这篇博客的初衷。