Win11安装detectron2和slowfast以及踩坑解决

最新推荐文章于 2024-04-22 17:47:19 发布

Yesung_Cloud1106

最新推荐文章于 2024-04-22 17:47:19 发布

阅读量881

点赞数 7

文章标签： python 深度学习开发语言

本文链接：https://blog.csdn.net/m0_60519450/article/details/136608182

版权

系统win11

软件anaconda3

CUDA version 12.0

第一步：新建虚拟环境python==3.8,直接从anaconda3里新建就可以

在prompt里激活环境

activate slowfast

第二步：安装pytorch

pip install torch==1.10.0+cu102 torchaudio==0.10.0+cu102 torchlibrosa==0.1.0 torchvision==0.11.1+cu102

第三步：下载并安装slowfast，fvcore,cocoapi,detectron2

下载

https://github.com/facebookresearch/SlowFast

https://github.com/facebookresearch/fvcore

https://github.com/philferriere/cocoapi

https://github.com/conansherry/detectron2

安装过程请参考自以下博文第三四五步，写得非常好：win10下detectron2编译安装方法及问题汇总_detectron2的安装nvcc 编译 cuda 源文件时出现了堆栈溢出的问题-CSDN博客

安装detectron2，在anaconda prompt里cd到detectron2的安装路径,编译安装

cd C:\softwaredownload\slowfast_need\detectron2-masterV2\detectron2-master

python setup.py build develop

安装过程中会提示缺什么包，就pip install什么包，直到编译成功。编译之后会在detectron2子文件夹生成一个.pyd文件

测试detectron2编译环境，新建一个test_windows_install.py文件，放在“C:\softwaredownload\slowfast_need\detectron2-masterV2\detectron2-master\tests”文件夹下，

test_windows_install.py代码内容请复制粘贴以下内容：

from detectron2.engine import DefaultPredictor
from detectron2.data import MetadataCatalog
from detectron2.config import get_cfg
from detectron2.utils.visualizer import ColorMode, Visualizer
from detectron2 import model_zoo

import cv2
import numpy as np
import requests

# Load an image
res = requests.get("https://live.staticflickr.com/700/33224654191_fdaee2e3f1_c_d.jpg")
image = np.asarray(bytearray(res.content), dtype="uint8")
image = cv2.imdecode(image, cv2.IMREAD_COLOR)

config_file = 'COCO-Detection/faster_rcnn_R_101_FPN_3x.yaml'
cfg = get_cfg()
cfg.merge_from_file(r'C:\softwaredownload\slowfast_need\detectron2-masterV2\detectron2-master\configs/'+config_file)#cfg.merge_from_file(model_zoo.get_config_file(config_file))
cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.75 # Threshold
cfg.MODEL.WEIGHTS ='https://dl.fbaipublicfiles.com/detectron2/COCO-Detection/faster_rcnn_R_101_FPN_3x/137851257/model_final_f6e8b1.pkl'
#cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url(config_file)
cfg.MODEL.DEVICE = "cpu" # cpu or cuda

# Create predictor
predictor = DefaultPredictor(cfg)

# Make prediction
output = predictor(image)
print(output)
v = Visualizer(image[:, :, ::-1],
               scale=0.8,
               metadata=MetadataCatalog.get(cfg.DATASETS.TRAIN[0]),
               instance_mode=ColorMode.IMAGE
               )
v = v.draw_instance_predictions(output["instances"].to("cpu"))
cv2.imshow('images', v.get_image()[:, :, ::-1])
cv2.waitKey(0)

如果运行过程中有以下报错，表示没有这个对象，需要从自己复制config文件路径

解决方案：

1，把cfg.merge_from_file(model_zoo.get_config_file(config_file))改成cfg.merge_from_file(r'C:\softwaredownload\slowfast_need\detectron2-masterV2\detectron2-master\configs/'+config_file)

2，把cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url(config_file)改成cfg.MODEL.WEIGHTS ='https://dl.fbaipublicfiles.com/detectron2/COCO-Detection/faster_rcnn_R_101_FPN_3x/137851257/model_final_f6e8b1.pkl'

第2步是直接重新从官网找的模型权重

官网链接选择的是R101-FPN，右键model复制链接即可粘贴到代码中

（上边的代码已经是解决了的。注释部分是原来的代码，原代码参考win10下detectron2编译安装方法及问题汇总_detectron2的安装nvcc 编译 cuda 源文件时出现了堆栈溢出的问题-CSDN博客）

运行结果如图：

第四步：编译slowfast

打开slowfast源码，找到setup.py文件，修改1个地方，把PIL改成pillow

安装slowfast

python setup.py build develop

conda install pywin32

在C:\softwaredownload\slowfast_need\SlowFast-main\demo\AVA下新建ava.json文件

{
    "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
}

参考自Windows 11 安装 SlowFast_slowfast_32x2_r101_50_50.pkl百度网盘-CSDN博客

下载模型权重文件modelzoo官网,选第3个，点link，下载到C:/softwaredownload/slowfast_need/SlowFast-main/demo/models/SLOWFAST_32x2_R101_50_50.pkl

第五步：修改C:\softwaredownload\slowfast_need\SlowFast-main\demo\AVA\SLOWFAST_32x2_R101_50_50.yaml

用记事本打开.yaml文件，修改这几行

添加模型路径

CHECKPOINT_FILE_PATH: "C:/softwaredownload/slowfast_need/SlowFast-main/demo/models/SLOWFAST_32x2_R101_50_50.pkl" #path to pretrain model

添加标签路径
LABEL_FILE_PATH: "C:/softwaredownload/slowfast_need/SlowFast-main/demo/AVA/ava.json" # Add local label file path here.

你输入的视频路径，写绝对路径
INPUT_VIDEO: "C:/softwaredownload/slowfast_need/SlowFast-main/vinput/1.mp4"

输出的是视频路径
OUTPUT_FILE: "C:/softwaredownload/slowfast_need/SlowFast-main/voutput/1_output.mp4"

注释掉以下四句

#TENSORBOARD:
# MODEL_VIS:
# TOPK: 2

#WEBCAM: -1

完整的修改后的文件如下：

TRAIN:
  ENABLE: False
  DATASET: ava
  BATCH_SIZE: 16
  EVAL_PERIOD: 1
  CHECKPOINT_PERIOD: 1
  AUTO_RESUME: True
  CHECKPOINT_FILE_PATH: "C:/softwaredownload/slowfast_need/SlowFast-main/demo/models/SLOWFAST_32x2_R101_50_50.pkl" #path to pretrain model
  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: "C:/softwaredownload/slowfast_need/SlowFast-main/demo/AVA/ava.json" # Add local label file path here.
  INPUT_VIDEO: "C:/softwaredownload/slowfast_need/SlowFast-main/vinput/1.mp4"
  OUTPUT_FILE: "C:/softwaredownload/slowfast_need/SlowFast-main/voutput/1_output.mp4"
  #WEBCAM: -1
  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

在slowfast根目录下运行

cd C:\softwaredownload\slowfast_need\SlowFast-main

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

_______________________________________

以下是踩坑记录：

1，首先,测试detectron2，出现报错

ImportError:DLL load failed while importing _C:找不到指定的程序路径

原因：解压detectron2压缩包时，丢了一个东西，应该就是这个DLL文件

解决方案：重新解压安装包，重新安装编译

如果抱的错是找不到_C，那么需要把_C文件复制到你的运行目录下。

2，运行slowfast，出现报错

OSError: Video C:/softwaredownload/slowfast_need/SlowFast-main/vinput/1.mp4 cannot be opened

原因：视频重命名的时候直接打“1.mp4”，打“1”就行了，“.mp4”是后缀

解决方案：重命名，记得把路径打对了

3，测试detectron2和slowfast，出现报错

AttributeError:module 'detectron2.model_zoo' has no attribute 'get_config_file'

原因：我安装的版本时0.1，也许是版本过低，需要在报错的.py文件下修改代码，先寻找configs的路径，把报错部分改成自己的路径+config_file

例如我在运行slowfast的时候就报了这个错误，

解决方案，找到predictor.py，修改报错的第174行。注释部分是原本的代码

（注：因为我在自己电脑的cpu运行，所以把cuda也改成cpu了，因为我的电脑cuda带不动，会出现：RuntimeError: Not compiled with GPU support）

self.cfg = get_cfg()
        self.cfg.merge_from_file(r'C:\softwaredownload\slowfast_need\detectron2-masterV2\detectron2-master\configs/'+cfg.DEMO.DETECTRON2_CFG)
        #self.cfg.merge_from_file(
        #    model_zoo.get_config_file(cfg.DEMO.DETECTRON2_CFG)
        #)
        self.cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = cfg.DEMO.DETECTRON2_THRESH
        #self.cfg.MODEL.WEIGHTS ='https://dl.fbaipublicfiles.com/detectron2/COCO-Detection/faster_rcnn_R_101_FPN_3x/137851257/model_final_f6e8b1.pkl'
        self.cfg.MODEL.WEIGHTS = cfg.DEMO.DETECTRON2_WEIGHTS
        self.cfg.INPUT.FORMAT = cfg.DEMO.INPUT_FORMAT
        if cfg.NUM_GPUS and gpu_id is None:
            gpu_id = torch.cuda.current_device()
        self.cfg.MODEL.DEVICE = "cpu"
        #self.cfg.MODEL.DEVICE = (
        #    "cuda:{}".format(gpu_id) if cfg.NUM_GPUS > 0 else "cpu"
        #)

4，测试slowfast出现报错：typeError:cannot unpack non-iterable NoneType object

原因：AVA.json文件里除了要复制的东西，不能写其他东西。把多余的删了，比如抬头默认的时间用户等。

# -*- coding: utf-8 -*-
"""
Created on Sat Mar  9 21:03:46 2024

@author: someone
"""

5，最后，运行slowfast报错

RuntimeError: CUDA out of memory. Tried to allocate 16.00 MiB (GPU 0; 4.00 GiB total capacity; 3.26 GiB already allocated; 0 bytes free; 3.44 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

逐一排查以上错误并解决后，顺畅无比，马上就要轻舟已过万重山了，谁知最后内存不足，泪了。

参考资料

【slowfast 代码解析输出检测数据】_slowfast导出数据-CSDN博客

Windows 11 安装 SlowFast_slowfast_32x2_r101_50_50.pkl百度网盘-CSDN博客

【SlowFast复现】SlowFast Networks for Video Recognition复现代码使用自己的视频进行demo检测_经典视频编码复现-CSDN博客

detectron2和slowfast的安装与使用_oserror: video 0 cannot be opened-CSDN博客

win10下detectron2编译安装方法及问题汇总_detectron2的安装nvcc 编译 cuda 源文件时出现了堆栈溢出的问题-CSDN博客

win10 下配置 slowfast 环境_slowfast算法复现windows-CSDN博客

Detectron2安装踩坑记录（比较详细版）-CSDN博客

win10安装detectron2_"no instance of overloaded function \"std::min\" m-CSDN博客