前言
做AI做的事兒锹杈，讓世界充滿AI
原定目標主要是對安卓手機進行UI實時監(jiān)測，思路是利用投屏原理拿到圖像幀迈着，在使用過程中進行無感知UI檢測竭望，我認為目前最大支持僅為60fps，作用能干嘛呢裕菠？作用可多了咬清，譬如訓練單圖標配合無障礙點擊進行智能化搶紅包，刷金幣奴潘，還有可能打王者旧烧，本篇為基礎搭建篇。
之所以放日志是因為工作環(huán)境是內網画髓，很多網站無法正常訪問掘剪，我是在家做一遍再在公司實現，主要用來對比執(zhí)行差異奈虾。
版本
Win10
Python 3.9.12
conda 22.9.0
Yolo5
全程高能注意
所有名稱不得出現中文名夺谁，否則報錯
目錄
- 安裝miniconda
- 配置conda國內下載鏡像
- 安裝Pytorch
- 創(chuàng)建conda虛擬環(huán)境
- 從git上下載Yolo5
- 圖片數據標記軟件
- 數據標記
- 數據籌備
- 訓練籌備
- 訓練可視化
- 檢測目標圖片
一、安裝miniconda
它是Anaconda的簡化版肉微，由于Anaconda每次安裝卸載太慢故用簡化版匾鸥，安裝卸載注意：
1.注冊表：\HKEY_CURRENT_USER\Software\Microsoft\Command Processor
下可能會多一個Autorun導致cmd打開就立即運行完，安裝完成后刪掉即可
2.安裝miniconda可能會由于權限等原因導致安裝的內容大量殘缺碉纳，典型的屬于缺失python.exe勿负，Library文件夾，如若發(fā)生建議先在C盤先安裝一遍再安裝到別的盤村象，之后為了節(jié)約C盤空間可以將其卸載
3.配置環(huán)境變量笆环，若在安裝過程中沒有勾選那個自動給配置環(huán)境變量的勾可以手動配置(事實上我勾了它也沒自動配置成功),主要配置如下：
D:\Miniconda3
D:\Miniconda3\Scripts
D:\Miniconda3\Library\bin
二、配置conda國內下載鏡像
由于原生下載鏈接有些包下載慢或下載不下來厚者，我們在這里使用清華鏡像（推薦使用豆瓣鏡像躁劣，清華鏡像有點小垃圾），按照官網教程做法生成.condarc文件巴拉巴拉后库菲，發(fā)現連conda命令都無法正常使用账忘，一直報某個python腳本的編碼錯誤，中間碰到一堆博客說啥https改http，去除-defaults都無法正常下載鳖擒，最終找到如下鏡像配置替換溉浙，conda命令方既能正常使用，下載還飛起蒋荚，至此鏡像配置完成：
channels:
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/win-64/
- defaults
show_channel_urls: true
channel_alias: https://mirrors.tuna.tsinghua.edu.cn/anaconda
default_channels:
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/r
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/pro
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/msys2
custom_channels:
conda-forge: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
msys2: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
bioconda: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
menpo: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
pytorch: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
pytorch-lts: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
simpleitk: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
三戳稽、安裝Pytorch
因工作電腦是集成顯卡，未防顯卡燒壞期升，我在Pytorch官網下載CPU版本惊奇，命令如下：
conda install pytorch torchvision torchaudio cpuonly -c pytorch
注：如若你的電腦需要cpu與gpu版本，建議在conda創(chuàng)建的虛擬環(huán)境中使用pip方式下載播赁，這樣下載的東西應該與虛擬環(huán)境綁定先舷，s可以執(zhí)行多個版本撵溃。
執(zhí)行日志：(簡書Markdown無折疊語法，真垃圾)
C:\Users\NewBeeChina>conda install pytorch torchvision torchaudio cpuonly -c pytorch
Collecting package metadata (current_repodata.json): done
Solving environment: |
Warning: >10 possible package resolutions (only showing differing packages):
- https://repo.anaconda.com/pkgs/main/noarch/noarch::colorama-0.4.4-pyhd3eb1b0_0, https://repo.anaconda.com/pkgs/main/win-64/win-64::console_shortcut-0.1.1-4, https://repo.anaconda.com/pkgs/main/win-64/win-64::powershell_shortcut-0.0.1-3, https://repo.anaconda.com/pkgs/main/win-64/win-64::win_inet_pton-1.1.0-py39haa95532_0
- defaults/noarch::colorama-0.4.4-pyhd3eb1b0_0, https://repo.anaconda.com/pkgs/main/win-64/win-64::console_shortcut-0.1.1-4, https://repo.anaconda.com/pkgs/main/win-64/win-64::powershell_shortcut-0.0.1-3, https://repo.anaconda.com/pkgs/main/win-64/win-64::win_inet_pton-1.1.0-py39haa95532_0
- defaults/win-64::console_shortcut-0.1.1-4, https://repo.anaconda.com/pkgs/main/noarch/noarch::colorama-0.4.4-pyhd3eb1b0_0, https://repo.anaconda.com/pkgs/main/win-64/win-64::powershell_shortcut-0.0.1-3, https://repo.anaconda.com/pkgs/main/win-64/win-64::win_inet_pton-1.1.0-py39haa95532_0
- defaults/noarch::colorama-0.4.4-pyhd3eb1b0_0, defaults/win-64::console_shortcut-0.1.1-4, https://repo.anaconda.com/pkgs/main/win-64/win-64::powershell_shortcut-0.0.1-3, https://repo.anaconda.com/pkgs/main/win-64/win-64::win_inet_pton-1.1.0-py39haa95532_0
- defaults/win-64::console_shortcut-0.1.1-4, defaults/win-64::powershell_shortcut-0.0.1-3, https://repo.anaconda.com/pkgs/main/noarch/noarch::colorama-0.4.4-pyhd3eb1b0_0, https://repo.anaconda.com/pkgs/main/win-64/win-64::win_inet_pton-1.1.0-py39haa95532_0
- defaults/noarch::colorama-0.4.4-pyhd3eb1b0_0, defaults/win-64::console_shortcut-0.1.1-4, defaults/win-64::powershell_shortcut-0.0.1-3, https://repo.anaconda.com/pkgs/main/win-64/win-64::win_inet_pton-1.1.0-py39haa95532_0
- defaults/win-64::powershell_shortcut-0.0.1-3, https://repo.anaconda.com/pkgs/main/noarch/noarch::colorama-0.4.4-pyhd3eb1b0_0, https://repo.anaconda.com/pkgs/main/win-64/win-64::console_shortcut-0.1.1-4, https://repo.anaconda.com/pkgs/main/win-64/win-64::win_inet_pton-1.1.0-py39haa95532_0
- defaults/noarch::colorama-0.4.4-pyhd3eb1b0_0, defaults/win-64::powershell_shortcut-0.0.1-3, https://repo.anaconda.com/pkgs/main/win-64/win-64::console_shortcut-0.1.1-4, https://repo.anaconda.com/pkgs/main/win-64/win-64::win_inet_pton-1.1.0-py39haa95532_0
- defaults/win-64::powershell_shortcut-0.0.1-3, defaults/win-64::win_inet_pton-1.1.0-py39haa95532_0, https://repo.anaconda.com/pkgs/main/noarch/noarch::colorama-0.4.4-pyhd3eb1b0_0, https://repo.anaconda.com/pkgs/main/win-64/win-64::console_shortcut-0.1.1-4
- defaults/win-64::win_inet_pton-1.1.0-py39haa95532_0, https://repo.anaconda.com/pkgs/main/noarch/noarch::colorama-0.4.4-pyhd3eb1b0_0, https://repo.anaconda.com/pkgs/main/win-64/win-64::console_shortcut-0.1.1-4, https://repo.anaconda.com/pkgs/main/win-64/win-64::powershell_shortcut-0.0.1-3
... and othedone
## Package Plan ##
environment location: D:\Miniconda3
added / updated specs:
- cpuonly
- pytorch
- torchaudio
- torchvision
The following packages will be downloaded:
package | build
---------------------------|-----------------
blas-1.0 | mkl 6 KB defaults
ca-certificates-2022.10.11 | haa95532_0 125 KB defaults
certifi-2022.9.24 | py39haa95532_0 154 KB defaults
conda-22.9.0 | py39haa95532_0 888 KB defaults
cpuonly-2.0 | 0 2 KB pytorch
freetype-2.12.1 | ha860e81_0 490 KB defaults
intel-openmp-2021.4.0 | haa95532_3556 2.2 MB defaults
jpeg-9e | h2bbff1b_0 292 KB defaults
lerc-3.0 | hd77b12b_0 120 KB defaults
libdeflate-1.8 | h2bbff1b_5 46 KB defaults
libpng-1.6.37 | h2a8f88b_0 333 KB defaults
libtiff-4.4.0 | h8a3f274_1 832 KB defaults
libuv-1.40.0 | he774522_0 255 KB defaults
libwebp-1.2.4 | h2bbff1b_0 67 KB defaults
libwebp-base-1.2.4 | h2bbff1b_0 279 KB defaults
lz4-c-1.9.3 | h2bbff1b_1 132 KB defaults
mkl-2021.4.0 | haa95532_640 114.9 MB defaults
mkl-service-2.4.0 | py39h2bbff1b_0 51 KB defaults
mkl_fft-1.3.1 | py39h277e83a_0 139 KB defaults
mkl_random-1.2.2 | py39hf11a4ad_0 225 KB defaults
numpy-1.23.3 | py39h3b20f71_0 11 KB defaults
numpy-base-1.23.3 | py39h4da318b_0 5.0 MB defaults
openssl-1.1.1s | h2bbff1b_0 5.5 MB defaults
pillow-9.2.0 | py39hdc2b20a_1 908 KB defaults
pytorch-1.13.0 | py3.9_cpu_0 138.2 MB pytorch
pytorch-mutex-1.0 | cpu 3 KB pytorch
tk-8.6.12 | h2bbff1b_0 3.1 MB defaults
toolz-0.12.0 | py39haa95532_0 106 KB defaults
torchaudio-0.13.0 | py39_cpu 4.5 MB pytorch
torchvision-0.14.0 | py39_cpu 6.3 MB pytorch
typing_extensions-4.3.0 | py39haa95532_0 42 KB defaults
xz-5.2.6 | h8cc25b3_0 240 KB defaults
zlib-1.2.13 | h8cc25b3_0 113 KB defaults
zstd-1.5.2 | h19a0ad4_0 509 KB defaults
------------------------------------------------------------
Total: 285.8 MB
The following NEW packages will be INSTALLED:
blas pkgs/main/win-64::blas-1.0-mkl
cpuonly pytorch/noarch::cpuonly-2.0-0
freetype pkgs/main/win-64::freetype-2.12.1-ha860e81_0
intel-openmp pkgs/main/win-64::intel-openmp-2021.4.0-haa95532_3556
jpeg pkgs/main/win-64::jpeg-9e-h2bbff1b_0
lerc pkgs/main/win-64::lerc-3.0-hd77b12b_0
libdeflate pkgs/main/win-64::libdeflate-1.8-h2bbff1b_5
libpng pkgs/main/win-64::libpng-1.6.37-h2a8f88b_0
libtiff pkgs/main/win-64::libtiff-4.4.0-h8a3f274_1
libuv pkgs/main/win-64::libuv-1.40.0-he774522_0
libwebp pkgs/main/win-64::libwebp-1.2.4-h2bbff1b_0
libwebp-base pkgs/main/win-64::libwebp-base-1.2.4-h2bbff1b_0
lz4-c pkgs/main/win-64::lz4-c-1.9.3-h2bbff1b_1
mkl pkgs/main/win-64::mkl-2021.4.0-haa95532_640
mkl-service pkgs/main/win-64::mkl-service-2.4.0-py39h2bbff1b_0
mkl_fft pkgs/main/win-64::mkl_fft-1.3.1-py39h277e83a_0
mkl_random pkgs/main/win-64::mkl_random-1.2.2-py39hf11a4ad_0
numpy pkgs/main/win-64::numpy-1.23.3-py39h3b20f71_0
numpy-base pkgs/main/win-64::numpy-base-1.23.3-py39h4da318b_0
pillow pkgs/main/win-64::pillow-9.2.0-py39hdc2b20a_1
pytorch pytorch/win-64::pytorch-1.13.0-py3.9_cpu_0
pytorch-mutex pytorch/noarch::pytorch-mutex-1.0-cpu
tk pkgs/main/win-64::tk-8.6.12-h2bbff1b_0
toolz pkgs/main/win-64::toolz-0.12.0-py39haa95532_0
torchaudio pytorch/win-64::torchaudio-0.13.0-py39_cpu
torchvision pytorch/win-64::torchvision-0.14.0-py39_cpu
typing_extensions pkgs/main/win-64::typing_extensions-4.3.0-py39haa95532_0
xz pkgs/main/win-64::xz-5.2.6-h8cc25b3_0
zlib pkgs/main/win-64::zlib-1.2.13-h8cc25b3_0
zstd pkgs/main/win-64::zstd-1.5.2-h19a0ad4_0
The following packages will be UPDATED:
ca-certificates 2022.3.29-haa95532_1 --> 2022.10.11-haa95532_0
certifi 2021.10.8-py39haa95532_2 --> 2022.9.24-py39haa95532_0
conda 4.12.0-py39haa95532_0 --> 22.9.0-py39haa95532_0
openssl 1.1.1n-h2bbff1b_0 --> 1.1.1s-h2bbff1b_0
Proceed ([y]/n)? y
Downloading and Extracting Packages
xz-5.2.6 | 240 KB | ############################################################################ | 100%
zlib-1.2.13 | 113 KB | ############################################################################ | 100%
zstd-1.5.2 | 509 KB | ############################################################################ | 100%
jpeg-9e | 292 KB | ############################################################################ | 100%
toolz-0.12.0 | 106 KB | ############################################################################ | 100%
torchaudio-0.13.0 | 4.5 MB | ############################################################################ | 100%
certifi-2022.9.24 | 154 KB | ############################################################################ | 100%
cpuonly-2.0 | 2 KB | ############################################################################ | 100%
torchvision-0.14.0 | 6.3 MB | ############################################################################ | 100%
conda-22.9.0 | 888 KB | ############################################################################ | 100%
pytorch-mutex-1.0 | 3 KB | ############################################################################ | 100%
libpng-1.6.37 | 333 KB | ############################################################################ | 100%
lz4-c-1.9.3 | 132 KB | ############################################################################ | 100%
libdeflate-1.8 | 46 KB | ############################################################################ | 100%
libuv-1.40.0 | 255 KB | ############################################################################ | 100%
lerc-3.0 | 120 KB | ############################################################################ | 100%
libtiff-4.4.0 | 832 KB | ############################################################################ | 100%
typing_extensions-4. | 42 KB | ############################################################################ | 100%
numpy-1.23.3 | 11 KB | ############################################################################ | 100%
ca-certificates-2022 | 125 KB | ############################################################################ | 100%
libwebp-base-1.2.4 | 279 KB | ############################################################################ | 100%
freetype-2.12.1 | 490 KB | ############################################################################ | 100%
mkl-2021.4.0 | 114.9 MB | ############################################################################ | 100%
mkl_fft-1.3.1 | 139 KB | ############################################################################ | 100%
numpy-base-1.23.3 | 5.0 MB | ############################################################################ | 100%
intel-openmp-2021.4. | 2.2 MB | ############################################################################ | 100%
mkl_random-1.2.2 | 225 KB | ############################################################################ | 100%
tk-8.6.12 | 3.1 MB | ############################################################################ | 100%
libwebp-1.2.4 | 67 KB | ############################################################################ | 100%
blas-1.0 | 6 KB | ############################################################################ | 100%
openssl-1.1.1s | 5.5 MB | ############################################################################ | 100%
pytorch-1.13.0 | 138.2 MB | ############################################################################ | 100%
mkl-service-2.4.0 | 51 KB | ############################################################################ | 100%
pillow-9.2.0 | 908 KB | ############################################################################ | 100%
Preparing transaction: done
Verifying transaction: done
Executing transaction: done
C:\Users\NewBeeChina>
三浮庐、創(chuàng)建conda虛擬環(huán)境
conda的主要作用就是其可以管理不同版本的python,最典型的場景就是我同時有py2與py3項目裂逐，通過conda創(chuàng)建不同的環(huán)境就可以同時執(zhí)行兩個版本的py工程.
創(chuàng)建格式：conda create --name 你起的環(huán)境名 python=你想創(chuàng)建的py版本
conda create --name LgpLoYo5 python=3.9.12
需要聯網下一點包输涕，然后敲命令conda activate 你起的環(huán)境名手動激活剛創(chuàng)建的虛擬環(huán)境展东，若忘了名字可以在磁盤D:\Miniconda3\envs目錄下看到你剛創(chuàng)建的虛擬環(huán)境,激活后就如下所示
注意：
1.激活與關閉命令也在如下所示的注釋
2.若第一次搭建環(huán)境在激活虛擬環(huán)境前需關閉重開cmd窗口否則可能報錯
done
#
# To activate this environment, use
#
# $ conda activate LgpLoYo5
#
# To deactivate an active environment, use
#
# $ conda deactivate
Retrieving notices: ...working... done
可能的報錯：
CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'.
If using 'conda activate' from a batch script, change your
invocation to 'CALL conda.bat activate'.
To initialize your shell, run
$ conda init <SHELL_NAME>
Currently supported shells are:
- bash
- cmd.exe
- fish
- tcsh
- xonsh
- zsh
- powershell
See 'conda init --help' for more information and options.
IMPORTANT: You may need to close and restart your shell after running 'conda init'.
幾個常用參考的命令：
查看版本號：conda -V
初始化：conda init
創(chuàng)建虛擬環(huán)境：conda create --name LgpYoLo5 python=3.9.12 -y
激活虛擬環(huán)境：conda activate LgpYoLo5
設置自動激活虛擬環(huán)境：conda config --set auto_activate_base true
查看所有虛擬環(huán)境：conda env list
退出虛擬環(huán)境：conda deactivate
四蜒滩、從git上下載Yolo5
在剛才激活的命令行環(huán)境中cd到從git下載的Yolo5工程的目錄,我一開始從別人博客復制命令pip install -r requirement.txt報了錯,后發(fā)現名稱錯了，真名為requirements.txt）
若未進行文件夾的命令行切換則報錯,所以在pip前一定要切換命令行的目錄到工程根目錄下
#未切目錄執(zhí)行命令報錯
ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt'
第二次從別人博客上復制的得滤，由于博客將requirements.txt寫成requirement.txt報了如下錯
(LgpLoYo5) E:\AIWorkSpace\yolov5-master>cd E:\AIWorkSpace\yolov5-master
(LgpLoYo5) E:\AIWorkSpace\yolov5-master>pip install -r requirement.txt
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirement.txt'
第三次我是從文件夾里找到了這個文件抬纸，手動敲的回車鍵,開始下載requirements.txt里列出的包，但是報了錯耿戚，
(LgpLoYo5) E:\AIWorkSpace\yolov5-master>pip install -r requirements.txt
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
Collecting ipython
Downloading ipython-8.6.0-py3-none-any.whl (761 kB)
---------------------------------------- 761.1/761.1 kB 18.4 kB/s eta 0:00:00
Collecting matplotlib>=3.2.2
Downloading matplotlib-3.6.2-cp39-cp39-win_amd64.whl (7.2 MB)
---------------------------------------- 0.1/7.2 MB 14.5 kB/s eta 0:08:14
ERROR: Exception:
Traceback (most recent call last):
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_vendor\urllib3\response.py", line 435, in _error_catcher
yield
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_vendor\urllib3\response.py", line 516, in read
data = self._fp.read(amt) if not fp_closed else b""
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_vendor\cachecontrol\filewrapper.py", line 90, in read
data = self.__fp.read(amt)
File "D:\Miniconda3\envs\LgpLoYo5\lib\http\client.py", line 463, in read
n = self.readinto(b)
File "D:\Miniconda3\envs\LgpLoYo5\lib\http\client.py", line 507, in readinto
n = self.fp.readinto(b)
File "D:\Miniconda3\envs\LgpLoYo5\lib\socket.py", line 704, in readinto
return self._sock.recv_into(b)
File "D:\Miniconda3\envs\LgpLoYo5\lib\ssl.py", line 1241, in recv_into
return self.read(nbytes, buffer)
File "D:\Miniconda3\envs\LgpLoYo5\lib\ssl.py", line 1099, in read
return self._sslobj.read(len, buffer)
socket.timeout: The read operation timed out
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\cli\base_command.py", line 167, in exc_logging_wrapper
status = run_func(*args)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\cli\req_command.py", line 247, in wrapper
return func(self, options, args)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\commands\install.py", line 369, in run
requirement_set = resolver.resolve(
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\resolution\resolvelib\resolver.py", line 92, in resolve
result = self._result = resolver.resolve(
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_vendor\resolvelib\resolvers.py", line 481, in resolve
state = resolution.resolve(requirements, max_rounds=max_rounds)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_vendor\resolvelib\resolvers.py", line 348, in resolve
self._add_to_criteria(self.state.criteria, r, parent=None)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_vendor\resolvelib\resolvers.py", line 172, in _add_to_criteria
if not criterion.candidates:
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_vendor\resolvelib\structs.py", line 151, in __bool__
return bool(self._sequence)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\resolution\resolvelib\found_candidates.py", line 155, in __bool__
return any(self)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\resolution\resolvelib\found_candidates.py", line 143, in <genexpr>
return (c for c in iterator if id(c) not in self._incompatible_ids)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\resolution\resolvelib\found_candidates.py", line 47, in _iter_built
candidate = func()
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\resolution\resolvelib\factory.py", line 206, in _make_candidate_from_link
self._link_candidate_cache[link] = LinkCandidate(
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\resolution\resolvelib\candidates.py", line 297, in __init__
super().__init__(
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\resolution\resolvelib\candidates.py", line 162, in __init__
self.dist = self._prepare()
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\resolution\resolvelib\candidates.py", line 231, in _prepare
dist = self._prepare_distribution()
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\resolution\resolvelib\candidates.py", line 308, in _prepare_distribution
return preparer.prepare_linked_requirement(self._ireq, parallel_builds=True)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\operations\prepare.py", line 438, in prepare_linked_requirement
return self._prepare_linked_requirement(req, parallel_builds)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\operations\prepare.py", line 483, in _prepare_linked_requirement
local_file = unpack_url(
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\operations\prepare.py", line 165, in unpack_url
file = get_http_url(
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\operations\prepare.py", line 106, in get_http_url
from_path, content_type = download(link, temp_dir.path)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\network\download.py", line 147, in __call__
for chunk in chunks:
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\cli\progress_bars.py", line 53, in _rich_progress_bar
for chunk in iterable:
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_internal\network\utils.py", line 63, in response_chunks
for chunk in response.raw.stream(
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_vendor\urllib3\response.py", line 573, in stream
data = self.read(amt=amt, decode_content=decode_content)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_vendor\urllib3\response.py", line 538, in read
raise IncompleteRead(self._fp_bytes_read, self.length_remaining)
File "D:\Miniconda3\envs\LgpLoYo5\lib\contextlib.py", line 137, in __exit__
self.gen.throw(typ, value, traceback)
File "D:\Miniconda3\envs\LgpLoYo5\lib\site-packages\pip\_vendor\urllib3\response.py", line 440, in _error_catcher
raise ReadTimeoutError(self._pool, None, "Read timed out.")
pip._vendor.urllib3.exceptions.ReadTimeoutError: HTTPSConnectionPool(host='files.pythonhosted.org', port=443): Read timed out.
打開requirements.txt文件查看,很明顯是下載第二個包matplotlib時發(fā)生了錯誤
# YOLOv5 ?? requirements
# Usage: pip install -r requirements.txt
# Base ------------------------------------------------------------------------
ipython # interactive notebook
matplotlib>=3.2.2
numpy>=1.18.5
opencv-python>=4.1.1
Pillow>=7.1.2
psutil # system resources
PyYAML>=5.3.1
requests>=2.23.0
scipy>=1.4.1
thop>=0.1.1 # FLOPs computation
torch>=1.7.0 # see https://pytorch.org/get-started/locally (recommended)
torchvision>=0.8.1
tqdm>=4.64.0
# protobuf<=3.20.1 # https://github.com/ultralytics/yolov5/issues/8012
# Logging ---------------------------------------------------------------------
tensorboard>=2.4.1
# clearml>=1.2.0
# comet
# Plotting --------------------------------------------------------------------
pandas>=1.1.4
seaborn>=0.11.0
# Export ----------------------------------------------------------------------
# coremltools>=6.0 # CoreML export
# onnx>=1.9.0 # ONNX export
# onnx-simplifier>=0.4.1 # ONNX simplifier
# nvidia-pyindex # TensorRT export
# nvidia-tensorrt # TensorRT export
# scikit-learn<=1.1.2 # CoreML quantization
# tensorflow>=2.4.1 # TF exports (-cpu, -aarch64, -macos)
# tensorflowjs>=3.9.0 # TF.js export
# openvino-dev # OpenVINO export
# Deploy ----------------------------------------------------------------------
# tritonclient[all]~=2.24.0
# Extras ----------------------------------------------------------------------
# mss # screenshots
# albumentations>=1.0.3
# pycocotools>=2.0 # COCO mAP
# roboflow
# ultralytics # HUB https://hub.ultralytics.com
不服輸，又把命令敲了一次,但是又發(fā)生了報錯
本次：
Collecting matplotlib>=3.2.2
Downloading matplotlib-3.6.2-cp39-cp39-win_amd64.whl (7.2 MB)
----------------------------- ---------- 5.4/7.2 MB 4.2 kB/s eta 0:07:20
上一次：
Collecting matplotlib>=3.2.2
Downloading matplotlib-3.6.2-cp39-cp39-win_amd64.whl (7.2 MB)
---------------------------------------- 0.1/7.2 MB 14.5 kB/s eta 0:08:14
巧了么不是阿趁，這TNND網絡不穩(wěn)定還是咋地膜蛔，后來查博客發(fā)現了如下命令，它是有效的(不要試圖將此鏈接添加至.condarc文件中使用脖阵，它就是個臨時工官網詳情)
pip install -i https://pypi.tuna.tsinghua.edu.cn/simple -r requirements.txt
安裝日志如下：
Microsoft Windows [版本 10.0.19044.2130]
(c) Microsoft Corporation皂股。保留所有權利。
C:\Users\NewBeeChina>e:
E:\>cd E:\AIWorkSpace\yolov5-master
E:\AIWorkSpace\yolov5-master>conda activate LgpLoYo5
(LgpLoYo5) E:\AIWorkSpace\yolov5-master>pip install -i https://pypi.tuna.tsinghua.edu.cn/simple -r requirements.txt
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
Collecting ipython
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/c7/53/072d677a16fd61f5806d80218c65202cc0ee77b831088af8f79ef59efcf2/ipython-8.6.0-py3-none-any.whl (761 kB)
---------------------------------------- 761.1/761.1 kB 1.2 MB/s eta 0:00:00
Collecting matplotlib>=3.2.2
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/78/af/4c83c99656c500ca0db7fe6f349d6309372ea8bad9c78d5c161930977bfd/matplotlib-3.6.2-cp39-cp39-win_amd64.whl (7.2 MB)
---------------------------------------- 7.2/7.2 MB 2.1 MB/s eta 0:00:00
Collecting numpy>=1.18.5
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/af/74/c02ece94ef88bed0a7f266959fd9bb2c97140345bc792f281b7db390eea9/numpy-1.23.4-cp39-cp39-win_amd64.whl (14.7 MB)
---------------------------------------- 14.7/14.7 MB 2.4 MB/s eta 0:00:00
Collecting opencv-python>=4.1.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/cf/09/b24c266cd61ddeed101b90c92a26f54d060b06f4a1b102eb891576d6e9e2/opencv_python-4.6.0.66-cp36-abi3-win_amd64.whl (35.6 MB)
---------------------------------------- 35.6/35.6 MB 3.0 MB/s eta 0:00:00
Collecting Pillow>=7.1.2
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/c0/8f/dfa473f3a6241bff91ae8bb905bd0afceb827f37de2917a94b5c4b1112bf/Pillow-9.3.0-cp39-cp39-win_amd64.whl (2.5 MB)
---------------------------------------- 2.5/2.5 MB 2.7 MB/s eta 0:00:00
Collecting psutil
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/25/6e/ba97809175c90cbdcd33b470e466ebf0854d15d1506e605cc0ddd284d5b6/psutil-5.9.4-cp36-abi3-win_amd64.whl (252 kB)
---------------------------------------- 252.5/252.5 kB 3.9 MB/s eta 0:00:00
Collecting PyYAML>=5.3.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/08/f4/ffa743f860f34a5e8c60abaaa686f82c9ac7a2b50e5a1c3b1eb564d59159/PyYAML-6.0-cp39-cp39-win_amd64.whl (151 kB)
---------------------------------------- 151.6/151.6 kB 4.4 MB/s eta 0:00:00
Collecting requests>=2.23.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/ca/91/6d9b8ccacd0412c08820f72cebaa4f0c0441b5cda699c90f618b6f8a1b42/requests-2.28.1-py3-none-any.whl (62 kB)
---------------------------------------- 62.8/62.8 kB ? eta 0:00:00
Collecting scipy>=1.4.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/d0/96/4f6eac3fea18f836a0e403539556b1684e6f3361fa39aa5d5797dedecd75/scipy-1.9.3-cp39-cp39-win_amd64.whl (40.2 MB)
---------------------------------------- 40.2/40.2 MB 2.9 MB/s eta 0:00:00
Collecting thop>=0.1.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/bb/0f/72beeab4ff5221dc47127c80f8834b4bcd0cb36f6ba91c0b1d04a1233403/thop-0.1.1.post2209072238-py3-none-any.whl (15 kB)
Collecting torch>=1.7.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/4c/d9/713853e06954bb657607d1e59d29e5896e1933e5d7fb50847a5730ad7325/torch-1.13.0-cp39-cp39-win_amd64.whl (167.2 MB)
---------------------------------------- 167.2/167.2 MB 2.3 MB/s eta 0:00:00
Collecting torchvision>=0.8.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/a7/f3/aaac29c2cdb84b0be1302aa17a68a7c39b05d9bca810d144e42c7131fb0d/torchvision-0.14.0-cp39-cp39-win_amd64.whl (1.1 MB)
---------------------------------------- 1.1/1.1 MB 2.4 MB/s eta 0:00:00
Collecting tqdm>=4.64.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/47/bb/849011636c4da2e44f1253cd927cfb20ada4374d8b3a4e425416e84900cc/tqdm-4.64.1-py2.py3-none-any.whl (78 kB)
---------------------------------------- 78.5/78.5 kB 1.5 MB/s eta 0:00:00
Collecting tensorboard>=2.4.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/05/70/ee7968f4a92ff9f95354d0ccaa9c0ba17b2644a33472ea845d92dd4e4821/tensorboard-2.11.0-py3-none-any.whl (6.0 MB)
---------------------------------------- 6.0/6.0 MB 1.6 MB/s eta 0:00:00
Collecting pandas>=1.1.4
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/60/53/619c0bcdc45b0a2ac94fc840c67073f8ca3f69344383c7dca0ed20e1ea73/pandas-1.5.1-cp39-cp39-win_amd64.whl (10.9 MB)
---------------------------------------- 10.9/10.9 MB 2.7 MB/s eta 0:00:00
Collecting seaborn>=0.11.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/77/18/7354cb68dd7906d5a3118e0ed3e30c37502f9e6253139ecfcf4fa33af210/seaborn-0.12.1-py3-none-any.whl (288 kB)
---------------------------------------- 288.2/288.2 kB 658.7 kB/s eta 0:00:00
Collecting traitlets>=5
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/ed/f9/caefd8c90955184e7426ef930e38c185e047169b520b35bdd57d341d03f4/traitlets-5.5.0-py3-none-any.whl (107 kB)
---------------------------------------- 107.4/107.4 kB 2.1 MB/s eta 0:00:00
Collecting decorator
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/d5/50/83c593b07763e1161326b3b8c6686f0f4b0f24d5526546bee538c89837d6/decorator-5.1.1-py3-none-any.whl (9.1 kB)
Collecting colorama
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/d1/d6/3965ed04c63042e047cb6a3e6ed1a63a35087b6a609aa3a15ed8ac56c221/colorama-0.4.6-py2.py3-none-any.whl (25 kB)
Collecting pygments>=2.4.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/4f/82/672cd382e5b39ab1cd422a672382f08a1fb3d08d9e0c0f3707f33a52063b/Pygments-2.13.0-py3-none-any.whl (1.1 MB)
---------------------------------------- 1.1/1.1 MB 725.7 kB/s eta 0:00:00
Collecting stack-data
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/0b/d3/87a41424a1d24d2cb9f5ae4ef4a97c7437ad81eb37d21049ce5decd13d70/stack_data-0.6.0-py3-none-any.whl (24 kB)
Collecting pickleshare
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/9a/41/220f49aaea88bc6fa6cba8d05ecf24676326156c23b991e80b3f2fc24c77/pickleshare-0.7.5-py2.py3-none-any.whl (6.9 kB)
Collecting backcall
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/4c/1c/ff6546b6c12603d8dd1070aa3c3d273ad4c07f5771689a7b69a550e8c951/backcall-0.2.0-py2.py3-none-any.whl (11 kB)
Collecting jedi>=0.16
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/b3/0e/836f12ec50075161e365131f13f5758451645af75c2becf61c6351ecec39/jedi-0.18.1-py2.py3-none-any.whl (1.6 MB)
---------------------------------------- 1.6/1.6 MB 748.9 kB/s eta 0:00:00
Collecting matplotlib-inline
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/f2/51/c34d7a1d528efaae3d8ddb18ef45a41f284eacf9e514523b191b7d0872cc/matplotlib_inline-0.1.6-py3-none-any.whl (9.4 kB)
Collecting prompt-toolkit<3.1.0,>3.0.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/03/22/784990e865d847384c28a05ff33ed09791251b320c212f957c62a11bd2ab/prompt_toolkit-3.0.32-py3-none-any.whl (382 kB)
---------------------------------------- 382.8/382.8 kB 335.9 kB/s eta 0:00:00
Collecting pyparsing>=2.2.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/6c/10/a7d0fa5baea8fe7b50f448ab742f26f52b80bfca85ac2be9d35cdd9a3246/pyparsing-3.0.9-py3-none-any.whl (98 kB)
---------------------------------------- 98.3/98.3 kB 704.8 kB/s eta 0:00:00
Collecting kiwisolver>=1.0.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/49/b9/edd9b69e1f2a8339347bcfcfbb14ce19db4a81158d01d8fd26fc3a088109/kiwisolver-1.4.4-cp39-cp39-win_amd64.whl (55 kB)
---------------------------------------- 55.4/55.4 kB ? eta 0:00:00
Collecting python-dateutil>=2.7
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/36/7a/87837f39d0296e723bb9b62bbb257d0355c7f6128853c78955f57342a56d/python_dateutil-2.8.2-py2.py3-none-any.whl (247 kB)
---------------------------------------- 247.7/247.7 kB 562.7 kB/s eta 0:00:00
Collecting packaging>=20.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/05/8e/8de486cbd03baba4deef4142bd643a3e7bbe954a784dc1bb17142572d127/packaging-21.3-py3-none-any.whl (40 kB)
---------------------------------------- 40.8/40.8 kB 1.9 MB/s eta 0:00:00
Collecting fonttools>=4.22.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/e3/d9/e9bae85e84737e76ebbcbea13607236da0c0699baed0ae4f1151b728a608/fonttools-4.38.0-py3-none-any.whl (965 kB)
---------------------------------------- 965.4/965.4 kB 955.1 kB/s eta 0:00:00
Collecting cycler>=0.10
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/5c/f9/695d6bedebd747e5eb0fe8fad57b72fdf25411273a39791cde838d5a8f51/cycler-0.11.0-py3-none-any.whl (6.4 kB)
Collecting contourpy>=1.0.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/a6/43/05ca3815a88650734860766e4d25a98ee7b8bf9d5f4fe280438c07ba5f4f/contourpy-1.0.6-cp39-cp39-win_amd64.whl (161 kB)
---------------------------------------- 161.3/161.3 kB 807.9 kB/s eta 0:00:00
Collecting idna<4,>=2.5
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/fc/34/3030de6f1370931b9dbb4dad48f6ab1015ab1d32447850b9fc94e60097be/idna-3.4-py3-none-any.whl (61 kB)
---------------------------------------- 61.5/61.5 kB 814.3 kB/s eta 0:00:00
Collecting charset-normalizer<3,>=2
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/db/51/a507c856293ab05cdc1db77ff4bc1268ddd39f29e7dc4919aa497f0adbec/charset_normalizer-2.1.1-py3-none-any.whl (39 kB)
Collecting urllib3<1.27,>=1.21.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/6f/de/5be2e3eed8426f871b170663333a0f627fc2924cc386cd41be065e7ea870/urllib3-1.26.12-py2.py3-none-any.whl (140 kB)
---------------------------------------- 140.4/140.4 kB 834.2 kB/s eta 0:00:00
Requirement already satisfied: certifi>=2017.4.17 in d:\miniconda3\envs\lgployo5\lib\site-packages (from requests>=2.23.0->-r requirements.txt (line 12)) (2022.9.24)
Collecting typing-extensions
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/0b/8e/f1a0a5a76cfef77e1eb6004cb49e5f8d72634da638420b9ea492ce8305e8/typing_extensions-4.4.0-py3-none-any.whl (26 kB)
Requirement already satisfied: wheel>=0.26 in d:\miniconda3\envs\lgployo5\lib\site-packages (from tensorboard>=2.4.1->-r requirements.txt (line 21)) (0.37.1)
Collecting tensorboard-data-server<0.7.0,>=0.6.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/74/69/5747a957f95e2e1d252ca41476ae40ce79d70d38151d2e494feb7722860c/tensorboard_data_server-0.6.1-py3-none-any.whl (2.4 kB)
Collecting werkzeug>=1.0.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/c8/27/be6ddbcf60115305205de79c29004a0c6bc53cec814f733467b1bb89386d/Werkzeug-2.2.2-py3-none-any.whl (232 kB)
---------------------------------------- 232.7/232.7 kB 158.1 kB/s eta 0:00:00
Requirement already satisfied: setuptools>=41.0.0 in d:\miniconda3\envs\lgployo5\lib\site-packages (from tensorboard>=2.4.1->-r requirements.txt (line 21)) (65.5.0)
Collecting grpcio>=1.24.3
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/e1/20/38ea842e338fb62384629d65d17f494c0f348bc3c16e81df607b31eb70ff/grpcio-1.50.0-cp39-cp39-win_amd64.whl (3.7 MB)
---------------------------------------- 3.7/3.7 MB 256.7 kB/s eta 0:00:00
Collecting google-auth<3,>=1.6.3
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/9b/9b/f40ea5c60762eabeb17cebdc05c395f44584c5c7fd7ce636a869c4f1e05d/google_auth-2.14.1-py2.py3-none-any.whl (175 kB)
---------------------------------------- 175.4/175.4 kB 330.2 kB/s eta 0:00:00
Collecting google-auth-oauthlib<0.5,>=0.4.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/b1/0e/0636cc1448a7abc444fb1b3a63655e294e0d2d49092dc3de05241be6d43c/google_auth_oauthlib-0.4.6-py2.py3-none-any.whl (18 kB)
Collecting tensorboard-plugin-wit>=1.6.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/e0/68/e8ecfac5dd594b676c23a7f07ea34c197d7d69b3313afdf8ac1b0a9905a2/tensorboard_plugin_wit-1.8.1-py3-none-any.whl (781 kB)
---------------------------------------- 781.3/781.3 kB 262.5 kB/s eta 0:00:00
Collecting protobuf<4,>=3.9.2
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/11/a5/e52b731415ad6ef3d841e9e6e337a690249e800cc7c06f0749afab26348c/protobuf-3.20.3-cp39-cp39-win_amd64.whl (904 kB)
---------------------------------------- 904.2/904.2 kB 457.7 kB/s eta 0:00:00
Collecting markdown>=2.6.8
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/86/be/ad281f7a3686b38dd8a307fa33210cdf2130404dfef668a37a4166d737ca/Markdown-3.4.1-py3-none-any.whl (93 kB)
---------------------------------------- 93.3/93.3 kB 532.4 kB/s eta 0:00:00
Collecting absl-py>=0.4
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/2c/39/ba081c6f7837366a39c9286fa1bc9dbf217249df80e133f25c62b05d1a53/absl_py-1.3.0-py3-none-any.whl (124 kB)
---------------------------------------- 124.6/124.6 kB 522.1 kB/s eta 0:00:00
Collecting pytz>=2020.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/85/ac/92f998fc52a70afd7f6b788142632afb27cd60c8c782d1452b7466603332/pytz-2022.6-py2.py3-none-any.whl (498 kB)
---------------------------------------- 498.1/498.1 kB 578.0 kB/s eta 0:00:00
Collecting six>=1.9.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/d9/5a/e7c31adbe875f2abbb91bd84cf2dc52d792b5a01506781dbcf25c91daf11/six-1.16.0-py2.py3-none-any.whl (11 kB)
Collecting rsa<5,>=3.1.4
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/49/97/fa78e3d2f65c02c8e1268b9aba606569fe97f6c8f7c2d74394553347c145/rsa-4.9-py3-none-any.whl (34 kB)
Collecting pyasn1-modules>=0.2.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/95/de/214830a981892a3e286c3794f41ae67a4495df1108c3da8a9f62159b9a9d/pyasn1_modules-0.2.8-py2.py3-none-any.whl (155 kB)
---------------------------------------- 155.3/155.3 kB 422.8 kB/s eta 0:00:00
Collecting cachetools<6.0,>=2.0.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/68/aa/5fc646cae6e997c3adf3b0a7e257cda75cff21fcba15354dffd67789b7bb/cachetools-5.2.0-py3-none-any.whl (9.3 kB)
Collecting requests-oauthlib>=0.7.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/6f/bb/5deac77a9af870143c684ab46a7934038a53eb4aa975bc0687ed6ca2c610/requests_oauthlib-1.3.1-py2.py3-none-any.whl (23 kB)
Collecting parso<0.9.0,>=0.8.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/05/63/8011bd08a4111858f79d2b09aad86638490d62fbf881c44e434a6dfca87b/parso-0.8.3-py2.py3-none-any.whl (100 kB)
---------------------------------------- 100.8/100.8 kB 131.6 kB/s eta 0:00:00
Collecting importlib-metadata>=4.4
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/b5/64/ef29a63cf08f047bb7fb22ab0f1f774b87eed0bb46d067a5a524798a4af8/importlib_metadata-5.0.0-py3-none-any.whl (21 kB)
Collecting wcwidth
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/59/7c/e39aca596badaf1b78e8f547c807b04dae603a433d3e7a7e04d67f2ef3e5/wcwidth-0.2.5-py2.py3-none-any.whl (30 kB)
Collecting MarkupSafe>=2.1.1
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/71/dc/41cbfe0d9aefdf14226dbf4eccfd0079a0e297809a17c5b902c9a7a3cc9a/MarkupSafe-2.1.1-cp39-cp39-win_amd64.whl (17 kB)
Collecting asttokens>=2.1.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/23/f7/19e20888d0b1e44b40c2795894a05e7be1f631c09949f7e0b177df1ab7a2/asttokens-2.1.0-py2.py3-none-any.whl (26 kB)
Collecting pure-eval
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/2b/27/77f9d5684e6bce929f5cfe18d6cfbe5133013c06cb2fbf5933670e60761d/pure_eval-0.2.2-py3-none-any.whl (11 kB)
Collecting executing>=1.2.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/28/3c/bc3819dd8b1a1588c9215a87271b6178cc5498acaa83885211f5d4d9e693/executing-1.2.0-py2.py3-none-any.whl (24 kB)
Collecting zipp>=0.5
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/40/8a/d63273ed0fa4a3d06f77e7b043f6577d8894e95515b0c187c52e2c0efabb/zipp-3.10.0-py3-none-any.whl (6.2 kB)
Collecting pyasn1<0.5.0,>=0.4.6
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/62/1e/a94a8d635fa3ce4cfc7f506003548d0a2447ae76fd5ca53932970fe3053f/pyasn1-0.4.8-py2.py3-none-any.whl (77 kB)
---------------------------------------- 77.1/77.1 kB 214.4 kB/s eta 0:00:00
Collecting oauthlib>=3.0.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/7e/80/cab10959dc1faead58dc8384a781dfbf93cb4d33d50988f7a69f1b7c9bbe/oauthlib-3.2.2-py3-none-any.whl (151 kB)
---------------------------------------- 151.7/151.7 kB 274.7 kB/s eta 0:00:00
Installing collected packages: wcwidth, tensorboard-plugin-wit, pytz, pyasn1, pure-eval, pickleshare, executing, backcall, zipp, urllib3, typing-extensions, traitlets, tensorboard-data-server, six, rsa, PyYAML, pyparsing, pygments, pyasn1-modules, psutil, protobuf, prompt-toolkit, Pillow, parso, oauthlib, numpy, MarkupSafe, kiwisolver, idna, fonttools, decorator, cycler, colorama, charset-normalizer, cachetools, absl-py, werkzeug, tqdm, torch, scipy, requests, python-dateutil, packaging, opencv-python, matplotlib-inline, jedi, importlib-metadata, grpcio, google-auth, contourpy, asttokens, torchvision, thop, stack-data, requests-oauthlib, pandas, matplotlib, markdown, seaborn, ipython, google-auth-oauthlib, tensorboard
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
Successfully installed MarkupSafe-2.1.1 Pillow-9.3.0 PyYAML-6.0 absl-py-1.3.0 asttokens-2.1.0 backcall-0.2.0 cachetools-5.2.0 charset-normalizer-2.1.1 colorama-0.4.6 contourpy-1.0.6 cycler-0.11.0 decorator-5.1.1 executing-1.2.0 fonttools-4.38.0 google-auth-2.14.1 google-auth-oauthlib-0.4.6 grpcio-1.50.0 idna-3.4 importlib-metadata-5.0.0 ipython-8.6.0 jedi-0.18.1 kiwisolver-1.4.4 markdown-3.4.1 matplotlib-3.6.2 matplotlib-inline-0.1.6 numpy-1.23.4 oauthlib-3.2.2 opencv-python-4.6.0.66 packaging-21.3 pandas-1.5.1 parso-0.8.3 pickleshare-0.7.5 prompt-toolkit-3.0.32 protobuf-3.20.3 psutil-5.9.4 pure-eval-0.2.2 pyasn1-0.4.8 pyasn1-modules-0.2.8 pygments-2.13.0 pyparsing-3.0.9 python-dateutil-2.8.2 pytz-2022.6 requests-2.28.1 requests-oauthlib-1.3.1 rsa-4.9 scipy-1.9.3 seaborn-0.12.1 six-1.16.0 stack-data-0.6.0 tensorboard-2.11.0 tensorboard-data-server-0.6.1 tensorboard-plugin-wit-1.8.1 thop-0.1.1.post2209072238 torch-1.13.0 torchvision-0.14.0 tqdm-4.64.1 traitlets-5.5.0 typing-extensions-4.4.0 urllib3-1.26.12 wcwidth-0.2.5 werkzeug-2.2.2 zipp-3.10.0
(LgpLoYo5) E:\AIWorkSpace\yolov5-master>
(LgpLoYo5) E:\AIWorkSpace\yolov5-master>
五命黔、圖片數據標記軟件
數據標注軟件市面上有很多呜呐，也有一些是自家開發(fā)的，此處介紹兩種悍募，轉換為yolo5所需要的數據蘑辑，本文使用的是精靈標注助手。
- 精靈標注助手
這款軟件免費強大坠宴，可以多點標記洋魂，但目前掌握的腳本只支持4點矩形轉換為Yolo的訓練數據，所以畫矩形，導出pascal-voc格式的數據副砍。 - labelimg(他是pyqt5寫的衔肢，支持三種框架，但只能畫4個點的矩形框)
此處使用了臨時鏡像豁翎，原語句為pip install labelimg
pip install -i https://pypi.tuna.tsinghua.edu.cn/simple labelimg
安裝日志：
(LgpLoYo5) E:\AIWorkSpace\yolov5-master>pip install -i https://pypi.tuna.tsinghua.edu.cn/simple labelimg
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
Collecting labelimg
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/c5/fb/9947097363fbbfde3921f7cf7ce9800c89f909d26a506145aec37c75cda7/labelImg-1.8.6.tar.gz (247 kB)
---------------------------------------- 247.7/247.7 kB 892.6 kB/s eta 0:00:00
Preparing metadata (setup.py) ... done
Collecting pyqt5
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/bd/85/31a12415765acb48fddac3e207cfffcbbae826fe194cf1d92179d8872f59/PyQt5-5.15.7-cp37-abi3-win_amd64.whl (6.8 MB)
---------------------------------------- 6.8/6.8 MB 2.8 MB/s eta 0:00:00
Collecting lxml
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/17/78/7cc7e269c7c58f0b94c4abdf6cf2bcce2fb0ca58d415e6e4e1b805cd9f17/lxml-4.9.1-cp39-cp39-win_amd64.whl (3.6 MB)
---------------------------------------- 3.6/3.6 MB 2.4 MB/s eta 0:00:00
Collecting PyQt5-Qt5>=5.15.0
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/37/97/5d3b222b924fa2ed4c2488925155cd0b03fd5d09ee1cfcf7c553c11c9f66/PyQt5_Qt5-5.15.2-py3-none-win_amd64.whl (50.1 MB)
---------------------------------------- 50.1/50.1 MB 864.1 kB/s eta 0:00:00
Collecting PyQt5-sip<13,>=12.11
Downloading https://pypi.tuna.tsinghua.edu.cn/packages/d1/85/e30b11daf7b8d4e00ed51c17204ec3df446041206d22087d2c5dc17d8543/PyQt5_sip-12.11.0-cp39-cp39-win_amd64.whl (78 kB)
---------------------------------------- 78.3/78.3 kB 1.4 MB/s eta 0:00:00
Building wheels for collected packages: labelimg
Building wheel for labelimg (setup.py) ... done
Created wheel for labelimg: filename=labelImg-1.8.6-py2.py3-none-any.whl size=261524 sha256=fbba7beb072b63726ff5f9032343755ca971f144630c1808903f87c5ff277364
Stored in directory: c:\users\newbeechina\appdata\local\pip\cache\wheels\d6\ba\ac\0b1de7ebb0074df1740d334b53680cbca868c0d4792450d895
Successfully built labelimg
Installing collected packages: PyQt5-Qt5, PyQt5-sip, lxml, pyqt5, labelimg
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
WARNING: Ignore distutils configs in setup.cfg due to encoding errors.
Successfully installed PyQt5-Qt5-5.15.2 PyQt5-sip-12.11.0 labelimg-1.8.6 lxml-4.9.1 pyqt5-5.15.7
六角骤、數據標記
精靈標注助手
新建?位置標注?填寫 項目名稱、設置標注圖片文件夾心剥、設置即將標注的分類邦尊，用矩形框進行標記，你要問我能不能用多邊形框進行標記刘陶，從我目前獲得的腳本來看只能轉換四點的bndbox無法轉換cubic_bezier胳赌。-
labelimg
在當前虛擬環(huán)境中敲擊labelimg會打開標記軟件，目前它僅支持PascalVOC匙隔、YOLO疑苫、CreateML三種標記，它默認是PascalVOC所以我們需要將其切換為YOLO
image.png
點擊Open Dir -> 選擇需要標注的文件夾 -> ok
七纷责、數據籌備
- 理論指引
訓練集(train set)：訓練模型
驗證集(val set)：評估模型捍掺，用來調整模型參數從而選擇最優(yōu)模型
測試集(test set)：一旦找到了最佳參數，就開始最終訓練
一組數據大致分為以上三類再膳，
在深度學習中挺勿，由于數據量本身很大，而且訓練神經網絡需要的數據很多喂柒，可以把更多的數據分給training不瓶，而相應減少validation和test，這三者一般的比例為training：validation：test = 2：1：1, 但是有些時候如果模型不需要很多調整只要擬合就可時灾杰，或者training本身就是training+validation (比如cross validation)時蚊丐，也可以比例為training：test =7：3，這一段描述從網上找的艳吠，準確性不敢保證但好歹我現在能配置比例了麦备。 - 實操
建立以下文件夾
1）根目錄/data/Annotations儲存pascal-voc格式.xml數據。
2）根目錄/data/ImageSets存儲腳本生成的train.txt昭娩、test.txt凛篙、val.txt、trainval.txt;這幾個文本文件存儲的待訓練圖片文件名稱栏渺。
3）根目錄/data/JPEGImages存儲所有訓練圖片
在根目錄創(chuàng)建preLabelsTxt.py,運行后在 ./data/ImageSets 會生成數據集分類txt文件,內容是一批圖片名
preLabelsTxt.py：
import os
import random
ROOT_PATH = 'E:/PyWorkSpace/LgpYolov53/'
trainval_percent = 0.3
train_percent = 0.7
xmlfilepath = ROOT_PATH + 'data/Annotations'
txtsavepath = ROOT_PATH + 'data/ImageSets'
total_xml = os.listdir(xmlfilepath)
num = len(total_xml)
list = range(num)
tv = int(num * trainval_percent)
tr = int(tv * train_percent)
trainval = random.sample(list, tv)
train = random.sample(trainval, tr)
ftrainval = open(ROOT_PATH + 'data/ImageSets/trainval.txt', 'w',encoding='utf-8')
ftest = open(ROOT_PATH + 'data/ImageSets/test.txt', 'w',encoding='utf-8')
ftrain = open(ROOT_PATH + 'data/ImageSets/train.txt', 'w',encoding='utf-8')
fval = open(ROOT_PATH + 'data/ImageSets/val.txt', 'w',encoding='utf-8')
for i in list:
name = total_xml[i][:-4] + '\n'
if i in trainval:
ftrainval.write(name)
if i in train:
ftest.write(name)
else:
fval.write(name)
else:
ftrain.write(name)
ftrainval.close()
ftrain.close()
fval.close()
ftest.close()
在根目錄創(chuàng)建convertLabels.py呛梆，運行后會生成根目錄/data/labels/...
在labels文件夾下就是Yolo5框架訓練所需的標注數據集，并且在/data/目錄下生成test.txt迈嘹、train.txt削彬、val.txt 三個帶儲存路徑的txt圖片數據集
convertLabels.py：
import xml.etree.ElementTree as ET
import pickle
import os
from os import listdir, getcwd
from os.path import join
ROOT_PATH = 'E:/PyWorkSpace/LgpYolov53/'
sets = ['train', 'test','val']
classes = ['狗']
def convert(size, box):
dw = 1. / size[0]
dh = 1. / size[1]
x = (box[0] + box[1]) / 2.0
y = (box[2] + box[3]) / 2.0
w = box[1] - box[0]
h = box[3] - box[2]
x = x * dw
w = w * dw
y = y * dh
h = h * dh
return (x, y, w, h)
def convert_annotation(image_id):
in_file = open(ROOT_PATH+'data/Annotations/%s.xml' % (image_id),'r',encoding='utf-8')
out_file = open(ROOT_PATH+'data/labels/%s.txt' % (image_id), 'w',encoding='utf-8')
tree = ET.parse(in_file)
root = tree.getroot()
size = root.find('size')
w = int(size.find('width').text)
h = int(size.find('height').text)
for obj in root.iter('object'):
difficult = obj.find('difficult').text
cls = obj.find('name').text
if cls not in classes or int(difficult) == 1:
continue
cls_id = classes.index(cls)
xmlbox = obj.find('bndbox')
b = (float(xmlbox.find('xmin').text), float(xmlbox.find('xmax').text), float(xmlbox.find('ymin').text),
float(xmlbox.find('ymax').text))
bb = convert((w, h), b)
out_file.write(str(cls_id) + " " + " ".join([str(a) for a in bb]) + '\n')
wd = getcwd()
print(wd)
for image_set in sets:
if not os.path.exists('data/labels/'):
os.makedirs('data/labels/')
image_ids = open(ROOT_PATH+'data/ImageSets/%s.txt' % (image_set),'r',encoding='utf-8').read().strip().split()
list_file = open(ROOT_PATH+'data/%s.txt' % (image_set), 'w',encoding='utf-8')
for image_id in image_ids:
list_file.write(ROOT_PATH+'data/JPEGImages/%s.png\n' % (image_id))
print("lgp："+image_id)
convert_annotation(image_id)
list_file.close()
八全庸、訓練籌備
1.修改yolov5l.yaml 的nc值
在根目錄/models/文件夾下找到y(tǒng)olov5l.yaml， n融痛、s壶笼、m、l雁刷、x幾個文件因為其配置的參數不同覆劈，所以需要訓練的時間依次增加，參數不要動只修改nc值沛励，nc:【 你需要訓練的類型數量】责语，我訓練的只有一個“狗”所以我的修改為nc:1
2.創(chuàng)建自己的訓練文件
根目錄/data/文件夾下創(chuàng)建【你自己要訓練的】.yaml，我的是LgpDog.yaml目派，我從coco.yaml復制部分內容并修改為
path: E:/PyWorkSpace/LgpYolov53/data # dataset root dir
train: train.txt # train images (relative to 'path') 118287 images
val: val.txt # train images (relative to 'path') 5000 images
test: test-dev.txt # 20288 of 40670 images, submit to https://competitions.codalab.org/competitions/20794
# Classes
nc: 1 # number of classes
names: ['狗'] # class names
文件中配置的train.txt與另兩個txt文件就是convertLabels.py腳本生成的三個訓練集文件坤候，里面存儲的是圖片的路徑。
3.修改根目錄/train.py文件
a）修改參數：
修改參數如下圖企蹭，
--weights是訓練的權重文件,初始訓練時是沒有訓練數據的白筹，所以用官方默認的，這里第1次不用改谅摄，之后可以改成自己訓練的.pt文件路徑徒河。
--cfg就是創(chuàng)建的
--data 就是自己建立的訓練配置，我的是LgpDos.yaml送漠，里面配置了訓練的類型顽照，此處，以及圖片源路徑
--epochs就是訓練迭代次數
--device暫時未弄清楚闽寡，貌似是后續(xù)檢測腳本detect.py執(zhí)行用的
--name保存訓練時.pt文件的文件夾名代兵，改不改都行
image.png
b）ROOT可能取得C盤路徑，所以直接注釋掉爷狈，這樣就是自己的項目路徑了奢人，這里需要注意
image.png
然后在虛擬環(huán)境中運行train.py，若是第1次執(zhí)行會下載一些東西淆院，譬如.pt文件，從官網上下載的yolos.pt會在根目錄下存在句惯，若根目錄下已經存在這個文件那就不會下載土辩，這樣省流量，在執(zhí)行過程中遇到過各種錯誤：
錯1：
ModuleNotFoundError: No module named 'yaml'
Requirement already satisfied: pyyaml in d:\miniconda3\envs\lgpyolo5\lib\sit
沒包一般就是環(huán)境進錯了抢野，關閉窗口使用conda命令重新進入目標虛擬環(huán)境并切換到你的工程目錄拷淘，之后再執(zhí)行train.py
錯2：
RuntimeError: result type Float can't be cast to the desired output type __int64
這是官方文件出問題了，解決方式未打開 根目錄/utils/loss.py文件
CTRL+F搜索for i in range(self.nl)
將
anchors, shape = self.anchors[i]
替換為
anchors, shape = self.anchors[i], p[i].shape
CTRL+F搜索indices.append
將
indices.append((b, a, gj.clamp_(0, gain[3] - 1), gi.clamp_(0, gain[2] - 1)))
替換為
indices.append((b, a, gj.clamp_(0, shape[2] - 1), gi.clamp_(0, shape[3] - 1))) # image, anchor, grid
錯3：說報一堆文件與labels找不到的錯誤
貌似是訓練的標注數據是通過圖片路徑查找的指孤，所以將標注的一批.txt文件與圖片文件放在一個文件夾里启涯，我這里圖片在JPEGImages贬堵，所以
九、訓練可視化
另起一個cmd窗口,如下指令结洼，激活虛擬環(huán)境黎做，敲擊命令tensorboard --logdir 【你的Yolo5的項目路徑】\runs松忍，在瀏覽器輸入http://localhost:6006
執(zhí)行日志：
C:\Users\22090201>conda activate LgpYolo5
(LgpYolo5) C:\Users\22090201>tensorboard --logdir E:\PyWorkSpace\LgpYolov53\runs
TensorFlow installation not found - running with reduced feature set.
Serving TensorBoard on localhost; to expose to the network, use a proxy or pass --bind_all
TensorBoard 2.11.0 at http://localhost:6006/ (Press CTRL+C to quit)
指標含義：
LOSS
loss分為cls_loss, box_loss, obj_loss三部分蒸殿。
cls_loss用于監(jiān)督類別分類，計算錨框與對應的標定分類是否正確鸣峭。
box_loss用于監(jiān)督檢測框的回歸宏所，預測框與標定框之間的誤差（CIoU）。
obj_loss用于監(jiān)督grid中是否存在物體摊溶，計算網絡的置信度爬骤。
mAP(IoU@0.75)，這是一個對檢測能力要求更高的標準莫换。
mAP(IoU@0.5)霞玄，跟Pascal VOC mAP標準計算方式一致；
mAP(IoU@[0.5:0.05:0.95])浓镜，需要計算10個IoU閾值下的mAP溃列，然后計算平均值。這個評估指標比僅考慮通用IoU閾值(0.5)評估指標更能體現出模型的精度膛薛。
參考資料：https://blog.csdn.net/u011994454/article/details/119564834
https://pytorch.apachecn.org/#/docs/1.7/19
十听隐、檢測目標圖片
訓練完后會在根目錄\runs\train\exp11\weights\ 文件夾下生成best.pt、last.pt兩個文件 我們使用best.pt權重文件
在虛擬環(huán)境中cd 到Yolo5所在目錄哄啄，敲擊命令
python detect.py --weights 【權重文件路徑】 --source 【目標檢測文件夾】 --device 【訓練顯卡或cpu】 --save-txt
示例：
python detect.py --weights runs/train/exp11/weights/best.pt --source ./data/JPEGImages --device cpu --save-txt
敲完之后會自動在runs/detect/exp生成文件夾雅任，里面存放著你目標檢測文件，如下圖上面有相似度咨跌，樣本+訓練資源+訓練時長+訓練參數+訓練框架=不同的效果沪么，這張圖的相似度很低，因為我訓練1.5h且使用1個cpu锌半，資源量很低禽车，另一方面我訓練的是狗但我檢測的是狼，我突然想這兩物種連人都很難分辨刊殉，AI檢測啥樣子殉摔，從效果來看這塊確實是人工智障，拆解成元素來看狼狗分辨主要是情感记焊，尾巴下垂 逸月，但據說哈士奇尾巴是可下可上的，這個就難搞遍膜，這點我還沒有想到有啥解碗硬，這個問題解決需要對兩者進行解構瓤湘，咱畢竟掌握信息元素不夠。
image.png
