Yolov11入门
一、安装MiniConda
Anaconda官网https://www.anaconda.com/download
清华镜像https://mirrors.tuna.tsinghua.edu.cn/help/anaconda/
二、下载yolov11源码
ultralyticshttps://github.com/ultralytics/ultralytics/tree/v8.3.39
三、 Conda使用
3.1 查看conda
conda list
C:\Users\wydcg>conda list
# packages in environment at E:\SoftWare\Miniconda3:
#
# Name Version Build Channel
anaconda-anon-usage 0.7.1 py313hfc23b7f_100
anaconda_powershell_prompt 1.1.0 haa95532_1
anaconda_prompt 1.1.0 haa95532_1
annotated-types 0.6.0 py313haa95532_0
archspec 0.2.3 pyhd3eb1b0_0
boltons 25.0.0 py313haa95532_0
brotlicffi 1.0.9.2 py313h5da7b33_1
bzip2 1.0.8 h2bbff1b_6
ca-certificates 2025.2.25 haa95532_0
certifi 2025.7.14 py313haa95532_0
cffi 1.17.1 py313h827c3e9_1
charset-normalizer 3.3.2 pyhd3eb1b0_0
colorama 0.4.6 py313haa95532_0
conda 25.5.1 py313haa95532_0
conda-anaconda-telemetry 0.2.0 py313haa95532_1
conda-anaconda-tos 0.2.1 py313haa95532_0
conda-content-trust 0.2.0 py313haa95532_1
conda-libmamba-solver 25.4.0 pyhd3eb1b0_0
conda-package-handling 2.4.0 py313haa95532_0
conda-package-streaming 0.12.0 py313haa95532_0
cpp-expected 1.1.0 h214f63a_0
cryptography 45.0.3 py313h51e0144_0
distro 1.9.0 py313haa95532_0
expat 2.7.1 h8ddb27b_0
fmt 9.1.0 h6d14046_1
frozendict 2.4.2 py313haa95532_0
idna 3.7 py313haa95532_0
jsonpatch 1.33 py313haa95532_1
jsonpointer 2.1 pyhd3eb1b0_0
libarchive 3.7.7 h9243413_0
libcurl 8.14.1 ha9f67de_0
libffi 3.4.4 hd77b12b_1
libiconv 1.16 h2bbff1b_3
libmamba 2.0.5 hcd6fe79_1
libmambapy 2.0.5 py313h214f63a_1
libmpdec 4.0.0 h827c3e9_0
libsolv 0.7.30 hf2fb9eb_1
libssh2 1.11.1 h2addb87_0
libxml2 2.13.8 h866ff63_0
lz4-c 1.9.4 h2bbff1b_1
markdown-it-py 2.2.0 py313haa95532_1
mdurl 0.1.0 py313haa95532_0
menuinst 2.3.0 py313h5da7b33_0
nlohmann_json 3.11.2 h6c2663c_0
openssl 3.0.16 h3f729d1_0
packaging 24.2 py313haa95532_0
pcre2 10.42 h0ff8eda_1
pip 25.1 pyhc872135_2
platformdirs 4.3.7 py313haa95532_0
pluggy 1.5.0 py313haa95532_0
pybind11-abi 5 hd3eb1b0_0
pycosat 0.6.6 py313h827c3e9_2
pycparser 2.21 pyhd3eb1b0_0
pydantic 2.11.7 py313haa95532_0
pydantic-core 2.33.2 py313h215eeae_0
pygments 2.19.1 py313haa95532_0
pysocks 1.7.1 py313haa95532_0
python 3.13.5 h286a616_100_cp313
python_abi 3.13 0_cp313
reproc 14.2.4 hd77b12b_2
reproc-cpp 14.2.4 hd77b12b_2
requests 2.32.4 py313haa95532_0
rich 13.9.4 py313haa95532_0
ruamel.yaml 0.18.10 py313h827c3e9_0
ruamel.yaml.clib 0.2.12 py313h827c3e9_0
setuptools 78.1.1 py313haa95532_0
simdjson 3.10.1 h214f63a_0
spdlog 1.11.0 h59b6b97_0
sqlite 3.50.2 hda9a48d_1
tk 8.6.14 h5e9d12e_1
tqdm 4.67.1 py313h4442805_0
truststore 0.10.0 py313haa95532_0
typing-extensions 4.12.2 py313haa95532_0
typing-inspection 0.4.0 py313haa95532_0
typing_extensions 4.12.2 py313haa95532_0
tzdata 2025b h04d1e81_0
ucrt 10.0.22621.0 haa95532_0
urllib3 2.5.0 py313haa95532_0
vc 14.3 h2df5915_10
vc14_runtime 14.44.35208 h4927774_10
vs2015_runtime 14.44.35208 ha6b5a95_10
wheel 0.45.1 py313haa95532_0
win_inet_pton 1.1.0 py313haa95532_0
xz 5.6.4 h4754444_1
yaml-cpp 0.8.0 hd77b12b_1
zlib 1.2.13 h8cc25b3_1
zstandard 0.23.0 py313h4fc1ca9_1
zstd 1.5.6 h8880b57_0
3.2 查看有哪些虚拟环境
- conda env list
- conda info --envs
C:\Users\wydcg>conda env list
# conda environments:
#
base E:\SoftWare\Miniconda3
C:\Users\wydcg> conda info --envs
# conda environments:
#
yolov11 C:\Users\wydcg\.conda\envs\yolov11
base E:\SoftWare\Miniconda3
3.3 设置默认conda目录
cond config --add envs_dirs E:/Code/Conda
设置了默认conda目录后可以直接在该目录下创建虚拟环境,不用再指定目录
conda create -n yolov11 python=3.11.9
C:\Windows\System32>conda info --env
# conda environments:
#
E:\Code\Conda\yolov11
base E:\SoftWare\Miniconda3
C:\Windows\System32>conda config --add envs_dirs E:/Code/Conda
C:\Windows\System32>conda info --env
# conda environments:
#
yolov11 E:\Code\Conda\yolov11
base E:\SoftWare\Miniconda3
3.4 创建conda环境
-
conda create -p E:/Code/Conda/yolov11 python=3.11.9
-
修复命名问题（无法使用项目名访问虚拟环境）
conda config --add envs_dirs E:\Code\Conda
C:\Users\wydcg>conda create -p E:/Code/Conda/yolov11 python=3.11.9
3 channel Terms of Service accepted
Channels:
- defaults
Platform: win-64
Collecting package metadata (repodata.json): done
Solving environment: done
## Package Plan ##
environment location: E:\Code\Conda\yolov11
added / updated specs:
- python=3.11.9
done
#
# To activate this environment, use
#
# $ conda activate E:\Code\Conda\yolov11
#
# To deactivate an active environment, use
#
# $ conda deactivate
3.5 激活环境
- conda activate E:\Code\Conda\yolov11
- conda activate yolov11 (设置了虚拟环境名称才可使用)
C:\Windows\System32>conda activate E:/Code/Conda/yolov11
(E:\Code\Conda\yolov11) C:\Windows\System32>
C:\Windows\System32>conda activate yolov11
(E:\Code\Conda\yolov11) C:\Windows\System32>
3.6 退出环境
conda deactivate
(E:\Code\Conda\yolov11) C:\Windows\System32>conda deactivate
C:\Windows\System32>
3.7 删除conda环境
- conda remove -n yolov11 --all
- conda env remove -p E:/Code/Conda/yolov11
C:\Users\wydcg>conda remove -n yolov11 --all
3 channel Terms of Service accepted
Remove all packages in environment C:\Users\wydcg\.conda\envs\yolov11:
## Package Plan ##
environment location: C:\Users\wydcg\.conda\envs\yolov11
The following packages will be REMOVED:
bzip2-1.0.8-h2bbff1b_6
ca-certificates-2025.2.25-haa95532_0
libffi-3.4.4-hd77b12b_1
openssl-3.0.17-h35632f6_0
pip-25.1-pyhc872135_2
python-3.11.9-he1021f5_0
setuptools-78.1.1-py311haa95532_0
sqlite-3.50.2-hda9a48d_1
tk-8.6.14-h5e9d12e_1
tzdata-2025b-h04d1e81_0
ucrt-10.0.22621.0-haa95532_0
vc-14.3-h2df5915_10
vc14_runtime-14.44.35208-h4927774_10
vs2015_runtime-14.44.35208-ha6b5a95_10
wheel-0.45.1-py311haa95532_0
xz-5.6.4-h4754444_1
zlib-1.2.13-h8cc25b3_1
Proceed ([y]/n)? y
Downloading and Extracting Packages:
Preparing transaction: done
Verifying transaction: done
Executing transaction: done
Everything found within the environment (C:\Users\wydcg\.conda\envs\yolov11), including any conda environment configurations and any non-conda files, will be deleted. Do you wish to continue?
(y/[n])? y
C:\Users\wydcg>conda env list
# conda environments:
#
base E:\SoftWare\Miniconda3
3.8 关闭自动激活的base环境
conda config --set auto_activate_base false
3.9 恢复自动激活base环境
conda config --set auto_activate_base true
3.10 命令总结
conda create 创建一个新的环境
conda activate 激活一个环境
conda deactivate 停用当前环境
conda remove 删除一个环境
conda list 列出已安装的包
conda install 安装一个包
conda update 更新已安装的包
conda search 搜索可用的包
conda env export 导出环境列表到一个文件中
conda env create -f file 从一个环境文件中创建一个新的环境
conda config --set auto_activate_base 修改配置文件来设置是否自动激活base环境
四、下载依赖
以下所有操作都在虚拟环境中进行
# 注意该依赖会自动安装cpu版本pytorch，要先卸载掉cpu版本再安装gpu版本
pip install ultralytics==8.3.39
# 查看pytorch版本
import torch
print(torch.__version__) # 查看PyTorch版本
# cu代表cuda，即GPU版，无需卸载
print(torch.cuda.is_available()) # 检查是否支持GPU
# True 表示可以使用GPU
# 卸载cpu版pytorch
pip uninstall torch torchvision torchaudio
五、yolov11应用
vscode中选择创建好的虚拟环境中的解释器
5.1 运行示例代码
from ultralytics import YOLO
# load the YOLOv11 model
model = YOLO("yolo11n.pt")
# perform inference on an image
results = model.predict(source="./ultralytics/assets/", conf=0.25, save=True)
5.2 图片打标
pip install labelimg
出错的处理办法：https://blog.csdn.net/wangpjpj/article/details/142451247
5.3 目录分级
目录 /dataset/yolov11/images/train\val /dataset/yolov11/labels/train\val
5.4 训练
-
数据集配置文件 data.yaml
path: ./datasets/animal # dataset root dir train: ./datasets/animal/train # train images (relative to 'path') 1281167 images val: ./datasets/animal/val # val images (relative to 'path') 50000 images test: # test images (optional) # Classes names: 0: dog 1: cat 2: cow 3: pig -
训练文件 train.py
from ultralytics import YOLO model = YOLO("yolo11n.pt") # train the model result = model.train(data="./data.yaml", epochs=100, imgsz=640, batch=16) # perform inference on an image results = model.predict(source="../dataset/animal/images/test", conf=0.25, save=True) # print the results print(results) -
推理文件 inference.py
# use the best model to perform inference on an image from ultralytics import YOLO # load the best model from training model = YOLO("../runs/detect/train/weights/best.pt") results = model.predict(source="../dataset/animal/images/test", conf=0.25, save=True) # print the results print(results)
评论区