YOLO Env Config

    目录: D:\gesture_recog\ultralytics-8.1.0


Mode                 LastWriteTime         Length Name
----                 -------------         ------ ----
d-----         2025/3/16     22:09                .github
d-----         2025/3/16     22:09                docker
d-----         2025/3/16     22:09                docs
d-----         2025/3/16     22:09                examples
d-----         2025/3/16     22:09                tests
d-----         2025/3/16     22:09                ultralytics
-a----         2025/3/16     22:09           2295 .gitignore
-a----         2025/3/16     22:09           2408 .pre-commit-config.yaml
-a----         2025/3/16     22:09            612 CITATION.cff
-a----         2025/3/16     22:09           5585 CONTRIBUTING.md
-a----         2025/3/16     22:09          34523 LICENSE
-a----         2025/3/16     22:09           6728 pyproject.toml
-a----         2025/3/16     22:09          35898 README.md
-a----         2025/3/16     22:09          34774 README.zh-CN.md


(yolov8) PS D:\gesture_recog\ultralytics-8.1.0> pip install -e .

pip install jupyterlab tensorboard

pip uninstall pillow # 删除原先的 9.3.0 版本
pip install pillow==9.3.0 # 重新下载一遍，好了

(yolov8) PS D:\gesture_recog\ultralytics-8.1.0> yolo
C:\Users\Chi\miniconda3\envs\yolov8\lib\site-packages\torchvision\io\image.py:14: UserWarning: Failed to load image Python extension: '[WinError 127] 找不到指定的程序。'If you don't plan on using image functionality from `torchvision.io`, you can ignore this warning. Otherwise, there might be something wrong with your environment. Did you have `libjpeg` or `libpng` installed before building `torchvision` from source?
  warn(

    Arguments received: ['yolo']. Ultralytics 'yolo' commands use the following syntax:

        yolo TASK MODE ARGS

        Where   TASK (optional) is one of ('detect', 'segment', 'classify', 'pose', 'obb')
                MODE (required) is one of ('train', 'val', 'predict', 'export', 'track', 'benchmark')
                ARGS (optional) are any number of custom 'arg=value' pairs like 'imgsz=320' that override defaults.
                    See all ARGS at https://docs.ultralytics.com/usage/cfg or with 'yolo cfg'

    1. Train a detection model for 10 epochs with an initial learning_rate of 0.01
        yolo train data=coco128.yaml model=yolov8n.pt epochs=10 lr0=0.01

    2. Predict a YouTube video using a pretrained segmentation model at image size 320:
        yolo predict model=yolov8n-seg.pt source='https://youtu.be/LNwODJXcvt4' imgsz=320

    3. Val a pretrained detection model at batch-size 1 and image size 640:
        yolo val model=yolov8n.pt data=coco128.yaml batch=1 imgsz=640

    4. Export a YOLOv8n classification model to ONNX format at image size 224 by 128 (no TASK required)
        yolo export model=yolov8n-cls.pt format=onnx imgsz=224,128

    5. Explore your datasets using semantic search and SQL with a simple GUI powered by Ultralytics Explorer API
        yolo explorer

    7. Run special commands:
        yolo help
        yolo checks
        yolo version
        yolo settings
        yolo copy-cfg
        yolo cfg

    Docs: https://docs.ultralytics.com
    Community: https://community.ultralytics.com
    GitHub: https://github.com/ultralytics/ultralytics

(yolov8) D:\gesture_recog\ultralytics-8.1.0>yolo mode='predict' task='detect' model='yolov8m.pt' source='00001.jpeg'
C:\Users\Chi\miniconda3\envs\yolov8\lib\site-packages\torchvision\io\image.py:14: UserWarning: Failed to load image Python extension: '[WinError 127] 找不到指定的程序。'If you don't plan on using image functionality from `torchvision.io`, you can ignore this warning. Otherwise, there might be something wrong with your environment. Did you have `libjpeg` or `libpng` installed before building `torchvision` from source?
  warn(
D:\gesture_recog\ultralytics-8.1.0\ultralytics\nn\tasks.py:634: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  return torch.load(file, map_location="cpu"), file  # load

image 1/1 D:\gesture_recog\ultralytics-8.1.0\00001.jpeg: 384x640 19 cars, 4 buss, 70.9ms
Speed: 7.6ms preprocess, 70.9ms inference, 148.1ms postprocess per image at shape (1, 3, 384, 640)
💡 Learn more at https://docs.ultralytics.com/modes/predict

(yolov8) D:\gesture_recog\ultralytics-8.1.0>conda install torchvision -c pytorch

(yolov8) D:\gesture_recog\ultralytics-8.1.0>yolo mode='predict' task='detect' model='yolov8m.pt' source='00001.jpeg'
Traceback (most recent call last):
  File "C:\Users\Chi\miniconda3\envs\yolov8\lib\runpy.py", line 196, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "C:\Users\Chi\miniconda3\envs\yolov8\lib\runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "C:\Users\Chi\miniconda3\envs\yolov8\Scripts\yolo.exe\__main__.py", line 4, in <module>
  File "D:\gesture_recog\ultralytics-8.1.0\ultralytics\__init__.py", line 5, in <module>
    from ultralytics.data.explorer.explorer import Explorer
  File "D:\gesture_recog\ultralytics-8.1.0\ultralytics\data\__init__.py", line 3, in <module>
    from .base import BaseDataset
  File "D:\gesture_recog\ultralytics-8.1.0\ultralytics\data\base.py", line 17, in <module>
    from ultralytics.utils import DEFAULT_CFG, LOCAL_RANK, LOGGER, NUM_THREADS, TQDM
  File "D:\gesture_recog\ultralytics-8.1.0\ultralytics\utils\__init__.py", line 19, in <module>
    import matplotlib.pyplot as plt
  File "C:\Users\Chi\miniconda3\envs\yolov8\lib\site-packages\matplotlib\__init__.py", line 161, in <module>
    from . import _api, _version, cbook, _docstring, rcsetup
  File "C:\Users\Chi\miniconda3\envs\yolov8\lib\site-packages\matplotlib\rcsetup.py", line 28, in <module>
    from matplotlib.colors import Colormap, is_color_like
  File "C:\Users\Chi\miniconda3\envs\yolov8\lib\site-packages\matplotlib\colors.py", line 52, in <module>
    from PIL import Image
  File "C:\Users\Chi\miniconda3\envs\yolov8\lib\site-packages\PIL\Image.py", line 100, in <module>
    from . import _imaging as core
ImportError: DLL load failed while importing _imaging: 找不到指定的模块。

conda install -c conda-forge pillow

from ultralytics import YOLO

yolo = YOLO(model='..\Models\yolov8m.pt', task='detect')

result=yolo(source=0, save=True)

ultralytics\cfg\default.yaml # 存放参数和一些默认参数形式

import torch
print(torch.cuda.is_available())

yolo task=detect mode=train model=..\Model\yolov8n.pt data=D:/gesture_recog/ASL/data.yaml epochs=50 imgsz=640 batch=4 device=0

# 验证模型
yolo task=detect mode=val model=runs/detect/train/weights/best.pt data=D:/gesture_recog/ASL/data.yaml

(yolov8) D:\facial_recog\ultralytics-8.1.0>yolo task=detect mode=val model=runs/detect/train3/weights/best.pt data=D:\facial_recog\Faces\data.yaml     

(yolov8) D:\gesture_recog\ultralytics-8.1.0>yolo task=detect mode=val model=runs/detect/train/weights/best.pt data=D:/gesture_recog/ASL/data.yaml
D:\gesture_recog\ultralytics-8.1.0\ultralytics\nn\tasks.py:634: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  return torch.load(file, map_location="cpu"), file  # load
Ultralytics YOLOv8.1.0 🚀 Python-3.10.16 torch-2.5.1 CUDA:0 (NVIDIA GeForce RTX 3050 Ti Laptop GPU, 4096MiB)
Model summary (fused): 168 layers, 3010718 parameters, 0 gradients, 8.1 GFLOPs
val: Scanning D:\gesture_recog\ASL\valid\labels.cache... 1320 images, 60 backgrounds, 0 corrupt: 100%|██████████| 1320/1320 [00:00<?, ?it/s]
                 Class     Images  Instances      Box(P          R      mAP50  mAP50-95): 100%|██████████| 83/83 [00:13<00:00,  6.27it/s]
                   all       1320       1270      0.921      0.889      0.945      0.785
                     A       1320         41      0.989      0.927      0.963      0.812
                     B       1320         53      0.877          1      0.989      0.839
                     C       1320         51      0.944      0.922       0.98      0.777
                     D       1320         44      0.906      0.876      0.949      0.761
                     E       1320         46          1      0.909      0.949      0.862
                     F       1320         49      0.981      0.898      0.983      0.895
                     G       1320         51      0.843      0.841      0.897      0.761
                     H       1320         46      0.903      0.957      0.956      0.773
                     I       1320         48      0.721      0.729      0.782      0.536
                     J       1320         48      0.954      0.861      0.938      0.697
                     K       1320         51      0.899      0.804      0.893      0.726
                     L       1320         56          1      0.905      0.984      0.826
                     M       1320         50      0.934       0.92      0.982      0.845
                     N       1320         29      0.891       0.85      0.968       0.83
                     O       1320         54      0.958      0.836      0.947      0.701
                     P       1320         44      0.943      0.886      0.939      0.687
                     Q       1320         44      0.831      0.818      0.858      0.744
                     R       1320         52      0.909      0.957      0.974      0.875
                     S       1320         67      0.962      0.881      0.965      0.851
                     T       1320         45      0.926      0.911      0.959       0.89
                     U       1320         53      0.938      0.943      0.977      0.852
                     V       1320         48      0.937      0.929      0.961      0.798
                     W       1320         46      0.977      0.942      0.984      0.842
                     X       1320         57      0.981      0.895       0.96      0.723
                     Y       1320         55      0.928      0.938       0.97      0.802
                     Z       1320         42      0.824      0.786      0.867      0.716
Speed: 0.4ms preprocess, 4.3ms inference, 0.0ms loss, 1.3ms postprocess per image
Results saved to runs\detect\val2
💡 Learn more at https://docs.ultralytics.com/modes/val

# 使用模型预测
yolo task=detect mode=predict model=runs/detect/train/weights/best.pt source=path/to/images_or_video

(yolov8) D:\gesture_recog\Flask>pip install flask