Skip to content

docs: 更新CUDA提醒 #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
SuperWaterGod wants to merge 1 commit into
base: main
Choose a base branch
from
Open

docs: 更新CUDA提醒 #2

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 5 additions & 2 deletions NeuralNetworkDetect/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ MaaFW 使用 YOLO 标准的输入输出格式，若您有 YOLO 训练经验，
如果你有一块 Nvidia GPU

```bash
# CUDA
# CUDA 11.8版本
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu118
```

Expand All @@ -32,7 +32,7 @@ pip install torch torchvision --index-url https://download.pytorch.org/whl/cu118
pip install torch torchvision
```

更多其他版本请参考 [PyTorch 官网](https://pytorch.org/get-started/locally/)。
更多其他版本请参考 [PyTorch 官网](https://pytorch.org/get-started/locally/)，特别是50系之后的Blackwell架构，至少使用12.8版本。

finally, 安装其他依赖：

Expand Down Expand Up @@ -111,8 +111,11 @@ yolo detect train data=./dataset/data.yaml model=yolo11n.pt epochs=500 imgsz=640
**注意：模型规模越大，最后导出的 onnx 模型也越大，其大小约为预训练模型的 2 ~ 3 倍**

- epochs: 训练轮次，可以适当调大些

- patience: 在多少轮训练之后，如果mA等指标无明显提升则提前中止训练
Copy link
Copy Markdown

@sourcery-ai sourcery-ai Bot Jan 2, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

issue (typo): “mA” 很可能是 “mAP” 的拼写错误，并且“等”前面应该加一个空格。

可以考虑改成 如果 mAP 等指标无明显提升则提前中止训练，以修正度量名称并补上“等”前面缺失的空格。

Suggested change
- patience: 在多少轮训练之后，如果mA等指标无明显提升则提前中止训练
- patience: 在多少轮训练之后，如果 mAP 等指标无明显提升则提前中止训练
Original comment in English

issue (typo): “mA” is likely a typo for “mAP” and could use a space before “等”.

Consider updating this to 如果 mAP 等指标无明显提升则提前中止训练 to correct the metric name and add the missing space before “等”。

Suggested change
- patience: 在多少轮训练之后，如果mA等指标无明显提升则提前中止训练
- patience: 在多少轮训练之后，如果 mAP 等指标无明显提升则提前中止训练


- cos_lr: 使用余弦学习率调度器，可以有效提高模型的收敛效果

- batch: 每次加载多少图片到显卡中，填写小数会根据显存自动决定，如 0.8 会占用 80% 的显存。多卡训练时不能填小数

更多参数参见 [train-settings](https://docs.ultralytics.com/modes/train/#train-settings)
Expand Down