EADST

YOLOv5: Train the Model

YOLOv5: Train the Model

Download YOLOv5 link

Create a yaml file under ./data/our_data.yaml, change the image path, class number, and class names


# train and val data as 1) directory: path/images/, 2) file: path/images.txt, or 3) list: [path1/images/, path2/images/]
train: /dfs/data/others/byolov5/dataset/yolo_data/train/images
val: /dfs/data/others/byolov5/dataset/yolo_data/val/images

# number of classes
nc: 2

# class names
names: ['b', 't']

Download YOLOv5s model link and put it to ./weights.

Create a yaml file under ./models/our_model.yaml from yolov5s.yaml, change number of classes (nc)

YOLOv5 🚀 by Ultralytics, GPL-3.0 license

Parameters

nc: 2 # number of classes depth_multiple: 0.33 # model depth multiple width_multiple: 0.50 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32

YOLOv5 v6.0 backbone

backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 9 ]

YOLOv5 v6.0 head

head: [[-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 13

[-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 17 (P3/8-small)

[-1, 1, Conv, [256, 3, 2]], [[-1, 14], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 20 (P4/16-medium)

[-1, 1, Conv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [1024, False]], # 23 (P5/32-large)

[[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ]

Run the following command to train the model

python train.py --data data/our_data.yaml --cfg models/our_model.yaml  --weights weights/yolov5s.pt --device 0

Reference:

yolov5训练自己的VOC数据集

相关标签
About Me
XD
Goals determine what you are going to be.
Category
标签云
Conda Use GIT Translation Shortcut git-lfs Food Algorithm 报税 Qwen2.5 Jetson Magnet Plotly git Bin 论文 MD5 PyCharm Animate C++ Hilton Review SQL BeautifulSoup NLTK Tensor BF16 SQLite TensorRT Video logger Quantization Web Password Google Pandas Zip Claude ms-swift uWSGI WebCrawler VSCode Jupyter 长期记忆 GPT4 Logo UI OCR LLM Datetime LaTeX Gemma Diagram Attention QWEN Safetensors Vmess 证件照 Augmentation LoRA Harness XML Windows diffusers ModelScope Jev hf IndexTTS2 RL JSON transformers FP32 News 继承 COCO Markdown Image2Text Excel ChatGPT 阿里云 域名 Permission 关于博主 FP8 Search 版权 Breakpoint Nginx Tracking NameSilo 个人博客 v2ray Knowledge PDB Git 顶会 第一性原理 GPTQ TTS Baidu Transformers Firewall 飞书 uwsgi DeepStream 签证 Base64 v0.dev LeetCode Land CSV SVR 净利润 算法题 Distillation SPIE CTC printf BTC DeepSeek 云服务器 递归学习法 Qwen2 Domain PDF VPN TensorFlow ResNet-50 RAR Anaconda Disk InvalidArgumentError Website icon YOLO NLP mmap 公式 强化学习 Card 自我改进 WAN 多线程 AI Color PyTorch Docker Vim Interview Pillow 图形思考法 LLAMA FP64 VGG-16 Statistics CC scipy Michelin TSV torchinfo CEIR SAM 音频 图标 Plate Qwen 搞笑 HaggingFace GoogLeNet Sklearn FlashAttention Ubuntu Pytorch UNIX API网关 多进程 llama.cpp OpenCV Django Rebuttal Input Hotel Mixtral CLAP Bitcoin 财报 Quantize Paddle Tiktoken Miniforge ONNX GGML Streamlit 论文速读 Bert HuggingFace Github Bipartite XGBoost Hungarian API FastAPI Template EXCEL PIP Data Freesound tqdm CAM Cloudreve tar Pickle Linux FP16 Math CUDA Paper Clash OpenAI Random Dataset Python Heatmap Crawler CV 腾讯云 Ptyhon Agent Llama Numpy Proxy RGB
站点统计

本站现有博文340篇,共被浏览1019175次

本站已经建立2719天!

热门文章
文章归档
回到顶部