yolov5 用自己的数据集进行训练

在训练之前先要按照一定目录格式准备数据：

VOC标签格式转yolo格式并划分训练集和测试集_爱钓鱼的歪猴的博客-CSDN博客

1、修改数据配置文件

2、修改模型配置文件

3、训练

1、修改数据配置文件

coco.yaml

拷贝data/scripts/coco.yaml文件，

path 修改为VOCdevkit文件夹所在目录

train：修改为yolov5_train.txt

val：修改为yolov5_val.txt

names也进行修改，保存为my-anther.yaml

具体如下：

这样模仿VOC数据集的目录结构，与yolov5项目里的代码所匹配

2、修改模型配置文件

把项目models/yolov5n.yaml文件拷贝一份，只修改类别数量（num of classes）就成，保存为my-yolov5n.yaml

3、训练

在项目目录下开启终端

运行：

 python train.py  --data my-anther.yaml --epochs 1 --weights yolov5n.pt --cfg my-yolov5n.yaml --batch-size 24

如果有GPU ，命令后面添加 --device 0。这里没有GPU，所以只跑一个 epoch。

batch可选择调为16、24、40、64、128....。先选一个小的，保证能跑起来，然后慢慢加大，如果内存不足报错，就返回上一个batch大小。

输出信息：

(yolo) jason@honor:~/PycharmProjects/pytorch_learn/yolo/yolov5-7.0$ python train.py  --data my-anther.yaml --epochs 1 --weights yolov5n.pt --cfg my-yolov5n.yaml --batch-size 24
train: weights=yolov5n.pt, cfg=my-yolov5n.yaml, data=my-anther.yaml, hyp=data/hyps/hyp.scratch-low.yaml, epochs=1, batch_size=24, imgsz=640, rect=False, resume=False, nosave=False, noval=False, noautoanchor=False, noplots=False, evolve=None, bucket=, cache=None, image_weights=False, device=, multi_scale=False, single_cls=False, optimizer=SGD, sync_bn=False, workers=8, project=runs/train, name=exp, exist_ok=False, quad=False, cos_lr=False, label_smoothing=0.0, patience=100, freeze=[0], save_period=-1, seed=0, local_rank=-1, entity=None, upload_dataset=False, bbox_interval=-1, artifact_alias=latest
github: skipping check (not a git repository), for updates see https://github.com/ultralytics/yolov5
YOLOv5 🚀 2022-11-22 Python-3.8.13 torch-2.0.0+cu117 CPU

hyperparameters: lr0=0.01, lrf=0.01, momentum=0.937, weight_decay=0.0005, warmup_epochs=3.0, warmup_momentum=0.8, warmup_bias_lr=0.1, box=0.05, cls=0.5, cls_pw=1.0, obj=1.0, obj_pw=1.0, iou_t=0.2, anchor_t=4.0, fl_gamma=0.0, hsv_h=0.015, hsv_s=0.7, hsv_v=0.4, degrees=0.0, translate=0.1, scale=0.5, shear=0.0, perspective=0.0, flipud=0.0, fliplr=0.5, mosaic=1.0, mixup=0.0, copy_paste=0.0
ClearML: run 'pip install clearml' to automatically track, visualize and remotely train YOLOv5 🚀 in ClearML
Comet: run 'pip install comet_ml' to automatically track and visualize YOLOv5 🚀 runs in Comet
TensorBoard: Start with 'tensorboard --logdir runs/train', view at http://localhost:6006/

                 from  n    params  module                                  arguments                     
  0                -1  1      1760  models.common.Conv                      [3, 16, 6, 2, 2]              
  1                -1  1      4672  models.common.Conv                      [16, 32, 3, 2]                
  2                -1  1      4800  models.common.C3                        [32, 32, 1]                   
  3                -1  1     18560  models.common.Conv                      [32, 64, 3, 2]                
  4                -1  2     29184  models.common.C3                        [64, 64, 2]                   
  5                -1  1     73984  models.common.Conv                      [64, 128, 3, 2]               
  6                -1  3    156928  models.common.C3                        [128, 128, 3]                 
  7                -1  1    295424  models.common.Conv                      [128, 256, 3, 2]              
  8                -1  1    296448  models.common.C3                        [256, 256, 1]                 
  9                -1  1    164608  models.common.SPPF                      [256, 256, 5]                 
 10                -1  1     33024  models.common.Conv                      [256, 128, 1, 1]              
 11                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          
 12           [-1, 6]  1         0  models.common.Concat                    [1]                           
 13                -1  1     90880  models.common.C3                        [256, 128, 1, False]          
 14                -1  1      8320  models.common.Conv                      [128, 64, 1, 1]               
 15                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          
 16           [-1, 4]  1         0  models.common.Concat                    [1]                           
 17                -1  1     22912  models.common.C3                        [128, 64, 1, False]           
 18                -1  1     36992  models.common.Conv                      [64, 64, 3, 2]                
 19          [-1, 14]  1         0  models.common.Concat                    [1]                           
 20                -1  1     74496  models.common.C3                        [128, 128, 1, False]          
 21                -1  1    147712  models.common.Conv                      [128, 128, 3, 2]              
 22          [-1, 10]  1         0  models.common.Concat                    [1]                           
 23                -1  1    296448  models.common.C3                        [256, 256, 1, False]          
 24      [17, 20, 23]  1      9471  models.yolo.Detect                      [2, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [64, 128, 256]]
my-YOLOv5n summary: 214 layers, 1766623 parameters, 1766623 gradients, 4.2 GFLOPs

Transferred 342/349 items from yolov5n.pt
optimizer: SGD(lr=0.01) with parameter groups 57 weight(decay=0.0), 60 weight(decay=0.0005625000000000001), 60 bias
train: Scanning /home/jason/work/my-datasets/yolov5_train.cache... 2276 images, 0 backgrounds, 0 corrupt: 100%|██████████| 2276/2276 00:00
val: Scanning /home/jason/work/my-datasets/yolov5_val.cache... 568 images, 0 backgrounds, 0 corrupt: 100%|██████████| 568/568 00:00

AutoAnchor: 6.38 anchors/target, 1.000 Best Possible Recall (BPR). Current anchors are a good fit to dataset ✅
Plotting labels to runs/train/exp2/labels.jpg... 
Image sizes 640 train, 640 val
Using 8 dataloader workers
Logging results to runs/train/exp2
Starting training for 1 epochs...

      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size
        0/0         0G    0.09858     0.3112    0.01833       1331        640: 100%|██████████| 95/95 10:55
                 Class     Images  Instances          P          R      mAP50   mAP50-95:   0%|          | 0/12 00:00WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:   8%|▊         | 1/12 00:08WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  17%|█▋        | 2/12 00:15WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  25%|██▌       | 3/12 00:23WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  33%|███▎      | 4/12 00:30WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  42%|████▏     | 5/12 00:38WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  50%|█████     | 6/12 00:47WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  58%|█████▊    | 7/12 00:54WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  67%|██████▋   | 8/12 01:00WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  75%|███████▌  | 9/12 01:07WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  83%|████████▎ | 10/12 01:14WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  92%|█████████▏| 11/12 01:21WARNING ⚠️ NMS time limit 2.500s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100%|██████████| 12/12 01:28
                   all        568      28591      0.613      0.262      0.108     0.0351

1 epochs completed in 0.207 hours.
Optimizer stripped from runs/train/exp2/weights/last.pt, 3.8MB
Optimizer stripped from runs/train/exp2/weights/best.pt, 3.8MB

Validating runs/train/exp2/weights/best.pt...
Fusing layers... 
my-YOLOv5n summary: 157 layers, 1761871 parameters, 0 gradients, 4.1 GFLOPs
                 Class     Images  Instances          P          R      mAP50   mAP50-95:   0%|          | 0/12 00:00WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:   8%|▊         | 1/12 00:08WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  17%|█▋        | 2/12 00:44WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  25%|██▌       | 3/12 01:07WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  33%|███▎      | 4/12 01:33WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  42%|████▏     | 5/12 01:42WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  50%|█████     | 6/12 01:49WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  58%|█████▊    | 7/12 01:57WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  67%|██████▋   | 8/12 02:04WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  75%|███████▌  | 9/12 02:10WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  83%|████████▎ | 10/12 02:18WARNING ⚠️ NMS time limit 2.900s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95:  92%|█████████▏| 11/12 02:24WARNING ⚠️ NMS time limit 2.500s exceeded
                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100%|██████████| 12/12 02:30
                   all        568      28591       0.61      0.228     0.0949     0.0317
                  open        568       2512          1          0    0.00273    0.00136
                 close        568      26079      0.219      0.456      0.187      0.062
Results saved to runs/train/exp2