YOLOv5白皮书-第Y6周：模型改进

一、课题背景和开发环境

📌第Y6周：模型改进📌

• 语言：Python3、Pytorch
• 📌本周任务：📌 本周任务：我修改了YOLOv5s的网络结构图，请根据网络结构图以及第Y1～Y5周的内容修改对应代码，并跑通程序。
  
  

开发环境

• 电脑系统：Windows 10
• 语言环境：Python 3.8.2
• 编译器：无（直接在cmd.exe内运行）
• 深度学习环境：Pytorch 1.8.1+cu111
• 显卡及显存：NVIDIA GeForce GTX 1660 Ti 12G
• CUDA版本：Release 10.2, V10.2.89（cmd输入nvcc -V或nvcc --version指令可查看）
• YOLOv5开源地址：YOLOv5开源地址
• 数据：🔗水果检测

---

二、调整模型

# 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]],            # 2
   [-1, 1, Conv, [256, 3, 2]],    # 3-P3/8
   [-1, 6, C2, [256]],            # 4-修改为C2*2
   [-1, 1, Conv, [512, 3, 2]],    # 5-P4/16
   [-1, 3, C3, [512]],            # 6-修改为C3*1
#   [-1, 1, Conv, [1024, 3, 2]],   # 7-删除P5/32
#   [-1, 3, C3, [1024]],           # 8-删除
   [-1, 1, SPPF, [512, 5]],      # 9-修改参数;层数变为7
  ]

# YOLOv5 v6.0 head
head:
  [[-1, 1, Conv, [512, 3, 2]], # 修改参数
   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
   [[-1, 6], 1, Concat, [1]],  # cat backbone P4
   [-1, 3, C3, [512, False]],  # 13->11

   [-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->15 (P3/8-small)

   [-1, 1, Conv, [256, 3, 2]],
   [[-1, 12], 1, Concat, [1]],  # cat head P4 修改层数-2
   [-1, 3, C3, [512, False]],  # 20->18 (P4/16-medium)

   [-1, 1, Conv, [512, 3, 2]],
   [[-1, 8], 1, Concat, [1]],  # cat head P5 修改层数-2
   [-1, 3, C3, [1024, False]],  # 23->21 (P5/32-large)

   [[15, 18, 21], 1, Detect, [nc, anchors]],  # Detect(P3, P4, P5) 修改层数-2
  ]

三、运行&打印模型查看

python train.py --img 640 --batch 8 --epoch 1 --data data/fruits.yaml --cfg models/yolov5s.yaml --weights weights/yolov5s.pt --device 0

(Pytorch) E:\WorkSpace_GuanXiang\0.学习资料\365天深度学习训练营\2.YOLOv5白皮书\yolov5-master>python train.py --img 640 --batch 8 --epoch 1 --data data/fruits.yaml --cfg models/yolov5s.yaml --weights weights/yolov5s.pt --device 0
train: weights=weights/yolov5s.pt, cfg=models/yolov5s.yaml, data=data/fruits.yaml, hyp=data\hyps\hyp.scratch-low.yaml, epochs=1, batch_size=8, imgsz=640, rect=False, resume=False, nosave=False, noval=False, noautoanchor=False, noplots=False, evolve=None, bucket=, cache=None, image_weights=False, device=0, 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  2022-12-8 Python-3.8.12 torch-1.8.1+cu111 CUDA:0 (NVIDIA GeForce GTX 1660 Ti, 6144MiB)

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/
Overriding model.yaml nc=80 with nc=4

                 from  n    params  module                                  arguments
  0                -1  1      3520  models.common.Conv                      [3, 32, 6, 2, 2]
  1                -1  1     18560  models.common.Conv                      [32, 64, 3, 2]
  2                -1  1     18816  models.common.C3                        [64, 64, 1]
  3                -1  1     73984  models.common.Conv                      [64, 128, 3, 2]
  4                -1  2    115712  models.common.C2                        [128, 128, 2]
  5                -1  1    295424  models.common.Conv                      [128, 256, 3, 2]
  6                -1  1    296448  models.common.C3                        [256, 256, 1]
  7                -1  1    164608  models.common.SPPF                      [256, 256, 5]
  8                -1  1    590336  models.common.Conv                      [256, 256, 3, 2]
  9                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']
 10           [-1, 6]  1         0  models.common.Concat                    [1]
 11                -1  1    361984  models.common.C3                        [512, 256, 1, False]
 12                -1  1     33024  models.common.Conv                      [256, 128, 1, 1]
 13                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']
 14           [-1, 4]  1         0  models.common.Concat                    [1]
 15                -1  1     90880  models.common.C3                        [256, 128, 1, False]
 16                -1  1    147712  models.common.Conv                      [128, 128, 3, 2]
 17          [-1, 12]  1         0  models.common.Concat                    [1]
 18                -1  1    296448  models.common.C3                        [256, 256, 1, False]
 19                -1  1    590336  models.common.Conv                      [256, 256, 3, 2]
 20           [-1, 8]  1         0  models.common.Concat                    [1]
 21                -1  1   1182720  models.common.C3                        [512, 512, 1, False]
 22      [15, 18, 21]  1     24273  models.yolo.Detect                      [4, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512]]
YOLOv5s summary: 179 layers, 4304785 parameters, 4304785 gradients, 13.2 GFLOPs

Transferred 126/289 items from weights\yolov5s.pt
AMP: checks passed
optimizer: SGD(lr=0.01) with parameter groups 47 weight(decay=0.0), 50 weight(decay=0.0005), 50 bias
train: Scanning E:\WorkSpace_GuanXiang\0.学习资料\365天深度学习训练营\2.YOLOv5白皮书\yolov5-master\paper_data\train...
train: WARNING  Cache directory E:\WorkSpace_GuanXiang\0.\365\2.YOLOv5\yolov5-master\paper_data is not writeable: [WinError 183] : 'E:\\WorkSpace_GuanXiang\\0.\\365\\2.YOLOv5\\yolov5-master\\paper_data\\train.cache.npy' -> 'E:\\WorkSpace_GuanXiang\\0.\\365\\2.YOLOv5\\yolov5-master\\paper_data\\train.cache'
val: Scanning E:\WorkSpace_GuanXiang\0.学习资料\365天深度学习训练营\2.YOLOv5白皮书\yolov5-master\paper_data\val.cache..

AutoAnchor: 5.35 anchors/target, 1.000 Best Possible Recall (BPR). Current anchors are a good fit to dataset
Plotting labels to runs\train\exp4\labels.jpg...
Image sizes 640 train, 640 val
Using 8 dataloader workers
Logging results to runs\train\exp4
Starting training for 1 epochs...

      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size
        0/0      1.71G        nan        nan        nan         42        640: 100%|██████████| 20/20 00:13
                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100%|██████████| 2/2 00:00
                   all         20         60          0          0          0          0

1 epochs completed in 0.004 hours.
Optimizer stripped from runs\train\exp4\weights\last.pt, 8.9MB
Optimizer stripped from runs\train\exp4\weights\best.pt, 8.9MB

Validating runs\train\exp4\weights\best.pt...
Fusing layers...
YOLOv5s summary: 132 layers, 4298225 parameters, 0 gradients, 13.0 GFLOPs
                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100%|██████████| 2/2 00:00
                   all         20         60          0          0          0          0
Results saved to runs\train\exp4