目的
機械学習の知識を得るため。YOLOv5の公式サイト：
実行環境
今回はGoogle Colaboratory環境で実行した。ハードウェア情報は以下の通りです。
GPU: Intel(R) iris(R) Xe Graphics (GPUメモリ20GB搭載)
メモリ: 40GB
セットアップ
YOLOv5_train_detect.ipynb
from google.colab import drive
drive.mount('/content/drive')
%cd ./drive/MyDrive
VOLOv5ライブラリーのインストール
YOLOv5_train_detect.ipynb
!git clone https://github.com/ultralytics/yolov5 # clone
%cd yolov5
パッケージをインストールする
YOLOv5_train_detect.ipynb
%pip install -qr requirements.txt
動作確認
YOLOv5_train_detect.ipynb
!python detect.py --source /content/drive/MyDrive/datasets/coco128/images/train2017/000000000560.jpg --conf 0.25 --img-size 640
検知結果を出力する
YOLOv5_train_detect.ipynb
import matplotlib.pyplot as plt
img = plt.imread('runs/detect/exp/000000000560.jpg')
plt.imshow(img)
plt.show()
学習用データの配置
データセットを以下のフォルダに配置します。# yolov5/dataset/
今回使用したデータセットは：https://public.roboflow.com/object-detection/mask-wearing
configファイルの修正
データ用のconfigファイルをdata/mask.yamlに作成して、以下を記述する。 特に難しい記述は不要な印象mask.yml
train: /content/drive/MyDrive/yolov5/datasets/train/images
val: /content/drive/MyDrive/yolov5/datasets/valid/images
nc: 2
names: ['mask', 'no-mask']
学習の実行
YOLOv5_train_detect.ipynb
!python train.py --img 640 --batch 16 --epochs 3 --data mask.yaml --weights yolov5s.pt --cache
実行のイメージ：
train: weights=yolov5s.pt, cfg=, data=mask.yaml, hyp=data/hyps/hyp.scratch-low.yaml, epochs=100, batch_size=16, imgsz=640, rect=False, resume=False, nosave=False, noval=False, noautoanchor=False, noplots=False, evolve=None, bucket=, cache=ram, 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: up to date with https://github.com/ultralytics/yolov5 ✅
YOLOv5 🚀 v7.0-27-g454dae1 Python-3.7.15 torch-1.12.1+cu113 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/
Overriding model.yaml nc=80 with nc=2
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.C3 [128, 128, 2]
5 -1 1 295424 models.common.Conv [128, 256, 3, 2]
6 -1 3 625152 models.common.C3 [256, 256, 3]
7 -1 1 1180672 models.common.Conv [256, 512, 3, 2]
8 -1 1 1182720 models.common.C3 [512, 512, 1]
9 -1 1 656896 models.common.SPPF [512, 512, 5]
10 -1 1 131584 models.common.Conv [512, 256, 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 361984 models.common.C3 [512, 256, 1, False]
14 -1 1 33024 models.common.Conv [256, 128, 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 90880 models.common.C3 [256, 128, 1, False]
18 -1 1 147712 models.common.Conv [128, 128, 3, 2]
19 [-1, 14] 1 0 models.common.Concat [1]
20 -1 1 296448 models.common.C3 [256, 256, 1, False]
21 -1 1 590336 models.common.Conv [256, 256, 3, 2]
22 [-1, 10] 1 0 models.common.Concat [1]
23 -1 1 1182720 models.common.C3 [512, 512, 1, False]
24 [17, 20, 23] 1 18879 models.yolo.Detect [2, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512]]
Model summary: 214 layers, 7025023 parameters, 7025023 gradients, 16.0 GFLOPs
Transferred 343/349 items from yolov5s.pt
optimizer: SGD(lr=0.01) with parameter groups 57 weight(decay=0.0), 60 weight(decay=0.0005), 60 bias
albumentations: Blur(p=0.01, blur_limit=(3, 7)), MedianBlur(p=0.01, blur_limit=(3, 7)), ToGray(p=0.01), CLAHE(p=0.01, clip_limit=(1, 4.0), tile_grid_size=(8, 8))
train: Scanning /content/drive/MyDrive/yolov5/datasets/train/labels.cache... 105 images, 0 backgrounds, 0 corrupt: 100% 105/105 [00:00<?, ?it/s]
train: Caching images (0.1GB ram): 100% 105/105 [00:02<00:00, 45.44it/s]
val: Scanning /content/drive/MyDrive/yolov5/datasets/valid/labels.cache... 29 images, 0 backgrounds, 0 corrupt: 100% 29/29 [00:00<?, ?it/s]
val: Caching images (0.0GB ram): 100% 29/29 [00:01<00:00, 28.73it/s]
AutoAnchor: 5.89 anchors/target, 0.999 Best Possible Recall (BPR). Current anchors are a good fit to dataset ✅
Plotting labels to runs/train/exp5/labels.jpg...
Image sizes 640 train, 640 val
Using 2 dataloader workers
Logging results to runs/train/exp5
Starting training for 100 epochs...
requirements: YOLOv5 requirement "gitpython" not found, attempting AutoUpdate...
Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/
Collecting gitpython
Downloading GitPython-3.1.29-py3-none-any.whl (182 kB)
Collecting gitdb<5,>=4.0.1
Downloading gitdb-4.0.10-py3-none-any.whl (62 kB)
Collecting smmap<6,>=3.0.1
Downloading smmap-5.0.0-py3-none-any.whl (24 kB)
Installing collected packages: smmap, gitdb, gitpython
Successfully installed gitdb-4.0.10 gitpython-3.1.29 smmap-5.0.0
requirements: 1 package updated per ['gitpython']
requirements: ⚠️ Restart runtime or rerun command for updates to take effect
train: weights=yolov5s.pt, cfg=, data=/content/drive/MyDrive/yolov5/data/mask.yaml, hyp=data/hyps/hyp.scratch-low.yaml, epochs=100, batch_size=16, imgsz=640, rect=False, resume=False, nosave=False, noval=False, noautoanchor=False, noplots=False, evolve=None, bucket=, cache=ram, 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
Command 'git fetch origin' timed out after 5 seconds
remote: Enumerating objects: 3, done.
remote: Counting objects: 100% (3/3), done.
remote: Compressing objects: 100% (2/2), done.
remote: Total 3 (delta 1), reused 1 (delta 1), pack-reused 0
requirements: YOLOv5 requirements "ipython" "thop>=0.1.1" not found, attempting AutoUpdate...
Unpacking objects: 100% (3/3), done.
From https://github.com/ultralytics/yolov5
443ef7f..357cde9 master -> origin/master
Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/
Requirement already satisfied: ipython in /usr/local/lib/python3.8/dist-packages (7.9.0)
Collecting thop>=0.1.1
Downloading thop-0.1.1.post2209072238-py3-none-any.whl (15 kB)
Requirement already satisfied: torch in /usr/local/lib/python3.8/dist-packages (from thop>=0.1.1) (1.13.0+cu116)
Requirement already satisfied: traitlets>=4.2 in /usr/local/lib/python3.8/dist-packages (from ipython) (5.6.0)
Requirement already satisfied: backcall in /usr/local/lib/python3.8/dist-packages (from ipython) (0.2.0)
Collecting jedi>=0.10
Downloading jedi-0.18.2-py2.py3-none-any.whl (1.6 MB)
Requirement already satisfied: setuptools>=18.5 in /usr/local/lib/python3.8/dist-packages (from ipython) (57.4.0)
Requirement already satisfied: prompt-toolkit<2.1.0,>=2.0.0 in /usr/local/lib/python3.8/dist-packages (from ipython) (2.0.10)
Requirement already satisfied: decorator in /usr/local/lib/python3.8/dist-packages (from ipython) (4.4.2)
Requirement already satisfied: pygments in /usr/local/lib/python3.8/dist-packages (from ipython) (2.6.1)
Requirement already satisfied: pexpect in /usr/local/lib/python3.8/dist-packages (from ipython) (4.8.0)
Requirement already satisfied: pickleshare in /usr/local/lib/python3.8/dist-packages (from ipython) (0.7.5)
Requirement already satisfied: parso<0.9.0,>=0.8.0 in /usr/local/lib/python3.8/dist-packages (from jedi>=0.10->ipython) (0.8.3)
Requirement already satisfied: six>=1.9.0 in /usr/local/lib/python3.8/dist-packages (from prompt-toolkit<2.1.0,>=2.0.0->ipython) (1.15.0)
Requirement already satisfied: wcwidth in /usr/local/lib/python3.8/dist-packages (from prompt-toolkit<2.1.0,>=2.0.0->ipython) (0.2.5)
Requirement already satisfied: ptyprocess>=0.5 in /usr/local/lib/python3.8/dist-packages (from pexpect->ipython) (0.7.0)
Requirement already satisfied: typing-extensions in /usr/local/lib/python3.8/dist-packages (from torch->thop>=0.1.1) (4.4.0)
Installing collected packages: jedi, thop
Successfully installed jedi-0.18.2 thop-0.1.1.post2209072238
requirements: 2 packages updated per /content/drive/MyDrive/yolov5/requirements.txt
requirements: ⚠️ Restart runtime or rerun command for updates to take effect
YOLOv5 🚀 v7.0-24-gf8539a6 Python-3.8.16 torch-1.13.0+cu116 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/
Overriding model.yaml nc=80 with nc=2
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.C3 [128, 128, 2]
5 -1 1 295424 models.common.Conv [128, 256, 3, 2]
6 -1 3 625152 models.common.C3 [256, 256, 3]
7 -1 1 1180672 models.common.Conv [256, 512, 3, 2]
8 -1 1 1182720 models.common.C3 [512, 512, 1]
9 -1 1 656896 models.common.SPPF [512, 512, 5]
10 -1 1 131584 models.common.Conv [512, 256, 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 361984 models.common.C3 [512, 256, 1, False]
14 -1 1 33024 models.common.Conv [256, 128, 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 90880 models.common.C3 [256, 128, 1, False]
18 -1 1 147712 models.common.Conv [128, 128, 3, 2]
19 [-1, 14] 1 0 models.common.Concat [1]
20 -1 1 296448 models.common.C3 [256, 256, 1, False]
21 -1 1 590336 models.common.Conv [256, 256, 3, 2]
22 [-1, 10] 1 0 models.common.Concat [1]
23 -1 1 1182720 models.common.C3 [512, 512, 1, False]
24 [17, 20, 23] 1 18879 models.yolo.Detect [2, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512]]
Model summary: 214 layers, 7025023 parameters, 7025023 gradients
Transferred 343/349 items from yolov5s.pt
optimizer: SGD(lr=0.01) with parameter groups 57 weight(decay=0.0), 60 weight(decay=0.0005), 60 bias
albumentations: Blur(p=0.01, blur_limit=(3, 7)), MedianBlur(p=0.01, blur_limit=(3, 7)), ToGray(p=0.01), CLAHE(p=0.01, clip_limit=(1, 4.0), tile_grid_size=(8, 8))
train: Scanning /content/drive/MyDrive/yolov5/datasets/train/labels.cache... 105 images, 0 backgrounds, 0 corrupt: 100% 105/105 [00:00<?, ?it/s]
train: Caching images (0.1GB ram): 100% 105/105 [00:02<00:00, 46.77it/s]
val: Scanning /content/drive/MyDrive/yolov5/datasets/valid/labels.cache... 29 images, 0 backgrounds, 0 corrupt: 100% 29/29 [00:00<?, ?it/s]
val: Caching images (0.0GB ram): 100% 29/29 [00:03<00:00, 9.37it/s]
AutoAnchor: 5.89 anchors/target, 0.999 Best Possible Recall (BPR). Current anchors are a good fit to dataset ✅
Plotting labels to runs/train/exp19/labels.jpg...
Image sizes 640 train, 640 val
Using 2 dataloader workers
Logging results to runs/train/exp19
Starting training for 100 epochs...
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
0/99 0G 0.1213 0.06643 0.02934 50 640: 100% 7/7 [02:55<00:00, 25.10s/it]
Class Images Instances P R mAP50 mAP50-95: 0% 0/1 [00:00<?, ?it/s]WARNING ⚠️ NMS time limit 1.950s exceeded
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:16<00:00, 16.34s/it]
all 29 162 0.00209 0.0106 0.00107 0.000142
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
1/99 0G 0.1127 0.07942 0.02828 83 640: 100% 7/7 [02:42<00:00, 23.15s/it]
Class Images Instances P R mAP50 mAP50-95: 0% 0/1 [00:00<?, ?it/s]WARNING ⚠️ NMS time limit 1.950s exceeded
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:16<00:00, 16.54s/it]
all 29 162 0.00162 0.032 0.000846 0.000169
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
2/99 0G 0.1019 0.0914 0.02684 112 640: 100% 7/7 [02:44<00:00, 23.55s/it]
Class Images Instances P R mAP50 mAP50-95: 0% 0/1 [00:00<?, ?it/s]WARNING ⚠️ NMS time limit 1.950s exceeded
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:16<00:00, 16.56s/it]
all 29 162 0.00543 0.0458 0.00388 0.00145
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
3/99 0G 0.09635 0.0999 0.02387 81 640: 100% 7/7 [02:40<00:00, 22.94s/it]
Class Images Instances P R mAP50 mAP50-95: 0% 0/1 [00:00<?, ?it/s]WARNING ⚠️ NMS time limit 1.950s exceeded
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:16<00:00, 16.71s/it]
all 29 162 0.592 0.0282 0.0265 0.00832
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
4/99 0G 0.08721 0.09012 0.02096 106 640: 100% 7/7 [02:43<00:00, 23.40s/it]
Class Images Instances P R mAP50 mAP50-95: 0% 0/1 [00:00<?, ?it/s]WARNING ⚠️ NMS time limit 1.950s exceeded
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:16<00:00, 16.03s/it]
all 29 162 0.557 0.0915 0.0287 0.0105
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
5/99 0G 0.08064 0.07276 0.01982 72 640: 100% 7/7 [02:39<00:00, 22.75s/it]
Class Images Instances P R mAP50 mAP50-95: 0% 0/1 [00:00<?, ?it/s]WARNING ⚠️ NMS time limit 1.950s exceeded
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:15<00:00, 15.71s/it]
all 29 162 0.611 0.081 0.0514 0.018
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
6/99 0G 0.07785 0.07415 0.01932 129 640: 100% 7/7 [02:37<00:00, 22.49s/it]
Class Images Instances P R mAP50 mAP50-95: 0% 0/1 [00:00<?, ?it/s]WARNING ⚠️ NMS time limit 1.950s exceeded
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:15<00:00, 15.98s/it]
all 29 162 0.689 0.141 0.126 0.0377
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
7/99 0G 0.07889 0.06491 0.01934 52 640: 100% 7/7 [02:37<00:00, 22.48s/it]
Class Images Instances P R mAP50 mAP50-95: 0% 0/1 [00:00<?, ?it/s]WARNING ⚠️ NMS time limit 1.950s exceeded
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:15<00:00, 15.87s/it]
all 29 162 0.607 0.187 0.0846 0.0313
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
8/99 0G 0.07932 0.06554 0.01618 98 640: 100% 7/7 [02:38<00:00, 22.60s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:15<00:00, 15.21s/it]
all 29 162 0.621 0.254 0.168 0.0579
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
9/99 0G 0.0745 0.06559 0.01803 101 640: 100% 7/7 [02:41<00:00, 23.08s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:15<00:00, 15.33s/it]
all 29 162 0.619 0.236 0.191 0.0693
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
10/99 0G 0.07774 0.05583 0.01737 80 640: 100% 7/7 [02:36<00:00, 22.38s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:14<00:00, 14.73s/it]
all 29 162 0.623 0.218 0.189 0.0568
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
11/99 0G 0.07289 0.06256 0.01686 88 640: 100% 7/7 [02:37<00:00, 22.50s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:14<00:00, 14.57s/it]
all 29 162 0.672 0.278 0.259 0.0795
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
12/99 0G 0.0782 0.05958 0.01681 108 640: 100% 7/7 [02:39<00:00, 22.85s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:14<00:00, 14.88s/it]
all 29 162 0.665 0.232 0.225 0.0865
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
13/99 0G 0.07574 0.06196 0.01782 76 640: 100% 7/7 [02:36<00:00, 22.35s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:15<00:00, 15.74s/it]
all 29 162 0.153 0.583 0.179 0.0587
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
14/99 0G 0.0815 0.05924 0.0177 151 640: 100% 7/7 [02:38<00:00, 22.60s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:15<00:00, 15.97s/it]
all 29 162 0.639 0.229 0.205 0.0822
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
15/99 0G 0.07427 0.0546 0.01764 98 640: 100% 7/7 [02:35<00:00, 22.25s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:14<00:00, 14.25s/it]
all 29 162 0.669 0.32 0.244 0.0895
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
16/99 0G 0.07037 0.05111 0.01617 130 640: 100% 7/7 [02:44<00:00, 23.55s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:14<00:00, 14.34s/it]
all 29 162 0.655 0.268 0.231 0.0599
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
17/99 0G 0.06646 0.05831 0.01645 203 640: 100% 7/7 [02:39<00:00, 22.73s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:13<00:00, 13.91s/it]
all 29 162 0.68 0.254 0.342 0.129
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
18/99 0G 0.06685 0.05112 0.0178 124 640: 100% 7/7 [02:37<00:00, 22.53s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:14<00:00, 14.27s/it]
all 29 162 0.695 0.278 0.272 0.105
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
19/99 0G 0.0637 0.05462 0.01597 62 640: 100% 7/7 [02:41<00:00, 23.09s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:14<00:00, 14.33s/it]
all 29 162 0.166 0.547 0.234 0.0909
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
20/99 0G 0.06474 0.05678 0.01562 110 640: 100% 7/7 [02:34<00:00, 22.14s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:13<00:00, 13.71s/it]
all 29 162 0.328 0.535 0.385 0.151
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
21/99 0G 0.05212 0.05954 0.01771 102 640: 100% 7/7 [02:33<00:00, 21.98s/it]
Class Images Instances P R mAP50 mAP50-95: 100% 1/1 [00:13<00:00, 13.61s/it]
all 29 162 0.195 0.586 0.26 0.107
Epoch GPU_mem box_loss obj_loss cls_loss Instances Size
22/99 0G 0.06745 0.04546 0.01458 146 640: 14% 1/7 [00:23<02:21, 23.66s/it]
・・・・・・
推論テスト
YOLOv5_train_detect.ipynb
!python detect.py --weights /content/yolov5/runs/train/exp4/weights/best.pt \
--source /content/drive/MyDrive/IMG_8585.JPG \
--conf 0.25 --img-size 640
実行のイメージ：
detect: weights=['/content/yolov5/runs/train/exp5/weights/best.pt'], source=/content/drive/MyDrive/IMG_8585.JPG, data=data/coco128.yaml, imgsz=[640, 640], conf_thres=0.25, iou_thres=0.45, max_det=1000, device=, view_img=False, save_txt=False, save_conf=False, save_crop=False, nosave=False, classes=None, agnostic_nms=False, augment=False, visualize=False, update=False, project=runs/detect, name=exp, exist_ok=False, line_thickness=3, hide_labels=False, hide_conf=False, half=False, dnn=False, vid_stride=1
YOLOv5 🚀 v7.0-27-g454dae1 Python-3.7.15 torch-1.12.1+cu113 CPU
Fusing layers...
Model summary: 157 layers, 7015519 parameters, 0 gradients, 15.8 GFLOPs
image 1/1 /content/drive/MyDrive/IMG_8585.JPG: 480x640 (no detections), 297.6ms
Speed: 1.3ms pre-process, 297.6ms inference, 0.5ms NMS per image at shape (1, 3, 640, 640)
Results saved to runs/detect/exp5
学習結果の出力
import matplotlib.pyplot as plt
img = plt.imread('runs/detect/exp5/IMG_8585.JPG')
plt.imshow(img)
plt.show()
課題
VOLOv7を利用したかったが、GPUの条件が満たさなかったみたいので、CUDA無化になった。