TTA
/
Detection_Experiment
18
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '56e7da7594'
${ noResults }
Detection_Experiment/custom_LP_detect/yolo_exp.ipynb

686 lines
66 KiB
Raw Normal View History Unescape

Organize my project
6 months ago
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"id": "a7676704",
"metadata": {},
"outputs": [],
"source": [
"from ultralytics import YOLO\n",
"\n",
"import torch"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "3e94066a",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"True"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"torch.cuda.is_available()"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "99b0442c",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"New https://pypi.org/project/ultralytics/8.3.228 available 😃 Update with 'pip install -U ultralytics'\n",
"Ultralytics 8.3.225 🚀 Python-3.10.18 torch-2.9.1+cu128 CUDA:0 (NVIDIA GeForce RTX 5090, 32087MiB)\n",
"\u001b[34m\u001b[1mengine/trainer: \u001b[0magnostic_nms=False, amp=True, augment=False, auto_augment=randaugment, batch=-1, bgr=0.0, box=7.5, cache=False, cfg=None, classes=None, close_mosaic=10, cls=0.5, compile=False, conf=None, copy_paste=0.0, copy_paste_mode=flip, cos_lr=False, cutmix=0.0, data=/home/cuuva/experiment/custom_LP_detect/custom_LP.yaml, degrees=0.0, deterministic=True, device=0, dfl=1.5, dnn=False, dropout=0.0, dynamic=False, embed=None, epochs=200, erasing=0.4, exist_ok=False, fliplr=0.5, flipud=0.0, format=torchscript, fraction=0.2, freeze=None, half=False, hsv_h=0.015, hsv_s=0.7, hsv_v=0.4, imgsz=640, int8=False, iou=0.7, keras=False, kobj=1.0, line_width=None, lr0=0.001, lrf=0.01, mask_ratio=4, max_det=300, mixup=0.0, mode=train, model=yolov8n.pt, momentum=0.937, mosaic=1.0, multi_scale=False, name=epo_200_frac_0_22, nbs=64, nms=False, opset=None, optimize=False, optimizer=AdamW, overlap_mask=True, patience=40, perspective=0.0, plots=True, pose=12.0, pretrained=True, profile=False, project=lp_detect, rect=False, resume=False, retina_masks=False, save=True, save_conf=False, save_crop=False, save_dir=/home/cuuva/experiment/custom_LP_detect/lp_detect/epo_200_frac_0_22, save_frames=False, save_json=False, save_period=-1, save_txt=False, scale=0.5, seed=0, shear=0.0, show=False, show_boxes=True, show_conf=True, show_labels=True, simplify=True, single_cls=False, source=None, split=val, stream_buffer=False, task=detect, time=None, tracker=botsort.yaml, translate=0.1, val=True, verbose=True, vid_stride=1, visualize=False, warmup_bias_lr=0.1, warmup_epochs=3.0, warmup_momentum=0.8, weight_decay=0.0005, workers=8, workspace=None\n",
"Overriding model.yaml nc=80 with nc=1\n",
"\n",
" from n params module arguments \n",
" 0 -1 1 464 ultralytics.nn.modules.conv.Conv [3, 16, 3, 2] \n",
" 1 -1 1 4672 ultralytics.nn.modules.conv.Conv [16, 32, 3, 2] \n",
" 2 -1 1 7360 ultralytics.nn.modules.block.C2f [32, 32, 1, True] \n",
" 3 -1 1 18560 ultralytics.nn.modules.conv.Conv [32, 64, 3, 2] \n",
" 4 -1 2 49664 ultralytics.nn.modules.block.C2f [64, 64, 2, True] \n",
" 5 -1 1 73984 ultralytics.nn.modules.conv.Conv [64, 128, 3, 2] \n",
" 6 -1 2 197632 ultralytics.nn.modules.block.C2f [128, 128, 2, True] \n",
" 7 -1 1 295424 ultralytics.nn.modules.conv.Conv [128, 256, 3, 2] \n",
" 8 -1 1 460288 ultralytics.nn.modules.block.C2f [256, 256, 1, True] \n",
" 9 -1 1 164608 ultralytics.nn.modules.block.SPPF [256, 256, 5] \n",
" 10 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest'] \n",
" 11 [-1, 6] 1 0 ultralytics.nn.modules.conv.Concat [1] \n",
" 12 -1 1 148224 ultralytics.nn.modules.block.C2f [384, 128, 1] \n",
" 13 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest'] \n",
" 14 [-1, 4] 1 0 ultralytics.nn.modules.conv.Concat [1] \n",
" 15 -1 1 37248 ultralytics.nn.modules.block.C2f [192, 64, 1] \n",
" 16 -1 1 36992 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2] \n",
" 17 [-1, 12] 1 0 ultralytics.nn.modules.conv.Concat [1] \n",
" 18 -1 1 123648 ultralytics.nn.modules.block.C2f [192, 128, 1] \n",
" 19 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2] \n",
" 20 [-1, 9] 1 0 ultralytics.nn.modules.conv.Concat [1] \n",
" 21 -1 1 493056 ultralytics.nn.modules.block.C2f [384, 256, 1] \n",
" 22 [15, 18, 21] 1 751507 ultralytics.nn.modules.head.Detect [1, [64, 128, 256]] \n",
"Model summary: 129 layers, 3,011,043 parameters, 3,011,027 gradients, 8.2 GFLOPs\n",
"\n",
"Transferred 319/355 items from pretrained weights\n",
"Freezing layer 'model.22.dfl.conv.weight'\n",
"\u001b[34m\u001b[1mAMP: \u001b[0mrunning Automatic Mixed Precision (AMP) checks...\n",
"\u001b[34m\u001b[1mAMP: \u001b[0mchecks passed ✅\n",
"\u001b[34m\u001b[1mtrain: \u001b[0mFast image access ✅ (ping: 0.0±0.0 ms, read: 1415.2±186.1 MB/s, size: 1748.6 KB)\n",
"\u001b[K\u001b[34m\u001b[1mtrain: \u001b[0mScanning /home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/labels/license_plate/ar01_01.cache... 160269 images, 0 backgrounds, 2 corrupt: 100% ━━━━━━━━━━━━ 160271/160271 514.3Mit/s 0.0s\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar01_01/C-220930_08_AR01_01_N0023.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar01_01/C-221011_13_AR01_01_N4945.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar01_01/C-221019_12_AR01_01_N5133.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar01_01/C-221019_14_AR01_01_N0089.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar03_03/C-221105_17_AR03_03_N0282.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar03_03/C-221109_14_AR03_03_N0673.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar04_04/C-221022_14_AR04_04_N0674.jpg: 2 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar09_01/C-220807_15_AR09_01_N0514.jpg: ignoring corrupt image/label: image file is truncated (107 bytes not processed)\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/cr01_02/C-220827_16_CR01_02_N0098.jpg: ignoring corrupt image/label: image file is truncated (96 bytes not processed)\n",
"\u001b[34m\u001b[1malbumentations: \u001b[0mBlur(p=0.01, blur_limit=(3, 7)), MedianBlur(p=0.01, blur_limit=(3, 7)), ToGray(p=0.01, method='weighted_average', num_output_channels=3), CLAHE(p=0.01, clip_limit=(1.0, 4.0), tile_grid_size=(8, 8))\n",
"\u001b[34m\u001b[1mAutoBatch: \u001b[0mComputing optimal batch size for imgsz=640 at 60.0% CUDA memory utilization.\n",
"\u001b[34m\u001b[1mAutoBatch: \u001b[0mCUDA:0 (NVIDIA GeForce RTX 5090) 31.33G total, 0.15G reserved, 0.05G allocated, 31.13G free\n",
" Params GFLOPs GPU_mem (GB) forward (ms) backward (ms) input output\n",
" 3011043 8.194 3.456 31.09 193.1 (1, 3, 640, 640) list\n",
" 3011043 16.39 3.997 4.507 23.43 (2, 3, 640, 640) list\n",
" 3011043 32.78 4.270 5.17 24.49 (4, 3, 640, 640) list\n",
" 3011043 65.55 4.979 5.493 31.44 (8, 3, 640, 640) list\n",
" 3011043 131.1 6.201 7.781 41.38 (16, 3, 640, 640) list\n",
" 3011043 262.2 5.014 15.01 55.64 (32, 3, 640, 640) list\n",
" 3011043 524.4 10.863 31.73 105.9 (64, 3, 640, 640) list\n",
"\u001b[34m\u001b[1mAutoBatch: \u001b[0mUsing batch-size 135 for CUDA:0 19.16G/31.33G (61%) ✅\n",
"\u001b[34m\u001b[1mtrain: \u001b[0mFast image access ✅ (ping: 0.0±0.0 ms, read: 1182.9±337.0 MB/s, size: 1333.9 KB)\n",
"\u001b[K\u001b[34m\u001b[1mtrain: \u001b[0mScanning /home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/labels/license_plate/ar01_01.cache... 160269 images, 0 backgrounds, 2 corrupt: 100% ━━━━━━━━━━━━ 160271/160271 317.1Mit/s 0.0s\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar01_01/C-220930_08_AR01_01_N0023.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar01_01/C-221011_13_AR01_01_N4945.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar01_01/C-221019_12_AR01_01_N5133.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar01_01/C-221019_14_AR01_01_N0089.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar03_03/C-221105_17_AR03_03_N0282.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar03_03/C-221109_14_AR03_03_N0673.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar04_04/C-221022_14_AR04_04_N0674.jpg: 2 duplicate labels removed\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/ar09_01/C-220807_15_AR09_01_N0514.jpg: ignoring corrupt image/label: image file is truncated (107 bytes not processed)\n",
"\u001b[34m\u001b[1mtrain: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Training/images/license_plate/cr01_02/C-220827_16_CR01_02_N0098.jpg: ignoring corrupt image/label: image file is truncated (96 bytes not processed)\n",
"\u001b[34m\u001b[1malbumentations: \u001b[0mBlur(p=0.01, blur_limit=(3, 7)), MedianBlur(p=0.01, blur_limit=(3, 7)), ToGray(p=0.01, method='weighted_average', num_output_channels=3), CLAHE(p=0.01, clip_limit=(1.0, 4.0), tile_grid_size=(8, 8))\n",
"\u001b[34m\u001b[1mval: \u001b[0mFast image access ✅ (ping: 0.0±0.0 ms, read: 1014.6±261.7 MB/s, size: 1776.7 KB)\n",
"\u001b[K\u001b[34m\u001b[1mval: \u001b[0mScanning /home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Validation/labels/license_plate/ar01_01.cache... 52168 images, 0 backgrounds, 0 corrupt: 100% ━━━━━━━━━━━━ 52168/52168 98.1Mit/s 0.0s\n",
"\u001b[34m\u001b[1mval: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Validation/images/license_plate/ar01_01/C-221030_13_AR01_01_N0451.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mval: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Validation/images/license_plate/sr02_01/C-220921_13_SR02_01_N0664.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mval: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Validation/images/license_plate/sr08_01/C-221026_17_SR08_01_N0368.jpg: 3 duplicate labels removed\n",
"\u001b[34m\u001b[1mval: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Validation/images/license_plate/sr10_01/C-221111_14_SR10_01_N2786.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mval: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Validation/images/license_plate/sr10_01/C-221112_10_SR10_01_N3102.jpg: 1 duplicate labels removed\n",
"\u001b[34m\u001b[1mval: \u001b[0m/home/cuuva/aihub_car/CCTV_car_or_licenseplate/data/Validation/images/license_plate/sr10_01/C-221112_13_SR10_01_N1269.jpg: 1 duplicate labels removed\n",
"Plotting labels to /home/cuuva/experiment/custom_LP_detect/lp_detect/epo_200_frac_0_22/labels.jpg... \n",
"\u001b[34m\u001b[1moptimizer:\u001b[0m AdamW(lr=0.001, momentum=0.937) with parameter groups 57 weight(decay=0.0), 64 weight(decay=0.0010546875), 63 bias(decay=0.0)\n",
"Image sizes 640 train, 640 val\n",
"Using 8 dataloader workers\n",
"Logging results to \u001b[1m/home/cuuva/experiment/custom_LP_detect/lp_detect/epo_200_frac_0_22\u001b[0m\n",
"Starting training for 200 epochs...\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 1/200 18.1G 2.084 2.448 0.8108 125 640: 100% ━━━━━━━━━━━━ 1188/1188 2.1it/s 9:14<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:340.7sss\n",
" all 52168 159183 0.719 0.475 0.53 0.234\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 2/200 18G 1.587 0.7441 0.7597 115 640: 100% ━━━━━━━━━━━━ 1188/1188 2.2it/s 8:56<0.7s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.7sss\n",
" all 52168 159183 0.817 0.527 0.613 0.329\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 3/200 18G 1.474 0.6842 0.7541 108 640: 100% ━━━━━━━━━━━━ 1188/1188 1.7it/s 11:43<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.9it/s 1:420.7sss\n",
" all 52168 159183 0.809 0.533 0.608 0.326\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 4/200 18G 1.389 0.6393 0.7508 123 640: 100% ━━━━━━━━━━━━ 1188/1188 1.7it/s 11:40<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:370.6sss\n",
" all 52168 159183 0.836 0.536 0.647 0.363\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 5/200 18G 1.327 0.6079 0.7478 100 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 12:49<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.9it/s 1:420.5sss\n",
" all 52168 159183 0.846 0.564 0.674 0.391\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 6/200 18G 1.289 0.5872 0.7467 130 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:20<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.2it/s 1:300.6sss\n",
" all 52168 159183 0.843 0.574 0.68 0.404\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 7/200 18G 1.26 0.5727 0.7452 133 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:09<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:340.6sss\n",
" all 52168 159183 0.848 0.57 0.674 0.388\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 8/200 18G 1.239 0.5616 0.7449 98 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 12:50<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:340.7sss\n",
" all 52168 159183 0.843 0.579 0.681 0.392\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 9/200 18G 1.224 0.5547 0.7435 119 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 12:51<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:380.6sss\n",
" all 52168 159183 0.843 0.583 0.685 0.396\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 10/200 18G 1.208 0.5464 0.7427 107 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:12<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:340.7sss\n",
" all 52168 159183 0.853 0.582 0.689 0.4\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 11/200 18G 1.196 0.5408 0.7427 111 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:22<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.9it/s 1:410.7sss\n",
" all 52168 159183 0.848 0.582 0.689 0.404\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 12/200 18G 1.187 0.5344 0.7425 104 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:36<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:380.6sss\n",
" all 52168 159183 0.849 0.582 0.691 0.41\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 13/200 18G 1.179 0.5308 0.7421 122 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:55<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:380.7sss\n",
" all 52168 159183 0.847 0.584 0.693 0.409\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 14/200 18G 1.168 0.527 0.7418 111 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:56<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:330.6sss\n",
" all 52168 159183 0.851 0.584 0.692 0.409\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 15/200 18G 1.162 0.5224 0.7415 106 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:42<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:320.5sss\n",
" all 52168 159183 0.852 0.585 0.692 0.411\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 16/200 18G 1.155 0.519 0.7416 125 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:38<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.2it/s 1:270.5sss\n",
" all 52168 159183 0.852 0.585 0.693 0.411\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 17/200 18G 1.147 0.514 0.741 106 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:29<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:350.6sss\n",
" all 52168 159183 0.852 0.586 0.695 0.411\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 18/200 18G 1.143 0.5129 0.7403 147 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:47<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:350.6sss\n",
" all 52168 159183 0.852 0.586 0.695 0.412\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 19/200 18G 1.139 0.5106 0.7407 135 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:24<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:390.6sss\n",
" all 52168 159183 0.852 0.586 0.696 0.412\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 20/200 18G 1.134 0.5083 0.7402 106 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:08<1.0s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.5sss\n",
" all 52168 159183 0.853 0.587 0.697 0.412\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 21/200 18G 1.13 0.5055 0.7408 113 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:10<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.7sss\n",
" all 52168 159183 0.853 0.587 0.699 0.414\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 22/200 18G 1.126 0.5043 0.7398 134 640: 100% ━━━━━━━━━━━━ 1188/1188 1.6it/s 12:41<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.2it/s 1:300.7sss\n",
" all 52168 159183 0.853 0.588 0.7 0.414\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 23/200 18G 1.119 0.5009 0.7401 111 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:51<0.9s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:350.5sss\n",
" all 52168 159183 0.855 0.587 0.701 0.415\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 24/200 18G 1.118 0.4995 0.7397 120 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:39<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.2it/s 1:300.6sss\n",
" all 52168 159183 0.854 0.588 0.702 0.416\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 25/200 18G 1.112 0.4975 0.74 124 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:17<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:350.6sss\n",
" all 52168 159183 0.855 0.588 0.703 0.416\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 26/200 18G 1.11 0.4956 0.7392 109 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:45<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.6sss\n",
" all 52168 159183 0.855 0.588 0.705 0.417\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 27/200 18G 1.108 0.4955 0.7395 116 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:42<0.9s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:350.6sss\n",
" all 52168 159183 0.857 0.589 0.707 0.418\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 28/200 18G 1.103 0.4925 0.7393 106 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:22<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:350.6sss\n",
" all 52168 159183 0.857 0.589 0.708 0.419\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 29/200 18G 1.102 0.4903 0.7399 101 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:36<0.9s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:370.6sss\n",
" all 52168 159183 0.857 0.59 0.709 0.419\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 30/200 18G 1.101 0.4911 0.7392 116 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 14:49<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:320.6sss\n",
" all 52168 159183 0.858 0.59 0.71 0.419\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 31/200 18G 1.097 0.4886 0.7395 111 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:54<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.7sss\n",
" all 52168 159183 0.857 0.591 0.711 0.42\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 32/200 18G 1.096 0.4877 0.7391 90 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:41<1.0s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.2it/s 1:270.5sss\n",
" all 52168 159183 0.858 0.591 0.712 0.421\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 33/200 18G 1.094 0.4862 0.7389 122 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:31<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:350.6sss\n",
" all 52168 159183 0.858 0.591 0.712 0.421\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 34/200 18G 1.092 0.4857 0.7389 126 640: 100% ━━━━━━━━━━━━ 1188/1188 1.2it/s 15:58<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:320.6sss\n",
" all 52168 159183 0.859 0.591 0.712 0.422\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 35/200 18G 1.087 0.484 0.7386 102 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:35<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:370.6sss\n",
" all 52168 159183 0.86 0.591 0.713 0.423\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 36/200 18G 1.086 0.4836 0.7385 117 640: 100% ━━━━━━━━━━━━ 1188/1188 1.6it/s 12:42<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:340.6sss\n",
" all 52168 159183 0.859 0.592 0.714 0.423\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 37/200 18G 1.084 0.4811 0.7389 97 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:46<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.9it/s 1:430.6sss\n",
" all 52168 159183 0.86 0.592 0.715 0.424\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 38/200 18G 1.079 0.4805 0.7381 111 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:34<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:310.6sss\n",
" all 52168 159183 0.86 0.593 0.716 0.425\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 39/200 18.4G 1.078 0.479 0.7387 110 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:02<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:320.5sss\n",
" all 52168 159183 0.86 0.593 0.716 0.426\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 40/200 18G 1.077 0.4789 0.739 111 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 15:05<1.0s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.2it/s 1:290.6sss\n",
" all 52168 159183 0.86 0.593 0.716 0.426\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 41/200 18G 1.075 0.4763 0.7384 107 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 15:10<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:330.6sss\n",
" all 52168 159183 0.86 0.593 0.716 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 42/200 18G 1.074 0.4767 0.7384 98 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:02<1.0s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.2it/s 1:300.6sss\n",
" all 52168 159183 0.861 0.593 0.717 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 43/200 18G 1.072 0.4752 0.7387 90 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:24<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:380.6sss\n",
" all 52168 159183 0.862 0.593 0.717 0.429\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 44/200 18G 1.073 0.4763 0.7384 102 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 15:18<1.0s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:380.6sss\n",
" all 52168 159183 0.863 0.593 0.717 0.43\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 45/200 18G 1.069 0.4748 0.7379 118 640: 100% ━━━━━━━━━━━━ 1188/1188 1.2it/s 16:39<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:320.7sss\n",
" all 52168 159183 0.863 0.593 0.718 0.431\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 46/200 18G 1.066 0.4747 0.7382 91 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 14:58<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.2it/s 1:290.6sss\n",
" all 52168 159183 0.863 0.594 0.718 0.431\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 47/200 18G 1.067 0.4751 0.738 95 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:33<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.9it/s 1:400.8sss\n",
" all 52168 159183 0.862 0.593 0.718 0.431\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 48/200 18G 1.065 0.4724 0.738 130 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:33<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:310.5sss\n",
" all 52168 159183 0.862 0.593 0.718 0.432\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 49/200 18G 1.065 0.4726 0.7379 126 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:53<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.9it/s 1:400.6sss\n",
" all 52168 159183 0.861 0.593 0.717 0.43\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 50/200 18G 1.062 0.4712 0.7381 113 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:60<1.0s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:330.6sss\n",
" all 52168 159183 0.862 0.593 0.718 0.431\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 51/200 18G 1.06 0.4701 0.7381 111 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:09<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:370.5sss\n",
" all 52168 159183 0.862 0.593 0.717 0.431\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 52/200 18G 1.057 0.4692 0.7378 111 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 15:09<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.6sss\n",
" all 52168 159183 0.862 0.593 0.717 0.43\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 53/200 18G 1.059 0.4697 0.7379 116 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:18<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:340.6sss\n",
" all 52168 159183 0.862 0.592 0.717 0.429\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 54/200 18G 1.056 0.4684 0.7376 113 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:22<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.2it/s 1:300.6sss\n",
" all 52168 159183 0.861 0.593 0.717 0.429\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 55/200 18G 1.054 0.4668 0.7376 108 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:37<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:350.8sss\n",
" all 52168 159183 0.862 0.592 0.717 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 56/200 18G 1.052 0.466 0.7376 127 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:25<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:320.6sss\n",
" all 52168 159183 0.862 0.592 0.717 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 57/200 18G 1.052 0.4661 0.7377 105 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:07<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.8it/s 1:460.6sss\n",
" all 52168 159183 0.862 0.592 0.717 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 58/200 18G 1.051 0.466 0.7374 112 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:27<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:330.7sss\n",
" all 52168 159183 0.863 0.592 0.717 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 59/200 18G 1.05 0.4645 0.7374 92 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 14:48<0.9s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:350.6sss\n",
" all 52168 159183 0.864 0.591 0.717 0.429\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 60/200 18G 1.049 0.4646 0.7376 116 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 15:09<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:330.6sss\n",
" all 52168 159183 0.863 0.591 0.717 0.429\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 61/200 18G 1.047 0.4628 0.738 122 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:25<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.6sss\n",
" all 52168 159183 0.863 0.591 0.717 0.429\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 62/200 18G 1.046 0.4625 0.7376 101 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:42<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:350.6sss\n",
" all 52168 159183 0.862 0.591 0.717 0.429\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 63/200 18G 1.045 0.462 0.7374 126 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 14:60<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.5sss\n",
" all 52168 159183 0.862 0.591 0.717 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 64/200 18G 1.047 0.4633 0.7374 110 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:19<0.9s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:340.4sss\n",
" all 52168 159183 0.861 0.591 0.718 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 65/200 18G 1.046 0.4622 0.7374 126 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:07<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:340.5sss\n",
" all 52168 159183 0.86 0.592 0.717 0.427\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 66/200 18G 1.044 0.4623 0.737 103 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 14:59<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:330.6sss\n",
" all 52168 159183 0.86 0.591 0.717 0.427\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 67/200 18G 1.042 0.4612 0.7373 113 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:20<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:370.5sss\n",
" all 52168 159183 0.86 0.591 0.717 0.427\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 68/200 18G 1.04 0.4606 0.7373 111 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:09<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:340.7sss\n",
" all 52168 159183 0.859 0.592 0.717 0.427\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 69/200 18G 1.04 0.4602 0.7374 107 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:13<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.9it/s 1:410.7sss\n",
" all 52168 159183 0.859 0.592 0.718 0.427\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 70/200 18G 1.036 0.4588 0.7374 122 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 15:04<1.0s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:340.7sss\n",
" all 52168 159183 0.859 0.592 0.718 0.427\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 71/200 18G 1.036 0.4582 0.7371 93 640: 100% ━━━━━━━━━━━━ 1188/1188 1.5it/s 13:38<1.0s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.9it/s 1:400.6sss\n",
" all 52168 159183 0.859 0.592 0.717 0.427\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 72/200 18G 1.038 0.4582 0.7371 119 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:37<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.5sss\n",
" all 52168 159183 0.859 0.592 0.718 0.427\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 73/200 18G 1.034 0.4574 0.7373 93 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:48<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.6sss\n",
" all 52168 159183 0.86 0.592 0.718 0.427\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 74/200 18G 1.036 0.4579 0.7369 103 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:12<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:350.6sss\n",
" all 52168 159183 0.86 0.592 0.719 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 75/200 18G 1.034 0.4565 0.7368 112 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:55<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.9it/s 1:420.6sss\n",
" all 52168 159183 0.861 0.593 0.719 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 76/200 18G 1.034 0.4559 0.7372 122 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 15:49<1.0s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.9it/s 1:400.6sss\n",
" all 52168 159183 0.862 0.593 0.719 0.429\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 77/200 18G 1.034 0.4562 0.7368 117 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 14:55<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 1.9it/s 1:420.6sss\n",
" all 52168 159183 0.862 0.593 0.719 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 78/200 18G 1.032 0.4559 0.7368 122 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 14:59<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:310.6sss\n",
" all 52168 159183 0.861 0.593 0.719 0.427\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 79/200 18G 1.03 0.4553 0.7373 112 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 14:49<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:310.6sss\n",
" all 52168 159183 0.862 0.593 0.719 0.427\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 80/200 18G 1.03 0.4559 0.7374 131 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 14:44<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.2it/s 1:290.5sss\n",
" all 52168 159183 0.862 0.593 0.72 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 81/200 18G 1.028 0.4541 0.7364 105 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:33<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:320.7sss\n",
" all 52168 159183 0.862 0.593 0.719 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 82/200 18G 1.028 0.4541 0.7371 110 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:25<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.2it/s 1:300.6sss\n",
" all 52168 159183 0.862 0.593 0.72 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 83/200 18G 1.027 0.4526 0.7366 98 640: 100% ━━━━━━━━━━━━ 1188/1188 1.3it/s 15:05<0.9s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:390.6sss\n",
" all 52168 159183 0.862 0.593 0.72 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 84/200 18G 1.028 0.4533 0.7365 99 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:21<0.3s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.7sss\n",
" all 52168 159183 0.862 0.592 0.72 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 85/200 18G 1.027 0.4521 0.7367 101 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:48<0.9s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.0it/s 1:360.6sss\n",
" all 52168 159183 0.862 0.592 0.72 0.428\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 86/200 18G 1.025 0.4525 0.7364 108 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 14:24<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:320.6sss\n",
" all 52168 159183 0.862 0.593 0.72 0.429\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 87/200 18G 1.023 0.4501 0.7367 91 640: 100% ━━━━━━━━━━━━ 1188/1188 1.4it/s 13:48<0.2s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:310.5sss\n",
" all 52168 159183 0.861 0.593 0.72 0.429\n",
"\n",
" Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size\n",
"\u001b[K 88/200 18G 1.023 0.451 0.7363 129 640: 100% ━━━━━━━━━━━━ 1188/1188 1.2it/s 16:46<0.8s\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.1it/s 1:320.7sss\n",
" all 52168 159183 0.861 0.593 0.721 0.429\n",
"\u001b[34m\u001b[1mEarlyStopping: \u001b[0mTraining stopped early as no improvement observed in last 40 epochs. Best results observed at epoch 48, best model saved as best.pt.\n",
"To update EarlyStopping(patience=40) pass a new patience value, i.e. `patience=300` or use `patience=0` to disable EarlyStopping.\n",
"\n",
"88 epochs completed in 23.032 hours.\n",
"Optimizer stripped from /home/cuuva/experiment/custom_LP_detect/lp_detect/epo_200_frac_0_22/weights/last.pt, 6.2MB\n",
"Optimizer stripped from /home/cuuva/experiment/custom_LP_detect/lp_detect/epo_200_frac_0_22/weights/best.pt, 6.2MB\n",
"\n",
"Validating /home/cuuva/experiment/custom_LP_detect/lp_detect/epo_200_frac_0_22/weights/best.pt...\n",
"Ultralytics 8.3.225 🚀 Python-3.10.18 torch-2.9.1+cu128 CUDA:0 (NVIDIA GeForce RTX 5090, 32087MiB)\n",
"Model summary (fused): 72 layers, 3,005,843 parameters, 0 gradients, 8.1 GFLOPs\n",
"\u001b[K Class Images Instances Box(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 194/194 2.3it/s 1:230.4sss\n",
" all 52168 159183 0.862 0.593 0.718 0.431\n",
"Speed: 0.0ms preprocess, 0.2ms inference, 0.0ms loss, 0.3ms postprocess per image\n",
"Results saved to \u001b[1m/home/cuuva/experiment/custom_LP_detect/lp_detect/epo_200_frac_0_22\u001b[0m\n"
]
}
],
"source": [
"# Load a pretrained YOLO11n model\n",
"model = YOLO(\"yolov8n.pt\")\n",
"\n",
"train_results = model.train(\n",
" data=\"/home/cuuva/experiment/custom_LP_detect/custom_LP.yaml\",\n",
" epochs=200,\n",
" imgsz=640,\n",
" batch= -1,\n",
" device=\"cuda\",\n",
" optimizer = 'AdamW',\n",
" lr0 = 0.001,\n",
" patience = 40,\n",
" project = 'lp_detect',\n",
" name = 'epo_200_frac_0_2',\n",
" fraction = 0.2\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "88d6a47e",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Ultralytics 8.3.225 🚀 Python-3.10.18 torch-2.9.1+cu128 CUDA:0 (NVIDIA GeForce RTX 5090, 32109MiB)\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"YOLOv8n summary (fused): 72 layers, 3,005,843 parameters, 0 gradients, 8.1 GFLOPs\n",
"\n",
"\u001b[34m\u001b[1mPyTorch:\u001b[0m starting from '/home/cuuva/experiment/custom_LP_detect/license_plate_detector.pt' with input shape (1, 3, 640, 640) BCHW and output shape(s) (1, 5, 8400) (6.0 MB)\n",
"\n",
"\u001b[34m\u001b[1mONNX:\u001b[0m starting export with onnx 1.19.1 opset 20...\n",
"\u001b[34m\u001b[1mONNX:\u001b[0m slimming with onnxslim 0.1.71...\n",
"\u001b[34m\u001b[1mONNX:\u001b[0m export success ✅ 0.3s, saved as '/home/cuuva/experiment/custom_LP_detect/license_plate_detector.onnx' (11.7 MB)\n",
"\n",
"Export complete (0.4s)\n",
"Results saved to \u001b[1m/home/cuuva/experiment/custom_LP_detect\u001b[0m\n",
"Predict: yolo predict task=detect model=/home/cuuva/experiment/custom_LP_detect/license_plate_detector.onnx imgsz=640 \n",
"Validate: yolo val task=detect model=/home/cuuva/experiment/custom_LP_detect/license_plate_detector.onnx imgsz=640 data=config.yaml \n",
"Visualize: https://netron.app\n"
]
},
{
"data": {
"text/plain": [
"'/home/cuuva/experiment/custom_LP_detect/license_plate_detector.onnx'"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# model = YOLO(\"/home/cuuva/experiment/custom_LP_detect/lp_detect/epo_200_frac_0_22/weights/best_lp_detect.pt\")\n",
"model = YOLO(\"/home/cuuva/experiment/custom_LP_detect/license_plate_detector.pt\")\n",
"model.export(format=\"onnx\", imgsz=640, device=0)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "0084e5f9",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "1stagedetect",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.18"
}
},
"nbformat": 4,
"nbformat_minor": 5
}