train_dairy,drink_yolov5s.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyPUE0HjdMzYr/7mPhZlKoVR",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "gpuClass": "premium"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/dbtjr1103/mainpj/blob/main/train_dairy%2Cdrink_yolov5s.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_15erAu3bXQF"
      },
      "outputs": [],
      "source": [
        "!pip install roboflow"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from roboflow import Roboflow\n",
        "rf = Roboflow(api_key=\"x2g0cc9LVwtb4OuDYTFj\")\n",
        "project = rf.workspace(\"last-s46sz\").project(\"dairy-nvrhm\")\n",
        "dataset = project.version(1).download(\"yolov5\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TOn8v5BZbbWE",
        "outputId": "486e7f6d-9d4f-4788-91fa-1dd9c6fd1f67"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "loading Roboflow workspace...\n",
            "loading Roboflow project...\n",
            "Downloading Dataset Version Zip in dairy-1 to yolov5pytorch: 100% [175307806 / 175307806] bytes\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Extracting Dataset Version Zip to dairy-1 in yolov5pytorch:: 100%|██████████| 7212/7212 [00:00<00:00, 7295.56it/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!git clone https://github.com/ultralytics/yolov5  # clone repo\n",
        "%cd yolov5"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Xx1BvBXMb5C9",
        "outputId": "e4558de4-f173-43dc-a7b3-60024e0fc00b"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cloning into 'yolov5'...\n",
            "remote: Enumerating objects: 15008, done.\u001b[K\n",
            "remote: Counting objects: 100% (6/6), done.\u001b[K\n",
            "remote: Compressing objects: 100% (6/6), done.\u001b[K\n",
            "remote: Total 15008 (delta 2), reused 2 (delta 0), pack-reused 15002\u001b[K\n",
            "Receiving objects: 100% (15008/15008), 14.03 MiB | 33.96 MiB/s, done.\n",
            "Resolving deltas: 100% (10293/10293), done.\n",
            "/content/yolov5\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install -qr requirements.txt  "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "h968royHcy-o",
        "outputId": "4ca3b18d-b32a-4c76-a6a3-170733630683"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[?25l     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0.0/184.0 KB\u001b[0m \u001b[31m?\u001b[0m eta \u001b[36m-:--:--\u001b[0m\r\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m184.0/184.0 KB\u001b[0m \u001b[31m5.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m62.7/62.7 KB\u001b[0m \u001b[31m8.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.6/1.6 MB\u001b[0m \u001b[31m35.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import utils\n",
        "from IPython.display import Image, clear_output  # to display images\n",
        "from utils.downloads import attempt_download\n",
        "%cat /content/dairy-1/data.yaml"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zYzo3WmfcwCj",
        "outputId": "e9b20dc6-485a-4f14-af46-c0c7048771c9"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "names:\n",
            "- \"갈바니리코타치즈425G\"\n",
            "- \"남양)아이꼬야아기랑쌀이랑1-080ML\"\n",
            "- \"남양)아이꼬야아기랑쌀이랑180ML\"\n",
            "- \"남양맛있는우유GT\"\n",
            "- \"남양맛있는우유GT슈퍼밀크\"\n",
            "- \"남양아인슈타인베이비\"\n",
            "- \"남양유업떠먹는불가리스생크림요거트85G\"\n",
            "- \"남양이오트웬티즈\"\n",
            "- \"남양초코에몽250ML\"\n",
            "- \"덴마크드링킹요구르트사과\"\n",
            "- \"덴마크드링킹요구르트자두\"\n",
            "- \"덴마크드링킹요구르트청포도\"\n",
            "- \"덴마크민트라떼\"\n",
            "- \"덴마크바나나우유\"\n",
            "- \"동원대니쉬THE건강한우유\"\n",
            "- \"동원덴마크요거샐러드퍼플150G\"\n",
            "- \"동원소와나무입안가득부드러운생크림요거트85G\"\n",
            "- \"매일루어팍가염버터250G\"\n",
            "- \"매일바이오드링킹요커트플레인\"\n",
            "- \"매일바이오딸기요거트\"\n",
            "- \"매일바이오플레인스위트\"\n",
            "- \"매일상하목장유기농우유저지방2_\"\n",
            "- \"매일썬업그린(과일야채샐러드)\"\n",
            "- \"매일아몬드브리즈뉴트리플러스프로틴\"\n",
            "- \"매일유업)바이오프로바이오틱생크림요거크플래인80G\"\n",
            "- \"매일유업)바이오프로바이오틱생크림요거트유자80G\"\n",
            "- \"매일유업바이오달지않은플레인450G\"\n",
            "- \"매일카스텔로까망베르125G\"\n",
            "- \"매일피크닉청포도\"\n",
            "- \"밀크씨슬간편한닥터캡슐\"\n",
            "- \"벨치즈코리아끼리그릭스타일200G\"\n",
            "- \"빙그레)요플레라이트80G\"\n",
            "- \"빙그레요플레닥터캡슐프로텍트플레인\"\n",
            "- \"빙그레요플레오리지널딸기\"\n",
            "- \"빙그레요플레오리지널딸기85G\"\n",
            "- \"빙그레요플레오리지널망고스틴80G\"\n",
            "- \"빙그레요플레오리지널복숭아85G\"\n",
            "- \"빙그레요플레오프룻복숭아\"\n",
            "- \"빙그레요플레오프룻블루베리\"\n",
            "- \"빙그레요플레플레인85G\"\n",
            "- \"상하목장)유기농베이비요구르트85G\"\n",
            "- \"서울에프앤비푸르밀바나나우유\"\n",
            "- \"서울에프앤비푸르밀생초콜릿우유\"\n",
            "- \"서울우유\"\n",
            "- \"서울우유(사각팩)\"\n",
            "- \"서울우유)서울연유500G\"\n",
            "- \"서울우유)치즈큐빅84G\"\n",
            "- \"서울우유귀리우유\"\n",
            "- \"서울우유성장기우유앙팡\"\n",
            "- \"서울우유속편한우유저지방\"\n",
            "- \"서울우유오징어치즈36G\"\n",
            "- \"서울우유저지방나100_\"\n",
            "- \"서울우유후레쉬밀크900ML\"\n",
            "- \"서울우유흑임자우유\"\n",
            "- \"소와나무덴마크드링킹요구르트베리믹스\"\n",
            "- \"소와나무덴마크드링킹요구르트파인애플\"\n",
            "- \"소와나무덴마크초코초코우유\"\n",
            "- \"소와나무덴마크커피커피\"\n",
            "- \"소와나무쿨피스오리지널복숭아\"\n",
            "- \"소와나무쿨피스오리지널파인애플\"\n",
            "- \"심플러스국내산1등급원유로만든우유\"\n",
            "- \"야쿠르트쿠퍼스프리미엄C140ML\"\n",
            "- \"영홍식품)엘사보르나쵸앤딥치즈175G\"\n",
            "- \"이마트그릭요거트그대로플레인20G\"\n",
            "- \"일동후디스그릭요거트무설탕저지방80G\"\n",
            "- \"크래프트하인즈코리아크래프트프리미엄크림치즈226G\"\n",
            "- \"태광웰푸드)처음먹는칼슘츄러스치즈45-5G\"\n",
            "- \"토핑오트_애플시나몬\"\n",
            "- \"파스퇴르)락토프리속편한쾌변150ML\"\n",
            "- \"파스퇴르)바른목장85G\"\n",
            "- \"파스퇴르_오직요구르트100_400G\"\n",
            "- \"파스퇴르바른목장85G\"\n",
            "- \"파스퇴르요구하이\"\n",
            "- \"파스퇴르위편한하루\"\n",
            "- \"폰타나파르마콰트로치즈430G\"\n",
            "- \"푸드웰뽀로로워터젤리포도\"\n",
            "- \"푸르밀)살아있는생유산균프로바이오요거트딸기85G\"\n",
            "- \"푸르밀)프로바이오사과150ml\"\n",
            "- \"푸르밀미숫가루우유\"\n",
            "- \"푸르밀뼈가튼튼고칼슘저지방\"\n",
            "- \"푸르밀흑당생크림요거트80G\"\n",
            "- \"풀무원)위솔루션130ML\"\n",
            "- \"풀무원다논더눈솔루션130ML\"\n",
            "- \"퓨어잇사과엔요거트16G\"\n",
            "- \"프로틴고단백질요거트\"\n",
            "- \"한국야구르트슈퍼100딸기90G\"\n",
            "- \"한국야쿠르트)얼려먹는야쿠르트110ML\"\n",
            "- \"한국야쿠르트건강한간케어쿠퍼스프리미엄140ML\"\n",
            "- \"헬리코박터프로젝트윌150ML\"\n",
            "- \"홈플러스)고다애에담치즈스틱75G\"\n",
            "- \"홈플러스마시는유산군DELIGHT\"\n",
            "- \"홈플러스마시는유산균\"\n",
            "- \"화인베지터블오일스프레드212G\"\n",
            "- \"후디스그릭요거트리얼블루베리320g(번들)\"\n",
            "- \"후디스그릭요거트리얼블루베리80G\"\n",
            "- \"후디스그릭요거트플레인80G\"\n",
            "- \"후디스코코포도그릭80G\"\n",
            "- \"후디스키요그릭요거트젤리블루베리50G\"\n",
            "- \"후디스키요그릭요거트젤리청포도50G\"\n",
            "nc: 99\n",
            "roboflow:\n",
            "  license: CC BY 4.0\n",
            "  project: dairy-nvrhm\n",
            "  url: https://universe.roboflow.com/project/dairy-nvrhm/dataset/1\n",
            "  version: 1\n",
            "  workspace: project\n",
            "test: ../test/images\n",
            "train: dairy-1/train/images\n",
            "val: dairy-1/valid/images\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# train yolov5s on custom data for 100 epochs\n",
        "# time its performance\n",
        "%%time\n",
        "%cd /content/yolov5/\n",
        "!python train.py --img 640 --batch 16 --epochs 500 --data /content/dairy-1/data.yaml --weights 'yolov5s.pt' --name yolov5s_results  --cache"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qBx6A4opc3wr",
        "outputId": "1c0b867f-6113-4f49-a0fa-184f02ad3117"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/yolov5\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mweights=yolov5s.pt, cfg=, data=/content/dairy-1/data.yaml, hyp=data/hyps/hyp.scratch-low.yaml, epochs=500, 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=yolov5s_results, 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\n",
            "\u001b[34m\u001b[1mgithub: \u001b[0mup to date with https://github.com/ultralytics/yolov5 ✅\n",
            "YOLOv5 🚀 v7.0-72-g064365d Python-3.8.10 torch-1.13.1+cu116 CUDA:0 (NVIDIA A100-SXM4-40GB, 40536MiB)\n",
            "\n",
            "\u001b[34m\u001b[1mhyperparameters: \u001b[0mlr0=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\n",
            "\u001b[34m\u001b[1mClearML: \u001b[0mrun 'pip install clearml' to automatically track, visualize and remotely train YOLOv5 🚀 in ClearML\n",
            "\u001b[34m\u001b[1mComet: \u001b[0mrun 'pip install comet_ml' to automatically track and visualize YOLOv5 🚀 runs in Comet\n",
            "\u001b[34m\u001b[1mTensorBoard: \u001b[0mStart with 'tensorboard --logdir runs/train', view at http://localhost:6006/\n",
            "2023-02-04 08:12:11.933174: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
            "Downloading https://ultralytics.com/assets/Arial.Unicode.ttf to /root/.config/Ultralytics/Arial.Unicode.ttf...\n",
            "100% 22.2M/22.2M [00:00<00:00, 46.6MB/s]\n",
            "Downloading https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5s.pt to yolov5s.pt...\n",
            "100% 14.1M/14.1M [00:00<00:00, 129MB/s]\n",
            "\n",
            "Overriding model.yaml nc=80 with nc=99\n",
            "\n",
            "                 from  n    params  module                                  arguments                     \n",
            "  0                -1  1      3520  models.common.Conv                      [3, 32, 6, 2, 2]              \n",
            "  1                -1  1     18560  models.common.Conv                      [32, 64, 3, 2]                \n",
            "  2                -1  1     18816  models.common.C3                        [64, 64, 1]                   \n",
            "  3                -1  1     73984  models.common.Conv                      [64, 128, 3, 2]               \n",
            "  4                -1  2    115712  models.common.C3                        [128, 128, 2]                 \n",
            "  5                -1  1    295424  models.common.Conv                      [128, 256, 3, 2]              \n",
            "  6                -1  3    625152  models.common.C3                        [256, 256, 3]                 \n",
            "  7                -1  1   1180672  models.common.Conv                      [256, 512, 3, 2]              \n",
            "  8                -1  1   1182720  models.common.C3                        [512, 512, 1]                 \n",
            "  9                -1  1    656896  models.common.SPPF                      [512, 512, 5]                 \n",
            " 10                -1  1    131584  models.common.Conv                      [512, 256, 1, 1]              \n",
            " 11                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 12           [-1, 6]  1         0  models.common.Concat                    [1]                           \n",
            " 13                -1  1    361984  models.common.C3                        [512, 256, 1, False]          \n",
            " 14                -1  1     33024  models.common.Conv                      [256, 128, 1, 1]              \n",
            " 15                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 16           [-1, 4]  1         0  models.common.Concat                    [1]                           \n",
            " 17                -1  1     90880  models.common.C3                        [256, 128, 1, False]          \n",
            " 18                -1  1    147712  models.common.Conv                      [128, 128, 3, 2]              \n",
            " 19          [-1, 14]  1         0  models.common.Concat                    [1]                           \n",
            " 20                -1  1    296448  models.common.C3                        [256, 256, 1, False]          \n",
            " 21                -1  1    590336  models.common.Conv                      [256, 256, 3, 2]              \n",
            " 22          [-1, 10]  1         0  models.common.Concat                    [1]                           \n",
            " 23                -1  1   1182720  models.common.C3                        [512, 512, 1, False]          \n",
            " 24      [17, 20, 23]  1    280488  models.yolo.Detect                      [99, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512]]\n",
            "Model summary: 214 layers, 7286632 parameters, 7286632 gradients, 16.8 GFLOPs\n",
            "\n",
            "Transferred 343/349 items from yolov5s.pt\n",
            "\u001b[34m\u001b[1mAMP: \u001b[0mchecks passed ✅\n",
            "\u001b[34m\u001b[1moptimizer:\u001b[0m SGD(lr=0.01) with parameter groups 57 weight(decay=0.0), 60 weight(decay=0.0005), 60 bias\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), CLAHE(p=0.01, clip_limit=(1, 4.0), tile_grid_size=(8, 8))\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mScanning /content/dairy-1/train/labels... 3150 images, 0 backgrounds, 0 corrupt: 100% 3150/3150 [00:00<00:00, 7873.65it/s]\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mNew cache created: /content/dairy-1/train/labels.cache\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mCaching images (3.6GB ram): 100% 3150/3150 [00:02<00:00, 1168.13it/s]\n",
            "\u001b[34m\u001b[1mval: \u001b[0mScanning /content/dairy-1/valid/labels... 300 images, 0 backgrounds, 0 corrupt: 100% 300/300 [00:00<00:00, 5115.11it/s]\n",
            "\u001b[34m\u001b[1mval: \u001b[0mNew cache created: /content/dairy-1/valid/labels.cache\n",
            "\u001b[34m\u001b[1mval: \u001b[0mCaching images (0.3GB ram): 100% 300/300 [00:00<00:00, 1049.04it/s]\n",
            "\n",
            "\u001b[34m\u001b[1mAutoAnchor: \u001b[0m3.71 anchors/target, 1.000 Best Possible Recall (BPR). Current anchors are a good fit to dataset ✅\n",
            "Plotting labels to runs/train/yolov5s_results/labels.jpg... \n",
            "Image sizes 640 train, 640 val\n",
            "Using 8 dataloader workers\n",
            "Logging results to \u001b[1mruns/train/yolov5s_results\u001b[0m\n",
            "Starting training for 500 epochs...\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      0/499       3.4G     0.1133    0.06992     0.1142        158        640:   0% 0/197 [00:01<?, ?it/s]Downloading https://ultralytics.com/assets/Arial.ttf to /root/.config/Ultralytics/Arial.ttf...\n",
            "      0/499      3.47G     0.1123    0.06962     0.1147        161        640:   1% 2/197 [00:03<04:23,  1.35s/it]Downloading https://ultralytics.com/assets/Arial.ttf to /root/.config/Ultralytics/Arial.ttf...\n",
            "      0/499      3.47G     0.1108    0.07154     0.1143        170        640:   2% 3/197 [00:03<02:37,  1.24it/s]Downloading https://ultralytics.com/assets/Arial.ttf to /root/.config/Ultralytics/Arial.ttf...\n",
            "      0/499      3.95G    0.06581    0.06243     0.1107        147        640: 100% 197/197 [00:23<00:00,  8.35it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:02<00:00,  3.37it/s]\n",
            "                   all        300        485    0.00721      0.936     0.0265     0.0146\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      1/499      4.85G    0.03937     0.0473     0.1069        104        640: 100% 197/197 [00:18<00:00, 10.54it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.69it/s]\n",
            "                   all        300        485    0.00822      0.965     0.0313      0.022\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      2/499      4.85G    0.03482    0.04216     0.1057        136        640: 100% 197/197 [00:18<00:00, 10.65it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.98it/s]\n",
            "                   all        300        485    0.00859      0.929     0.0372     0.0265\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      3/499      4.85G    0.03026    0.04029     0.1048        115        640: 100% 197/197 [00:18<00:00, 10.88it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.59it/s]\n",
            "                   all        300        485    0.00913      0.985     0.0407     0.0318\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      4/499      4.85G    0.02719    0.03825     0.1039        133        640: 100% 197/197 [00:17<00:00, 11.08it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.62it/s]\n",
            "                   all        300        485    0.00969      0.998     0.0579     0.0455\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      5/499      4.85G    0.02538    0.03721     0.1026        175        640: 100% 197/197 [00:17<00:00, 11.01it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.51it/s]\n",
            "                   all        300        485    0.00707      0.997     0.0646      0.048\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      6/499      4.85G    0.02409    0.03673     0.1006        160        640: 100% 197/197 [00:18<00:00, 10.93it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.75it/s]\n",
            "                   all        300        485    0.00995      0.996     0.0889     0.0674\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      7/499      4.85G    0.02388    0.03552    0.09815        126        640: 100% 197/197 [00:17<00:00, 11.07it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.50it/s]\n",
            "                   all        300        485      0.805     0.0983      0.106      0.088\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      8/499      4.85G    0.02374    0.03568    0.09536        139        640: 100% 197/197 [00:18<00:00, 10.80it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.66it/s]\n",
            "                   all        300        485      0.686      0.123       0.15      0.117\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      9/499      4.85G     0.0234    0.03367    0.09215        136        640: 100% 197/197 [00:17<00:00, 11.03it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.57it/s]\n",
            "                   all        300        485      0.657      0.189      0.164      0.133\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     10/499      4.85G     0.0231    0.03422    0.08919        140        640: 100% 197/197 [00:17<00:00, 11.02it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.56it/s]\n",
            "                   all        300        485      0.625      0.192      0.176       0.14\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     11/499      4.85G    0.02317    0.03387     0.0862        140        640: 100% 197/197 [00:18<00:00, 10.84it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.60it/s]\n",
            "                   all        300        485      0.508      0.314      0.233      0.178\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     12/499      4.85G      0.023    0.03318    0.08365        149        640: 100% 197/197 [00:18<00:00, 10.86it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.54it/s]\n",
            "                   all        300        485      0.537      0.355      0.269      0.225\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     13/499      4.85G    0.02289      0.033    0.08073        135        640: 100% 197/197 [00:18<00:00, 10.87it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.35it/s]\n",
            "                   all        300        485      0.503      0.373      0.279      0.233\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     14/499      4.85G    0.02298    0.03294     0.0775        114        640: 100% 197/197 [00:18<00:00, 10.86it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.59it/s]\n",
            "                   all        300        485      0.502      0.428      0.359      0.307\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     15/499      4.85G    0.02282     0.0331    0.07477        160        640: 100% 197/197 [00:17<00:00, 11.11it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.58it/s]\n",
            "                   all        300        485      0.445       0.43      0.356      0.299\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     16/499      4.85G    0.02354    0.03255    0.07164        133        640: 100% 197/197 [00:17<00:00, 11.04it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.55it/s]\n",
            "                   all        300        485      0.497      0.536      0.437      0.379\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     17/499      4.85G    0.02356    0.03215    0.06789        156        640: 100% 197/197 [00:18<00:00, 10.92it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.53it/s]\n",
            "                   all        300        485      0.435      0.582      0.514      0.446\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     18/499      4.85G    0.02331    0.03216    0.06474        107        640: 100% 197/197 [00:17<00:00, 11.05it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.58it/s]\n",
            "                   all        300        485      0.448      0.614      0.564      0.487\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     19/499      4.85G    0.02372    0.03227    0.06281        147        640: 100% 197/197 [00:17<00:00, 11.02it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.45it/s]\n",
            "                   all        300        485      0.449      0.703      0.613      0.519\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     20/499      4.85G     0.0233    0.03204    0.05897        143        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.51it/s]\n",
            "                   all        300        485      0.467      0.737      0.665      0.579\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     21/499      4.85G    0.02359    0.03171    0.05689        118        640: 100% 197/197 [00:18<00:00, 10.80it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.58it/s]\n",
            "                   all        300        485      0.577      0.705       0.72      0.634\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     22/499      4.85G    0.02395    0.03157    0.05323        152        640: 100% 197/197 [00:17<00:00, 11.02it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.67it/s]\n",
            "                   all        300        485      0.563      0.702      0.719      0.612\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     23/499      4.85G    0.02398    0.03198    0.05059        136        640: 100% 197/197 [00:18<00:00, 10.91it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.39it/s]\n",
            "                   all        300        485      0.641      0.762      0.776      0.661\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     24/499      4.85G    0.02409    0.03152    0.04821        149        640: 100% 197/197 [00:18<00:00, 10.94it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.73it/s]\n",
            "                   all        300        485      0.621      0.805      0.823      0.706\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     25/499      4.85G    0.02386    0.03159    0.04542        131        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.53it/s]\n",
            "                   all        300        485      0.682      0.825      0.862      0.758\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     26/499      4.85G    0.02395    0.03155     0.0443        113        640: 100% 197/197 [00:17<00:00, 11.09it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.67it/s]\n",
            "                   all        300        485      0.733      0.828      0.868      0.751\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     27/499      4.85G    0.02424     0.0314     0.0403        122        640: 100% 197/197 [00:17<00:00, 10.99it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.49it/s]\n",
            "                   all        300        485      0.738      0.847      0.888      0.772\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     28/499      4.85G    0.02456    0.03118    0.03864        119        640: 100% 197/197 [00:18<00:00, 10.79it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.65it/s]\n",
            "                   all        300        485      0.811      0.851      0.913      0.797\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     29/499      4.85G    0.02373    0.03108    0.03705        138        640: 100% 197/197 [00:17<00:00, 11.34it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.58it/s]\n",
            "                   all        300        485      0.778      0.886      0.931      0.807\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     30/499      4.85G    0.02438    0.03099    0.03647        150        640: 100% 197/197 [00:18<00:00, 10.69it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.43it/s]\n",
            "                   all        300        485      0.826      0.902      0.949       0.85\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     31/499      4.85G    0.02421    0.03144    0.03499        133        640: 100% 197/197 [00:17<00:00, 10.96it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.56it/s]\n",
            "                   all        300        485       0.86      0.902      0.951      0.839\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     32/499      4.85G    0.02378     0.0309    0.03355        127        640: 100% 197/197 [00:17<00:00, 10.96it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.48it/s]\n",
            "                   all        300        485       0.84      0.928      0.965      0.866\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     33/499      4.85G    0.02364    0.03102    0.03232        141        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.51it/s]\n",
            "                   all        300        485       0.84      0.939      0.966      0.863\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     34/499      4.85G    0.02395     0.0311    0.03063        132        640: 100% 197/197 [00:17<00:00, 11.11it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.56it/s]\n",
            "                   all        300        485      0.852       0.89      0.952      0.845\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     35/499      4.85G    0.02358    0.03061     0.0302        113        640: 100% 197/197 [00:18<00:00, 10.89it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.64it/s]\n",
            "                   all        300        485      0.848       0.92      0.952      0.832\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     36/499      4.85G    0.02349    0.03068    0.02917        123        640: 100% 197/197 [00:17<00:00, 11.04it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.71it/s]\n",
            "                   all        300        485      0.871      0.912      0.962       0.86\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     37/499      4.85G      0.024    0.03047    0.02752        140        640: 100% 197/197 [00:17<00:00, 10.96it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.64it/s]\n",
            "                   all        300        485      0.914      0.887      0.974      0.854\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     38/499      4.85G    0.02368    0.03061     0.0271        138        640: 100% 197/197 [00:18<00:00, 10.91it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.59it/s]\n",
            "                   all        300        485      0.866      0.919      0.964      0.838\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     39/499      4.85G    0.02329    0.03013    0.02536        127        640: 100% 197/197 [00:18<00:00, 10.89it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.63it/s]\n",
            "                   all        300        485      0.916      0.948      0.983      0.886\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     40/499      4.85G    0.02318    0.03052     0.0266        117        640: 100% 197/197 [00:17<00:00, 11.03it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.61it/s]\n",
            "                   all        300        485      0.901      0.883      0.968      0.872\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     41/499      4.85G    0.02363    0.03015    0.02548        126        640: 100% 197/197 [00:18<00:00, 10.91it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.65it/s]\n",
            "                   all        300        485      0.918      0.956      0.981      0.881\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     42/499      4.85G    0.02304     0.0302    0.02491        110        640: 100% 197/197 [00:17<00:00, 11.00it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.73it/s]\n",
            "                   all        300        485       0.91      0.942      0.979      0.872\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     43/499      4.85G    0.02354     0.0305    0.02433        131        640: 100% 197/197 [00:17<00:00, 11.01it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.65it/s]\n",
            "                   all        300        485      0.916      0.939      0.974      0.859\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     44/499      4.85G    0.02279     0.0304    0.02399        115        640: 100% 197/197 [00:17<00:00, 11.06it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.66it/s]\n",
            "                   all        300        485       0.94      0.935      0.983      0.887\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     45/499      4.85G    0.02299    0.02939     0.0242        157        640: 100% 197/197 [00:18<00:00, 10.92it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.52it/s]\n",
            "                   all        300        485      0.878      0.935      0.981      0.877\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     46/499      4.85G    0.02296    0.03018    0.02287        150        640: 100% 197/197 [00:18<00:00, 10.90it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.66it/s]\n",
            "                   all        300        485      0.899      0.945      0.979      0.886\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     47/499      4.85G    0.02272    0.02999    0.02246        118        640: 100% 197/197 [00:18<00:00, 10.85it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.67it/s]\n",
            "                   all        300        485      0.925      0.922      0.982      0.884\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     48/499      4.85G    0.02257    0.02947    0.02206        134        640: 100% 197/197 [00:18<00:00, 10.71it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.63it/s]\n",
            "                   all        300        485      0.943      0.943      0.987      0.888\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     49/499      4.85G    0.02271    0.03028    0.02146        130        640: 100% 197/197 [00:17<00:00, 11.08it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.67it/s]\n",
            "                   all        300        485      0.941      0.957      0.986       0.89\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     50/499      4.85G    0.02255    0.02976    0.02123        164        640: 100% 197/197 [00:17<00:00, 11.13it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.70it/s]\n",
            "                   all        300        485      0.912      0.967       0.99      0.898\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     51/499      4.85G    0.02189    0.02982    0.02137        124        640: 100% 197/197 [00:17<00:00, 10.98it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.52it/s]\n",
            "                   all        300        485      0.923      0.974      0.993      0.908\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     52/499      4.85G     0.0221    0.02912    0.02073        153        640: 100% 197/197 [00:18<00:00, 10.88it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.41it/s]\n",
            "                   all        300        485      0.944      0.929       0.99      0.894\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     53/499      4.85G    0.02195    0.02964    0.02099        137        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.63it/s]\n",
            "                   all        300        485      0.922      0.932      0.981      0.883\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     54/499      4.85G    0.02231    0.02959     0.0199        112        640: 100% 197/197 [00:18<00:00, 10.94it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.70it/s]\n",
            "                   all        300        485      0.917      0.941      0.976      0.868\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     55/499      4.85G    0.02223    0.02882    0.01945        127        640: 100% 197/197 [00:17<00:00, 11.05it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.73it/s]\n",
            "                   all        300        485      0.939       0.96      0.986      0.896\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     56/499      4.85G    0.02184      0.029    0.01951        171        640: 100% 197/197 [00:18<00:00, 10.93it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.71it/s]\n",
            "                   all        300        485      0.904      0.962      0.985       0.89\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     57/499      4.85G    0.02224    0.02925    0.01959        129        640: 100% 197/197 [00:17<00:00, 11.03it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.54it/s]\n",
            "                   all        300        485      0.926      0.975      0.989      0.893\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     58/499      4.85G    0.02176    0.02917    0.01888        119        640: 100% 197/197 [00:18<00:00, 10.85it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.59it/s]\n",
            "                   all        300        485      0.911      0.941      0.978      0.887\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     59/499      4.85G    0.02158    0.02887    0.01894        154        640: 100% 197/197 [00:18<00:00, 10.88it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.73it/s]\n",
            "                   all        300        485      0.926      0.946      0.985        0.9\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     60/499      4.85G    0.02196    0.02884    0.01869        126        640: 100% 197/197 [00:18<00:00, 10.91it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.63it/s]\n",
            "                   all        300        485      0.949      0.973      0.992      0.907\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     61/499      4.85G     0.0216    0.02887    0.01797        118        640: 100% 197/197 [00:17<00:00, 11.14it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.48it/s]\n",
            "                   all        300        485      0.934      0.969      0.989      0.897\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     62/499      4.85G    0.02152    0.02889    0.01843        121        640: 100% 197/197 [00:18<00:00, 10.89it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.37it/s]\n",
            "                   all        300        485      0.953      0.939       0.99      0.908\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     63/499      4.85G     0.0216     0.0291     0.0185        169        640: 100% 197/197 [00:17<00:00, 11.09it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.70it/s]\n",
            "                   all        300        485      0.951      0.963       0.99      0.907\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     64/499      4.85G    0.02163    0.02872     0.0182        116        640: 100% 197/197 [00:18<00:00, 10.88it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.64it/s]\n",
            "                   all        300        485      0.954       0.94      0.991      0.914\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     65/499      4.85G    0.02192    0.02898    0.01752        149        640: 100% 197/197 [00:18<00:00, 10.84it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.25it/s]\n",
            "                   all        300        485      0.936      0.982      0.993      0.913\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     66/499      4.85G     0.0212    0.02855    0.01857        130        640: 100% 197/197 [00:18<00:00, 10.87it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.65it/s]\n",
            "                   all        300        485      0.961       0.98      0.993      0.914\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     67/499      4.85G    0.02186    0.02844    0.01752        157        640: 100% 197/197 [00:17<00:00, 11.12it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.60it/s]\n",
            "                   all        300        485      0.958      0.966       0.99      0.915\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     68/499      4.85G     0.0217    0.02859    0.01656        146        640: 100% 197/197 [00:18<00:00, 10.80it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.60it/s]\n",
            "                   all        300        485       0.96       0.98      0.993      0.916\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     69/499      4.85G    0.02133    0.02835    0.01642        132        640: 100% 197/197 [00:17<00:00, 11.04it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.52it/s]\n",
            "                   all        300        485      0.963      0.973      0.992      0.916\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     70/499      4.85G     0.0208    0.02787    0.01621        144        640: 100% 197/197 [00:17<00:00, 11.27it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.56it/s]\n",
            "                   all        300        485      0.941      0.972       0.99      0.922\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "     71/499      4.85G    0.02119    0.02541    0.01633        133        640:   2% 4/197 [00:00<00:16, 11.39it/s]Exception ignored in: <generator object tqdm.__iter__ at 0x7f16427e86d0>\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.8/dist-packages/tqdm/std.py\", line 1210, in __iter__\n",
            "    self.close()\n",
            "  File \"/usr/local/lib/python3.8/dist-packages/tqdm/std.py\", line 1289, in close\n",
            "    self._decr_instances(self)\n",
            "  File \"/usr/local/lib/python3.8/dist-packages/tqdm/std.py\", line 593, in _decr_instances\n",
            "    with cls._lock:\n",
            "  File \"/usr/local/lib/python3.8/dist-packages/tqdm/std.py\", line 113, in __enter__\n",
            "    self.acquire()\n",
            "  File \"/usr/local/lib/python3.8/dist-packages/tqdm/std.py\", line 105, in acquire\n",
            "    for lock in self.locks:\n",
            "KeyboardInterrupt: \n",
            "Traceback (most recent call last):\n",
            "  File \"train.py\", line 634, in <module>\n",
            "    main(opt)\n",
            "  File \"train.py\", line 528, in main\n",
            "    train(opt.hyp, opt, device, callbacks)\n",
            "  File \"train.py\", line 317, in train\n",
            "    scaler.scale(loss).backward()\n",
            "  File \"/usr/local/lib/python3.8/dist-packages/torch/_tensor.py\", line 488, in backward\n",
            "    torch.autograd.backward(\n",
            "  File \"/usr/local/lib/python3.8/dist-packages/torch/autograd/__init__.py\", line 197, in backward\n",
            "Exception in thread Thread-18:\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/lib/python3.8/threading.py\", line 932, in _bootstrap_inner\n",
            "    Variable._execution_engine.run_backward(  # Calls into the C++ engine to run the backward pass\n",
            "KeyboardInterrupt\n",
            "    self.run()\n",
            "  File \"/usr/lib/python3.8/threading.py\", line 870, in run\n",
            "    self._target(*self._args, **self._kwargs)\n",
            "  File \"/usr/local/lib/python3.8/dist-packages/torch/utils/data/_utils/pin_memory.py\", line 49, in _pin_memory_loop\n",
            "    do_one_step()\n",
            "  File \"/usr/local/lib/python3.8/dist-packages/torch/utils/data/_utils/pin_memory.py\", line 26, in do_one_step\n",
            "    r = in_queue.get(timeout=MP_STATUS_CHECK_INTERVAL)\n",
            "  File \"/usr/lib/python3.8/multiprocessing/queues.py\", line 116, in get\n",
            "    return _ForkingPickler.loads(res)\n",
            "  File \"/usr/local/lib/python3.8/dist-packages/torch/multiprocessing/reductions.py\", line 305, in rebuild_storage_fd\n",
            "    fd = df.detach()\n",
            "  File \"/usr/lib/python3.8/multiprocessing/resource_sharer.py\", line 57, in detach\n",
            "    with _resource_sharer.get_connection(self._id) as conn:\n",
            "  File \"/usr/lib/python3.8/multiprocessing/resource_sharer.py\", line 87, in get_connection\n",
            "    c = Client(address, authkey=process.current_process().authkey)\n",
            "  File \"/usr/lib/python3.8/multiprocessing/connection.py\", line 502, in Client\n",
            "    c = SocketClient(address)\n",
            "  File \"/usr/lib/python3.8/multiprocessing/connection.py\", line 630, in SocketClient\n",
            "    s.connect(address)\n",
            "FileNotFoundError: [Errno 2] No such file or directory\n",
            "^C\n",
            "CPU times: user 21.9 s, sys: 3.51 s, total: 25.4 s\n",
            "Wall time: 24min 44s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!python val.py --weights /content/yolov5/runs/train/yolov5s_results/weights/best.pt --data /content/dairy-1/data.yaml --img 640"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zEmvIjMOkkcv",
        "outputId": "8657a397-cc00-492e-8ac3-d5b5c0fc1690"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[34m\u001b[1mval: \u001b[0mdata=/content/dairy-1/data.yaml, weights=['/content/yolov5/runs/train/yolov5s_results/weights/best.pt'], batch_size=32, imgsz=640, conf_thres=0.001, iou_thres=0.6, max_det=300, task=val, device=, workers=8, single_cls=False, augment=False, verbose=False, save_txt=False, save_hybrid=False, save_conf=False, save_json=False, project=runs/val, name=exp, exist_ok=False, half=False, dnn=False\n",
            "YOLOv5 🚀 v7.0-72-g064365d Python-3.8.10 torch-1.13.1+cu116 CUDA:0 (NVIDIA A100-SXM4-40GB, 40536MiB)\n",
            "\n",
            "Fusing layers... \n",
            "Model summary: 157 layers, 7277128 parameters, 0 gradients, 16.6 GFLOPs\n",
            "\u001b[34m\u001b[1mval: \u001b[0mScanning /content/dairy-1/valid/labels.cache... 300 images, 0 backgrounds, 0 corrupt: 100% 300/300 [00:00<?, ?it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:03<00:00,  2.58it/s]\n",
            "                   all        300        485      0.941      0.972       0.99      0.921\n",
            "Speed: 0.3ms pre-process, 2.0ms inference, 1.3ms NMS per image at shape (32, 3, 640, 640)\n",
            "Results saved to \u001b[1mruns/val/exp\u001b[0m\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "%cd /content/yolov5/\n",
        "!python export.py  --weights /content/yolov5/runs/train/yolov5s_results/weights/best.pt  --include tflite"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4CX52CB1exRy",
        "outputId": "d7854477-0368-4bf8-c76b-e6cc03e889c1"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/yolov5\n",
            "\u001b[34m\u001b[1mexport: \u001b[0mdata=data/coco128.yaml, weights=['/content/yolov5/runs/train/yolov5s_results/weights/best.pt'], imgsz=[640, 640], batch_size=1, device=cpu, half=False, inplace=False, keras=False, optimize=False, int8=False, dynamic=False, simplify=False, opset=17, verbose=False, workspace=4, nms=False, agnostic_nms=False, topk_per_class=100, topk_all=100, iou_thres=0.45, conf_thres=0.25, include=['tflite']\n",
            "YOLOv5 🚀 v7.0-72-g064365d Python-3.8.10 torch-1.13.1+cu116 CPU\n",
            "\n",
            "Fusing layers... \n",
            "Model summary: 157 layers, 7277128 parameters, 0 gradients, 16.6 GFLOPs\n",
            "\n",
            "\u001b[34m\u001b[1mPyTorch:\u001b[0m starting from /content/yolov5/runs/train/yolov5s_results/weights/best.pt with output shape (1, 25200, 104) (56.2 MB)\n",
            "2023-02-04 08:58:29.743495: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
            "\n",
            "\u001b[34m\u001b[1mTensorFlow SavedModel:\u001b[0m starting export with tensorflow 2.9.2...\n",
            "\n",
            "                 from  n    params  module                                  arguments                     \n",
            "2023-02-04 08:58:32.672813: E tensorflow/stream_executor/cuda/cuda_driver.cc:271] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected\n",
            "2023-02-04 08:58:32.672882: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: 7c1ade23ffb9\n",
            "2023-02-04 08:58:32.672896: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: 7c1ade23ffb9\n",
            "2023-02-04 08:58:32.672968: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:200] libcuda reported version is: 510.47.3\n",
            "2023-02-04 08:58:32.673007: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:204] kernel reported version is: 510.47.3\n",
            "2023-02-04 08:58:32.673021: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:310] kernel version seems to match DSO: 510.47.3\n",
            "2023-02-04 08:58:32.673449: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F AVX512_VNNI FMA\n",
            "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
            "  0                -1  1      3520  models.common.Conv                      [3, 32, 6, 2, 2]              \n",
            "  1                -1  1     18560  models.common.Conv                      [32, 64, 3, 2]                \n",
            "  2                -1  1     18816  models.common.C3                        [64, 64, 1]                   \n",
            "  3                -1  1     73984  models.common.Conv                      [64, 128, 3, 2]               \n",
            "  4                -1  1    115712  models.common.C3                        [128, 128, 2]                 \n",
            "  5                -1  1    295424  models.common.Conv                      [128, 256, 3, 2]              \n",
            "  6                -1  1    625152  models.common.C3                        [256, 256, 3]                 \n",
            "  7                -1  1   1180672  models.common.Conv                      [256, 512, 3, 2]              \n",
            "  8                -1  1   1182720  models.common.C3                        [512, 512, 1]                 \n",
            "  9                -1  1    656896  models.common.SPPF                      [512, 512, 5]                 \n",
            " 10                -1  1    131584  models.common.Conv                      [512, 256, 1, 1]              \n",
            " 11                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 12           [-1, 6]  1         0  models.common.Concat                    [1]                           \n",
            " 13                -1  1    361984  models.common.C3                        [512, 256, 1, False]          \n",
            " 14                -1  1     33024  models.common.Conv                      [256, 128, 1, 1]              \n",
            " 15                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 16           [-1, 4]  1         0  models.common.Concat                    [1]                           \n",
            " 17                -1  1     90880  models.common.C3                        [256, 128, 1, False]          \n",
            " 18                -1  1    147712  models.common.Conv                      [128, 128, 3, 2]              \n",
            " 19          [-1, 14]  1         0  models.common.Concat                    [1]                           \n",
            " 20                -1  1    296448  models.common.C3                        [256, 256, 1, False]          \n",
            " 21                -1  1    590336  models.common.Conv                      [256, 256, 3, 2]              \n",
            " 22          [-1, 10]  1         0  models.common.Concat                    [1]                           \n",
            " 23                -1  1   1182720  models.common.C3                        [512, 512, 1, False]          \n",
            " 24      [17, 20, 23]  1    280488  models.yolo.Detect                      [99, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512], [640, 640]]\n",
            "Model: \"model\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_1 (InputLayer)           [(1, 640, 640, 3)]   0           []                               \n",
            "                                                                                                  \n",
            " tf_conv (TFConv)               (1, 320, 320, 32)    3488        ['input_1[0][0]']                \n",
            "                                                                                                  \n",
            " tf_conv_1 (TFConv)             (1, 160, 160, 64)    18496       ['tf_conv[0][0]']                \n",
            "                                                                                                  \n",
            " tfc3 (TFC3)                    (1, 160, 160, 64)    18624       ['tf_conv_1[0][0]']              \n",
            "                                                                                                  \n",
            " tf_conv_7 (TFConv)             (1, 80, 80, 128)     73856       ['tfc3[0][0]']                   \n",
            "                                                                                                  \n",
            " tfc3_1 (TFC3)                  (1, 80, 80, 128)     115200      ['tf_conv_7[0][0]']              \n",
            "                                                                                                  \n",
            " tf_conv_15 (TFConv)            (1, 40, 40, 256)     295168      ['tfc3_1[0][0]']                 \n",
            "                                                                                                  \n",
            " tfc3_2 (TFC3)                  (1, 40, 40, 256)     623872      ['tf_conv_15[0][0]']             \n",
            "                                                                                                  \n",
            " tf_conv_25 (TFConv)            (1, 20, 20, 512)     1180160     ['tfc3_2[0][0]']                 \n",
            "                                                                                                  \n",
            " tfc3_3 (TFC3)                  (1, 20, 20, 512)     1181184     ['tf_conv_25[0][0]']             \n",
            "                                                                                                  \n",
            " tfsppf (TFSPPF)                (1, 20, 20, 512)     656128      ['tfc3_3[0][0]']                 \n",
            "                                                                                                  \n",
            " tf_conv_33 (TFConv)            (1, 20, 20, 256)     131328      ['tfsppf[0][0]']                 \n",
            "                                                                                                  \n",
            " tf_upsample (TFUpsample)       (1, 40, 40, 256)     0           ['tf_conv_33[0][0]']             \n",
            "                                                                                                  \n",
            " tf_concat (TFConcat)           (1, 40, 40, 512)     0           ['tf_upsample[0][0]',            \n",
            "                                                                  'tfc3_2[0][0]']                 \n",
            "                                                                                                  \n",
            " tfc3_4 (TFC3)                  (1, 40, 40, 256)     361216      ['tf_concat[0][0]']              \n",
            "                                                                                                  \n",
            " tf_conv_39 (TFConv)            (1, 40, 40, 128)     32896       ['tfc3_4[0][0]']                 \n",
            "                                                                                                  \n",
            " tf_upsample_1 (TFUpsample)     (1, 80, 80, 128)     0           ['tf_conv_39[0][0]']             \n",
            "                                                                                                  \n",
            " tf_concat_1 (TFConcat)         (1, 80, 80, 256)     0           ['tf_upsample_1[0][0]',          \n",
            "                                                                  'tfc3_1[0][0]']                 \n",
            "                                                                                                  \n",
            " tfc3_5 (TFC3)                  (1, 80, 80, 128)     90496       ['tf_concat_1[0][0]']            \n",
            "                                                                                                  \n",
            " tf_conv_45 (TFConv)            (1, 40, 40, 128)     147584      ['tfc3_5[0][0]']                 \n",
            "                                                                                                  \n",
            " tf_concat_2 (TFConcat)         (1, 40, 40, 256)     0           ['tf_conv_45[0][0]',             \n",
            "                                                                  'tf_conv_39[0][0]']             \n",
            "                                                                                                  \n",
            " tfc3_6 (TFC3)                  (1, 40, 40, 256)     295680      ['tf_concat_2[0][0]']            \n",
            "                                                                                                  \n",
            " tf_conv_51 (TFConv)            (1, 20, 20, 256)     590080      ['tfc3_6[0][0]']                 \n",
            "                                                                                                  \n",
            " tf_concat_3 (TFConcat)         (1, 20, 20, 512)     0           ['tf_conv_51[0][0]',             \n",
            "                                                                  'tf_conv_33[0][0]']             \n",
            "                                                                                                  \n",
            " tfc3_7 (TFC3)                  (1, 20, 20, 512)     1181184     ['tf_concat_3[0][0]']            \n",
            "                                                                                                  \n",
            " tf_detect (TFDetect)           ((1, 25200, 104),    280488      ['tfc3_5[0][0]',                 \n",
            "                                )                                 'tfc3_6[0][0]',                 \n",
            "                                                                  'tfc3_7[0][0]']                 \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 7,277,128\n",
            "Trainable params: 0\n",
            "Non-trainable params: 7,277,128\n",
            "__________________________________________________________________________________________________\n",
            "2023-02-04 08:58:35.770622: I tensorflow/core/grappler/devices.cc:66] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 0\n",
            "2023-02-04 08:58:35.770954: I tensorflow/core/grappler/clusters/single_machine.cc:358] Starting new session\n",
            "\u001b[34m\u001b[1mTensorFlow SavedModel:\u001b[0m export success ✅ 9.0s, saved as /content/yolov5/runs/train/yolov5s_results/weights/best_saved_model (28.0 MB)\n",
            "\n",
            "\u001b[34m\u001b[1mTensorFlow Lite:\u001b[0m starting export with tensorflow 2.9.2...\n",
            "WARNING:absl:Found untraced functions such as tf_conv_2_layer_call_fn, tf_conv_2_layer_call_and_return_conditional_losses, tf_conv_3_layer_call_fn, tf_conv_3_layer_call_and_return_conditional_losses, tf_conv_4_layer_call_fn while saving (showing 5 of 268). These functions will not be directly callable after loading.\n",
            "2023-02-04 08:59:30.190016: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:362] Ignored output_format.\n",
            "2023-02-04 08:59:30.190112: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:365] Ignored drop_control_dependency.\n",
            "2023-02-04 08:59:30.190984: I tensorflow/cc/saved_model/reader.cc:43] Reading SavedModel from: /tmp/tmpo33hdsbm\n",
            "2023-02-04 08:59:30.229240: I tensorflow/cc/saved_model/reader.cc:81] Reading meta graph with tags { serve }\n",
            "2023-02-04 08:59:30.229318: I tensorflow/cc/saved_model/reader.cc:122] Reading SavedModel debug info (if present) from: /tmp/tmpo33hdsbm\n",
            "2023-02-04 08:59:30.335863: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:354] MLIR V1 optimization pass is not enabled\n",
            "2023-02-04 08:59:30.352108: I tensorflow/cc/saved_model/loader.cc:228] Restoring SavedModel bundle.\n",
            "2023-02-04 08:59:30.636959: I tensorflow/cc/saved_model/loader.cc:212] Running initialization op on SavedModel bundle at path: /tmp/tmpo33hdsbm\n",
            "2023-02-04 08:59:30.788423: I tensorflow/cc/saved_model/loader.cc:301] SavedModel load for tags { serve }; Status: success: OK. Took 597446 microseconds.\n",
            "2023-02-04 08:59:31.219482: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:263] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n",
            "2023-02-04 08:59:31.747433: I tensorflow/compiler/mlir/lite/flatbuffer_export.cc:1972] Estimated count of arithmetic ops: 18.339 G  ops, equivalently 9.169 G  MACs\n",
            "\n",
            "Estimated count of arithmetic ops: 18.339 G  ops, equivalently 9.169 G  MACs\n",
            "\u001b[34m\u001b[1mTensorFlow Lite:\u001b[0m export success ✅ 53.7s, saved as /content/yolov5/runs/train/yolov5s_results/weights/best-fp16.tflite (14.0 MB)\n",
            "\n",
            "Export complete (64.1s)\n",
            "Results saved to \u001b[1m/content/yolov5/runs/train/yolov5s_results/weights\u001b[0m\n",
            "Detect:          python detect.py --weights /content/yolov5/runs/train/yolov5s_results/weights/best-fp16.tflite \n",
            "Validate:        python val.py --weights /content/yolov5/runs/train/yolov5s_results/weights/best-fp16.tflite \n",
            "PyTorch Hub:     model = torch.hub.load('ultralytics/yolov5', 'custom', '/content/yolov5/runs/train/yolov5s_results/weights/best-fp16.tflite')  \n",
            "Visualize:       https://netron.app\n",
            "CPU times: user 530 ms, sys: 98.5 ms, total: 629 ms\n",
            "Wall time: 1min 8s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from roboflow import Roboflow\n",
        "rf = Roboflow(api_key=\"x2g0cc9LVwtb4OuDYTFj\")\n",
        "project = rf.workspace(\"last-s46sz\").project(\"drink-keffs\")\n",
        "dataset = project.version(1).download(\"yolov5\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ug2mpGo6e0n2",
        "outputId": "107a9de6-d978-49a1-f344-0e2fb67e6752"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "loading Roboflow workspace...\n",
            "loading Roboflow project...\n",
            "Downloading Dataset Version Zip in drink-1 to yolov5pytorch: 100% [163552896 / 163552896] bytes\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Extracting Dataset Version Zip to drink-1 in yolov5pytorch:: 100%|██████████| 7212/7212 [00:00<00:00, 7461.07it/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%cat ./drink-1/data.yaml"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D0vwZoR7y_0x",
        "outputId": "86fd0bc3-752b-47b2-8351-966b73707a19"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "names:\n",
            "- \"CJ)새싹보리410ML\"\n",
            "- \"CJ컨디션헛개100ML\"\n",
            "- \"가야복숭아농장1-5L\"\n",
            "- \"가야알로에농장1-5L\"\n",
            "- \"가야토마토농장1-5L\"\n",
            "- \"가야토마토농장180ML\"\n",
            "- \"광동)발효홍삼100ml\"\n",
            "- \"광동)비타500(병)100ML\"\n",
            "- \"광동)알찬콩두유200ML\"\n",
            "- \"광동)옥수수수염차500ML\"\n",
            "- \"광동)힘찬하루헛개차1L\"\n",
            "- \"광동힘찬하루헛개차500ML\"\n",
            "- \"남양엣홈오렌지1-5L\"\n",
            "- \"델몬트허니애플시나몬280ML\"\n",
            "- \"동서TOP심플리스무스블랙240ML\"\n",
            "- \"동서TOP심플리스무스스위트아메리카노240ML\"\n",
            "- \"동아오츠카데미소다자몽250ML\"\n",
            "- \"동아오츠카마신다미네랄워터500ML\"\n",
            "- \"동아포카리스웨트500ML\"\n",
            "- \"동원시스템새벽헛개520ML\"\n",
            "- \"레쓰비그란데라떼500ML\"\n",
            "- \"롯데)델몬트듬뿍사과1L\"\n",
            "- \"롯데)술술풀리는하루작두콩차500ML\"\n",
            "- \"롯데)칸타타콘트라베이스(스위트블랙)500ML\"\n",
            "- \"롯데)콜드비타민플러스(포도)240ML\"\n",
            "- \"롯데)파머스주스바오렌지1L\"\n",
            "- \"롯데고려홍삼180ML\"\n",
            "- \"롯데레쓰비175ML\"\n",
            "- \"롯데마운틴듀355ML\"\n",
            "- \"롯데밀키스1-5L\"\n",
            "- \"롯데밀키스500ML\"\n",
            "- \"롯데사랑초톡톡스파클링330ML\"\n",
            "- \"롯데아이시스200ML\"\n",
            "- \"롯데옥수수수염차500ML\"\n",
            "- \"롯데잘빠진하루우엉차500ML\"\n",
            "- \"롯데칸타타아메리카노200ML\"\n",
            "- \"롯데칸타타카페모카(병)250ML\"\n",
            "- \"롯데클라우드클리어제로(캔)350ML\"\n",
            "- \"롯데핑크퐁오렌지망고235ML\"\n",
            "- \"롯데핫식스자몽250ML\"\n",
            "- \"루비키스트루비레드자몽1-89L\"\n",
            "- \"미닛메이드오리지널오렌지1-5L\"\n",
            "- \"빙그레)맑은하늘도라지차500ML\"\n",
            "- \"슈웹스레몬토닉350ML\"\n",
            "- \"신어산음료일화광천수500ML\"\n",
            "- \"썬키스트머스캣1-5L\"\n",
            "- \"썬키스트후레쉬알로에1L\"\n",
            "- \"썬키스트훼미리망고2L\"\n",
            "- \"썬키스트훼미리오렌지180ML\"\n",
            "- \"썬키스트훼미리제주감귤2L\"\n",
            "- \"웅진가야농장제주감귤농장1-5L\"\n",
            "- \"웅진고칼슘오렌지1-5L\"\n",
            "- \"웅진내사랑사과180ML\"\n",
            "- \"웅진아침햇살180ML\"\n",
            "- \"웅진아침햇살500ML\"\n",
            "- \"웅진자연은내사랑감귤1-5L\"\n",
            "- \"웅진자연은토마토500ML\"\n",
            "- \"웅진초록꿀매실180ML\"\n",
            "- \"웰치스그레이프500ml\"\n",
            "- \"유어스광동어성초500ml\"\n",
            "- \"일화)맥콜500ML\"\n",
            "- \"일화헛개꿀물175ML\"\n",
            "- \"쟈뎅로얄헤이즐넛1-1L\"\n",
            "- \"쟈뎅시그니처까페라떼1-1L\"\n",
            "- \"쟈뎅시그니처아메리카노블랙1-1L\"\n",
            "- \"쟈뎅시그니처아메리카노스위트1-1L\"\n",
            "- \"조지아고티카빈티지블랙390ML\"\n",
            "- \"지에스)맑은샘물1-5L\"\n",
            "- \"카카오닙스차500ML\"\n",
            "- \"칸타타더블드립원두커피900ML\"\n",
            "- \"코카)미닛메이드오렌지1-5L\"\n",
            "- \"코카스프라이트500ML\"\n",
            "- \"코카씨그램플레인350ML\"\n",
            "- \"코카콜라)더탄산애플민트450ML\"\n",
            "- \"코카콜라)코카콜라350ML\"\n",
            "- \"코카콜라)코카콜라오리지날테이스트245ml\"\n",
            "- \"코카콜라)킨사이다185ML\"\n",
            "- \"코카콜라1-5L\"\n",
            "- \"코카콜라355ML\"\n",
            "- \"코카콜라오리지널300ML\"\n",
            "- \"코카콜라제로250ML\"\n",
            "- \"코카콜라조지아고티카빈티지블랙270ML\"\n",
            "- \"코카쿠우포도300ML\"\n",
            "- \"코카환타오렌지1-5L\"\n",
            "- \"코카환타오렌지215ML\"\n",
            "- \"코카환타파인애플1-5L\"\n",
            "- \"코카환타포도250ML\"\n",
            "- \"팔도비락식혜500ML\"\n",
            "- \"팔도뽀로로보리차220ML\"\n",
            "- \"하이트진로)토닉워터300ML\"\n",
            "- \"해태)결명자서리태차350ML\"\n",
            "- \"해태갈배사이다238ML\"\n",
            "- \"해태갈배사이다355ML\"\n",
            "- \"해태레몬에이드350ML\"\n",
            "- \"해태써니텐블라스트포도1-5L\"\n",
            "- \"해태썬키스트망고바나나220ML\"\n",
            "- \"해태썬키스트자몽소다350ML(PET)\"\n",
            "- \"해태음료구론산바몬드오리지날액150ML\"\n",
            "- \"해태후레쉬포도50_1-5L\"\n",
            "nc: 99\n",
            "roboflow:\n",
            "  license: CC BY 4.0\n",
            "  project: drink-keffs\n",
            "  url: https://universe.roboflow.com/last-s46sz/drink-keffs/dataset/1\n",
            "  version: 1\n",
            "  workspace: last-s46sz\n",
            "test: ../test/images\n",
            "train: drink-1/train/images\n",
            "val: drink-1/valid/images\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# train yolov5s on custom data for 100 epochs\n",
        "# time its performance\n",
        "%%time\n",
        "%cd /content/yolov5/\n",
        "!python train.py --img 640 --batch 16 --epochs 70 --data ./drink-1/data.yaml --weights 'yolov5s.pt' --name yolov5s_results  --cache"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "itNdDprqzHOX",
        "outputId": "7dfff53f-0606-4a60-e6de-a1a98ad41630"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/yolov5\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mweights=yolov5s.pt, cfg=, data=./drink-1/data.yaml, hyp=data/hyps/hyp.scratch-low.yaml, epochs=70, 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=yolov5s_results, 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\n",
            "remote: Enumerating objects: 50, done.\u001b[K\n",
            "remote: Counting objects: 100% (50/50), done.\u001b[K\n",
            "remote: Compressing objects: 100% (25/25), done.\u001b[K\n",
            "remote: Total 50 (delta 30), reused 42 (delta 25), pack-reused 0\u001b[K\n",
            "Unpacking objects: 100% (50/50), 10.03 KiB | 604.00 KiB/s, done.\n",
            "From https://github.com/ultralytics/yolov5\n",
            " * [new branch]      docker     -> origin/docker\n",
            "\u001b[34m\u001b[1mgithub: \u001b[0mup to date with https://github.com/ultralytics/yolov5 ✅\n",
            "YOLOv5 🚀 v7.0-72-g064365d Python-3.8.10 torch-1.13.1+cu116 CUDA:0 (NVIDIA A100-SXM4-40GB, 40536MiB)\n",
            "\n",
            "\u001b[34m\u001b[1mhyperparameters: \u001b[0mlr0=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\n",
            "\u001b[34m\u001b[1mClearML: \u001b[0mrun 'pip install clearml' to automatically track, visualize and remotely train YOLOv5 🚀 in ClearML\n",
            "\u001b[34m\u001b[1mComet: \u001b[0mrun 'pip install comet_ml' to automatically track and visualize YOLOv5 🚀 runs in Comet\n",
            "\u001b[34m\u001b[1mTensorBoard: \u001b[0mStart with 'tensorboard --logdir runs/train', view at http://localhost:6006/\n",
            "2023-02-04 09:41:36.004498: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
            "Overriding model.yaml nc=80 with nc=99\n",
            "\n",
            "                 from  n    params  module                                  arguments                     \n",
            "  0                -1  1      3520  models.common.Conv                      [3, 32, 6, 2, 2]              \n",
            "  1                -1  1     18560  models.common.Conv                      [32, 64, 3, 2]                \n",
            "  2                -1  1     18816  models.common.C3                        [64, 64, 1]                   \n",
            "  3                -1  1     73984  models.common.Conv                      [64, 128, 3, 2]               \n",
            "  4                -1  2    115712  models.common.C3                        [128, 128, 2]                 \n",
            "  5                -1  1    295424  models.common.Conv                      [128, 256, 3, 2]              \n",
            "  6                -1  3    625152  models.common.C3                        [256, 256, 3]                 \n",
            "  7                -1  1   1180672  models.common.Conv                      [256, 512, 3, 2]              \n",
            "  8                -1  1   1182720  models.common.C3                        [512, 512, 1]                 \n",
            "  9                -1  1    656896  models.common.SPPF                      [512, 512, 5]                 \n",
            " 10                -1  1    131584  models.common.Conv                      [512, 256, 1, 1]              \n",
            " 11                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 12           [-1, 6]  1         0  models.common.Concat                    [1]                           \n",
            " 13                -1  1    361984  models.common.C3                        [512, 256, 1, False]          \n",
            " 14                -1  1     33024  models.common.Conv                      [256, 128, 1, 1]              \n",
            " 15                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 16           [-1, 4]  1         0  models.common.Concat                    [1]                           \n",
            " 17                -1  1     90880  models.common.C3                        [256, 128, 1, False]          \n",
            " 18                -1  1    147712  models.common.Conv                      [128, 128, 3, 2]              \n",
            " 19          [-1, 14]  1         0  models.common.Concat                    [1]                           \n",
            " 20                -1  1    296448  models.common.C3                        [256, 256, 1, False]          \n",
            " 21                -1  1    590336  models.common.Conv                      [256, 256, 3, 2]              \n",
            " 22          [-1, 10]  1         0  models.common.Concat                    [1]                           \n",
            " 23                -1  1   1182720  models.common.C3                        [512, 512, 1, False]          \n",
            " 24      [17, 20, 23]  1    280488  models.yolo.Detect                      [99, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512]]\n",
            "Model summary: 214 layers, 7286632 parameters, 7286632 gradients, 16.8 GFLOPs\n",
            "\n",
            "Transferred 343/349 items from yolov5s.pt\n",
            "\u001b[34m\u001b[1mAMP: \u001b[0mchecks passed ✅\n",
            "\u001b[34m\u001b[1moptimizer:\u001b[0m SGD(lr=0.01) with parameter groups 57 weight(decay=0.0), 60 weight(decay=0.0005), 60 bias\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), CLAHE(p=0.01, clip_limit=(1, 4.0), tile_grid_size=(8, 8))\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mScanning /content/yolov5/drink-1/train/labels... 3150 images, 0 backgrounds, 0 corrupt: 100% 3150/3150 [00:00<00:00, 8020.80it/s]\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mNew cache created: /content/yolov5/drink-1/train/labels.cache\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mCaching images (3.6GB ram): 100% 3150/3150 [00:02<00:00, 1170.38it/s]\n",
            "\u001b[34m\u001b[1mval: \u001b[0mScanning /content/yolov5/drink-1/valid/labels... 300 images, 0 backgrounds, 0 corrupt: 100% 300/300 [00:00<00:00, 6936.10it/s]\n",
            "\u001b[34m\u001b[1mval: \u001b[0mNew cache created: /content/yolov5/drink-1/valid/labels.cache\n",
            "\u001b[34m\u001b[1mval: \u001b[0mCaching images (0.3GB ram): 100% 300/300 [00:00<00:00, 871.54it/s]\n",
            "\n",
            "\u001b[34m\u001b[1mAutoAnchor: \u001b[0m3.74 anchors/target, 1.000 Best Possible Recall (BPR). Current anchors are a good fit to dataset ✅\n",
            "Plotting labels to runs/train/yolov5s_results2/labels.jpg... \n",
            "Image sizes 640 train, 640 val\n",
            "Using 8 dataloader workers\n",
            "Logging results to \u001b[1mruns/train/yolov5s_results2\u001b[0m\n",
            "Starting training for 70 epochs...\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "       0/69      3.63G    0.06514     0.0635     0.1112        152        640: 100% 197/197 [00:22<00:00,  8.73it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.51it/s]\n",
            "                   all        300        536    0.00594      0.876     0.0214     0.0131\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "       1/69      4.53G     0.0383    0.04489      0.108        114        640: 100% 197/197 [00:18<00:00, 10.54it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  6.07it/s]\n",
            "                   all        300        536     0.0077       0.93     0.0255     0.0122\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "       2/69      4.53G    0.03329    0.04119     0.1066        156        640: 100% 197/197 [00:17<00:00, 10.98it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  6.02it/s]\n",
            "                   all        300        536    0.00953      0.967     0.0306      0.019\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "       3/69      4.53G    0.02911    0.03843      0.106        144        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.89it/s]\n",
            "                   all        300        536    0.00727      0.897     0.0259     0.0162\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "       4/69      4.53G    0.02567    0.03622     0.1056        142        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.64it/s]\n",
            "                   all        300        536    0.00975      0.979     0.0327     0.0267\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "       5/69      4.53G    0.02373    0.03482     0.1054        180        640: 100% 197/197 [00:18<00:00, 10.82it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.73it/s]\n",
            "                   all        300        536     0.0101      0.978     0.0408     0.0317\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "       6/69      4.53G    0.02189    0.03417     0.1051        170        640: 100% 197/197 [00:18<00:00, 10.89it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.64it/s]\n",
            "                   all        300        536     0.0101      0.988     0.0433     0.0333\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "       7/69      4.53G    0.02109    0.03271     0.1049        132        640: 100% 197/197 [00:17<00:00, 11.21it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.59it/s]\n",
            "                   all        300        536    0.00971      0.973       0.04     0.0331\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "       8/69      4.53G    0.01984     0.0325     0.1048        151        640: 100% 197/197 [00:17<00:00, 10.96it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.70it/s]\n",
            "                   all        300        536       0.01      0.993     0.0402     0.0336\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "       9/69      4.53G    0.01956    0.03122     0.1047        144        640: 100% 197/197 [00:17<00:00, 11.10it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.76it/s]\n",
            "                   all        300        536     0.0102      0.979     0.0425     0.0354\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      10/69      4.53G    0.01841    0.03057     0.1043        133        640: 100% 197/197 [00:17<00:00, 11.16it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.65it/s]\n",
            "                   all        300        536     0.0102      0.982     0.0491     0.0396\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      11/69      4.53G     0.0185    0.02987      0.104        138        640: 100% 197/197 [00:17<00:00, 11.30it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.62it/s]\n",
            "                   all        300        536       0.01      0.992       0.05     0.0444\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      12/69      4.53G    0.01794    0.02963     0.1031        163        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.63it/s]\n",
            "                   all        300        536       0.01      0.992     0.0581      0.052\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      13/69      4.53G    0.01764    0.02902     0.1021        163        640: 100% 197/197 [00:18<00:00, 10.91it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.57it/s]\n",
            "                   all        300        536      0.017      0.796     0.0699     0.0615\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      14/69      4.53G    0.01764    0.02903     0.1009        123        640: 100% 197/197 [00:18<00:00, 10.77it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.57it/s]\n",
            "                   all        300        536       0.79     0.0703     0.0848     0.0697\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      15/69      4.53G    0.01744    0.02904    0.09929        162        640: 100% 197/197 [00:17<00:00, 11.09it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.42it/s]\n",
            "                   all        300        536      0.816      0.092     0.0959     0.0859\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      16/69      4.53G    0.01786     0.0281    0.09735        140        640: 100% 197/197 [00:17<00:00, 10.99it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.56it/s]\n",
            "                   all        300        536      0.777      0.155      0.136      0.118\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      17/69      4.53G    0.01774    0.02803    0.09513        178        640: 100% 197/197 [00:18<00:00, 10.91it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.61it/s]\n",
            "                   all        300        536      0.607      0.207      0.163      0.138\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      18/69      4.53G    0.01808    0.02789     0.0924        111        640: 100% 197/197 [00:17<00:00, 11.09it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.53it/s]\n",
            "                   all        300        536      0.618       0.21      0.199      0.173\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      19/69      4.53G    0.01819    0.02841    0.08949        172        640: 100% 197/197 [00:17<00:00, 11.01it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.48it/s]\n",
            "                   all        300        536      0.584      0.301      0.247      0.213\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      20/69      4.53G    0.01863    0.02798    0.08708        168        640: 100% 197/197 [00:17<00:00, 11.08it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.47it/s]\n",
            "                   all        300        536      0.599       0.28      0.263      0.222\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      21/69      4.53G    0.01877     0.0276    0.08388        129        640: 100% 197/197 [00:18<00:00, 10.92it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.53it/s]\n",
            "                   all        300        536      0.601      0.341      0.309      0.266\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      22/69      4.53G    0.01855     0.0273    0.08097        154        640: 100% 197/197 [00:17<00:00, 11.02it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.70it/s]\n",
            "                   all        300        536      0.472      0.412      0.354      0.298\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      23/69      4.53G    0.01904    0.02789    0.07844        147        640: 100% 197/197 [00:17<00:00, 10.97it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.62it/s]\n",
            "                   all        300        536      0.519      0.402      0.373      0.334\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      24/69      4.53G    0.01868    0.02767    0.07597        148        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.72it/s]\n",
            "                   all        300        536      0.412      0.576      0.464      0.399\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      25/69      4.53G    0.01863    0.02717    0.07253        150        640: 100% 197/197 [00:18<00:00, 10.94it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.54it/s]\n",
            "                   all        300        536      0.414      0.553      0.479      0.424\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      26/69      4.53G    0.01909      0.027    0.06971        139        640: 100% 197/197 [00:17<00:00, 11.16it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.57it/s]\n",
            "                   all        300        536      0.461      0.549      0.509      0.455\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      27/69      4.53G    0.01868    0.02663    0.06737        135        640: 100% 197/197 [00:17<00:00, 11.00it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.71it/s]\n",
            "                   all        300        536      0.494      0.639      0.601      0.535\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      28/69      4.53G    0.01895    0.02681    0.06504        145        640: 100% 197/197 [00:17<00:00, 11.03it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.68it/s]\n",
            "                   all        300        536      0.536      0.641      0.664      0.598\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      29/69      4.53G     0.0192    0.02683    0.06292        134        640: 100% 197/197 [00:18<00:00, 10.93it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.66it/s]\n",
            "                   all        300        536      0.535      0.641      0.657      0.591\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      30/69      4.53G    0.01919    0.02668    0.06042        161        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.57it/s]\n",
            "                   all        300        536      0.584      0.727      0.708      0.646\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      31/69      4.53G    0.01948    0.02681    0.05823        143        640: 100% 197/197 [00:17<00:00, 10.97it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.64it/s]\n",
            "                   all        300        536      0.602      0.718      0.739      0.663\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      32/69      4.53G    0.01885    0.02662    0.05548        125        640: 100% 197/197 [00:18<00:00, 10.87it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.63it/s]\n",
            "                   all        300        536        0.6      0.748      0.765      0.679\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      33/69      4.53G    0.01894    0.02626    0.05297        151        640: 100% 197/197 [00:18<00:00, 10.79it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.72it/s]\n",
            "                   all        300        536      0.679      0.724      0.775      0.689\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      34/69      4.53G    0.01909    0.02645    0.05093        149        640: 100% 197/197 [00:17<00:00, 11.00it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.15it/s]\n",
            "                   all        300        536      0.672      0.758        0.8       0.73\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      35/69      4.53G    0.01894    0.02572    0.04894        115        640: 100% 197/197 [00:17<00:00, 11.06it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.48it/s]\n",
            "                   all        300        536       0.66      0.805      0.819      0.745\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      36/69      4.53G    0.01926    0.02596    0.04669        139        640: 100% 197/197 [00:17<00:00, 10.99it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.68it/s]\n",
            "                   all        300        536       0.73      0.815      0.871      0.786\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      37/69      4.53G    0.01885    0.02556    0.04536        143        640: 100% 197/197 [00:17<00:00, 11.12it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.70it/s]\n",
            "                   all        300        536      0.728      0.832      0.858      0.789\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      38/69      4.53G    0.01917    0.02584    0.04384        151        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.30it/s]\n",
            "                   all        300        536      0.789      0.807      0.874      0.789\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      39/69      4.53G    0.01912    0.02578    0.04244        138        640: 100% 197/197 [00:18<00:00, 10.85it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.69it/s]\n",
            "                   all        300        536      0.771      0.808      0.875      0.801\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      40/69      4.53G    0.01869    0.02565    0.04049        134        640: 100% 197/197 [00:17<00:00, 11.02it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.55it/s]\n",
            "                   all        300        536      0.755      0.894      0.903      0.827\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      41/69      4.53G    0.01868    0.02537    0.03931        119        640: 100% 197/197 [00:18<00:00, 10.71it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.63it/s]\n",
            "                   all        300        536        0.8      0.874       0.92      0.834\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      42/69      4.53G    0.01838    0.02502    0.03766        135        640: 100% 197/197 [00:18<00:00, 10.83it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.65it/s]\n",
            "                   all        300        536      0.766      0.898      0.917      0.842\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      43/69      4.53G    0.01823    0.02538     0.0368        176        640: 100% 197/197 [00:18<00:00, 10.93it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.77it/s]\n",
            "                   all        300        536      0.827      0.864      0.922      0.843\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      44/69      4.53G    0.01842    0.02515    0.03549        118        640: 100% 197/197 [00:18<00:00, 10.92it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.62it/s]\n",
            "                   all        300        536      0.834      0.888      0.931      0.862\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      45/69      4.53G    0.01796    0.02478    0.03478        163        640: 100% 197/197 [00:17<00:00, 11.15it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.68it/s]\n",
            "                   all        300        536      0.836       0.91      0.947      0.868\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      46/69      4.53G    0.01795    0.02493    0.03342        159        640: 100% 197/197 [00:18<00:00, 10.87it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.77it/s]\n",
            "                   all        300        536       0.81      0.916      0.943      0.868\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      47/69      4.53G     0.0176    0.02456     0.0328        131        640: 100% 197/197 [00:17<00:00, 11.16it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.70it/s]\n",
            "                   all        300        536      0.788       0.93      0.945       0.88\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      48/69      4.53G    0.01731    0.02433    0.03169        148        640: 100% 197/197 [00:18<00:00, 10.94it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.76it/s]\n",
            "                   all        300        536      0.859      0.895      0.949      0.878\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      49/69      4.53G     0.0172    0.02418    0.03092        142        640: 100% 197/197 [00:18<00:00, 10.90it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.62it/s]\n",
            "                   all        300        536      0.823      0.927      0.954      0.889\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      50/69      4.53G    0.01744    0.02419    0.02999        156        640: 100% 197/197 [00:17<00:00, 11.11it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.75it/s]\n",
            "                   all        300        536       0.88       0.88      0.946      0.884\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      51/69      4.53G    0.01675    0.02378    0.02868        114        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.68it/s]\n",
            "                   all        300        536      0.823      0.926      0.954      0.893\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      52/69      4.53G    0.01675    0.02365    0.02855        164        640: 100% 197/197 [00:17<00:00, 10.96it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.64it/s]\n",
            "                   all        300        536      0.847      0.939      0.961      0.897\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      53/69      4.53G    0.01644    0.02334    0.02819        152        640: 100% 197/197 [00:17<00:00, 11.00it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.74it/s]\n",
            "                   all        300        536      0.882      0.904      0.961      0.898\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      54/69      4.53G    0.01625    0.02359    0.02701        119        640: 100% 197/197 [00:17<00:00, 11.24it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.64it/s]\n",
            "                   all        300        536      0.884      0.909      0.962      0.897\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      55/69      4.53G    0.01587    0.02299    0.02599        122        640: 100% 197/197 [00:17<00:00, 11.01it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.82it/s]\n",
            "                   all        300        536      0.894      0.935      0.966      0.901\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      56/69      4.53G    0.01574    0.02328    0.02482        183        640: 100% 197/197 [00:18<00:00, 10.87it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.75it/s]\n",
            "                   all        300        536      0.866      0.943      0.969      0.908\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      57/69      4.53G    0.01554    0.02304    0.02413        162        640: 100% 197/197 [00:18<00:00, 10.74it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.81it/s]\n",
            "                   all        300        536      0.874      0.925      0.964      0.897\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      58/69      4.53G    0.01504    0.02272    0.02463        119        640: 100% 197/197 [00:18<00:00, 10.89it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.73it/s]\n",
            "                   all        300        536      0.828      0.955      0.963      0.902\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      59/69      4.53G    0.01455    0.02242    0.02299        174        640: 100% 197/197 [00:18<00:00, 10.94it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.75it/s]\n",
            "                   all        300        536      0.895       0.93       0.97      0.918\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      60/69      4.53G    0.01453    0.02171     0.0237        142        640: 100% 197/197 [00:18<00:00, 10.74it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.73it/s]\n",
            "                   all        300        536      0.876      0.942      0.966       0.91\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      61/69      4.53G    0.01403    0.02208    0.02288        138        640: 100% 197/197 [00:18<00:00, 10.83it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.82it/s]\n",
            "                   all        300        536      0.867      0.939      0.969      0.915\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      62/69      4.53G    0.01369    0.02165    0.02173        118        640: 100% 197/197 [00:17<00:00, 11.06it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.77it/s]\n",
            "                   all        300        536      0.864      0.956      0.972      0.913\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      63/69      4.53G    0.01311    0.02129    0.02128        181        640: 100% 197/197 [00:17<00:00, 11.02it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.67it/s]\n",
            "                   all        300        536      0.889      0.945      0.974      0.923\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      64/69      4.53G     0.0128    0.02112    0.02085        139        640: 100% 197/197 [00:18<00:00, 10.93it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.79it/s]\n",
            "                   all        300        536      0.914       0.93      0.977      0.921\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      65/69      4.53G    0.01265     0.0213    0.02104        152        640: 100% 197/197 [00:17<00:00, 10.95it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.74it/s]\n",
            "                   all        300        536      0.895      0.933      0.974      0.922\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      66/69      4.53G    0.01211    0.02077    0.02035        138        640: 100% 197/197 [00:18<00:00, 10.93it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.66it/s]\n",
            "                   all        300        536      0.872      0.958       0.97      0.921\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      67/69      4.53G    0.01213    0.02054    0.01987        193        640: 100% 197/197 [00:18<00:00, 10.85it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.77it/s]\n",
            "                   all        300        536      0.873      0.961      0.976      0.931\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      68/69      4.53G    0.01157    0.02024    0.01909        164        640: 100% 197/197 [00:17<00:00, 11.03it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.73it/s]\n",
            "                   all        300        536      0.886      0.951       0.97      0.923\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "      69/69      4.53G    0.01138    0.02011    0.01952        139        640: 100% 197/197 [00:17<00:00, 11.01it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.78it/s]\n",
            "                   all        300        536      0.896      0.946      0.975      0.931\n",
            "\n",
            "70 epochs completed in 0.396 hours.\n",
            "Optimizer stripped from runs/train/yolov5s_results2/weights/last.pt, 15.0MB\n",
            "Optimizer stripped from runs/train/yolov5s_results2/weights/best.pt, 15.0MB\n",
            "\n",
            "Validating runs/train/yolov5s_results2/weights/best.pt...\n",
            "Fusing layers... \n",
            "Model summary: 157 layers, 7277128 parameters, 0 gradients, 16.6 GFLOPs\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:03<00:00,  3.21it/s]\n",
            "                   all        300        536      0.873      0.961      0.976      0.931\n",
            "          CJ)새싹보리410ML        300          2      0.799          1      0.995      0.945\n",
            "          CJ컨디션헛개100ML        300         16      0.974          1      0.995      0.956\n",
            "           가야복숭아농장1-5L        300          3      0.439      0.545      0.624      0.624\n",
            "           가야알로에농장1-5L        300          4      0.896          1      0.995      0.904\n",
            "           가야토마토농장1-5L        300          6      0.926          1      0.995      0.921\n",
            "          가야토마토농장180ML        300          9          1      0.967      0.995      0.937\n",
            "          광동)발효홍삼100ml        300          7       0.94          1      0.995      0.819\n",
            "      광동)비타500(병)100ML        300         11          1       0.89      0.995      0.895\n",
            "         광동)알찬콩두유200ML        300          5      0.897          1      0.995      0.924\n",
            "        광동)옥수수수염차500ML        300          1      0.719          1      0.995      0.995\n",
            "          광동)힘찬하루헛개차1L        300          6      0.918          1      0.995      0.931\n",
            "        광동힘찬하루헛개차500ML        300         10      0.945          1      0.995      0.995\n",
            "           남양엣홈오렌지1-5L        300          5          1      0.466      0.995       0.95\n",
            "       델몬트허니애플시나몬280ML        300          6      0.923          1      0.995      0.995\n",
            "    동서TOP심플리스무스블랙240ML        300          4      0.876          1      0.995      0.995\n",
            "동서TOP심플리스무스스위트아메리카노240ML        300          2      0.545          1      0.995      0.995\n",
            "      동아오츠카데미소다자몽250ML        300          6       0.92          1      0.995      0.995\n",
            "    동아오츠카마신다미네랄워터500ML        300          9      0.949          1      0.995      0.918\n",
            "         동아포카리스웨트500ML        300          1      0.852          1      0.995      0.895\n",
            "        동원시스템새벽헛개520ML        300         10      0.882          1      0.995      0.945\n",
            "         레쓰비그란데라떼500ML        300          7      0.965          1      0.995      0.981\n",
            "          롯데)델몬트듬뿍사과1L        300          7      0.925          1      0.995       0.91\n",
            "   롯데)술술풀리는하루작두콩차500ML        300          6      0.903          1      0.995      0.971\n",
            "롯데)칸타타콘트라베이스(스위트블랙)500ML        300          6      0.963        0.5      0.692      0.603\n",
            "  롯데)콜드비타민플러스(포도)240ML        300          8      0.935          1      0.995      0.983\n",
            "        롯데)파머스주스바오렌지1L        300          8      0.947          1      0.995       0.93\n",
            "           롯데고려홍삼180ML        300          6       0.89          1      0.995      0.995\n",
            "            롯데레쓰비175ML        300          9          1      0.871      0.984      0.967\n",
            "           롯데마운틴듀355ML        300          5      0.883          1      0.995      0.967\n",
            "             롯데밀키스1-5L        300          6      0.922          1      0.995      0.962\n",
            "            롯데밀키스500ML        300          1      0.743          1      0.995      0.995\n",
            "      롯데사랑초톡톡스파클링330ML        300          7          1      0.991      0.995      0.928\n",
            "           롯데아이시스200ML        300          3      0.841          1      0.995      0.995\n",
            "         롯데옥수수수염차500ML        300          7      0.869      0.857      0.857      0.857\n",
            "       롯데잘빠진하루우엉차500ML        300          6      0.886          1      0.995      0.946\n",
            "       롯데칸타타아메리카노200ML        300          8      0.931          1      0.995       0.91\n",
            "     롯데칸타타카페모카(병)250ML        300          1      0.623          1      0.995      0.995\n",
            "   롯데클라우드클리어제로(캔)350ML        300          4      0.879          1      0.995      0.952\n",
            "       롯데핑크퐁오렌지망고235ML        300          2       0.79          1      0.995      0.895\n",
            "          롯데핫식스자몽250ML        300          8      0.818          1      0.982      0.946\n",
            "      루비키스트루비레드자몽1-89L        300          4      0.881          1      0.995      0.958\n",
            "      미닛메이드오리지널오렌지1-5L        300          1      0.691          1      0.995      0.895\n",
            "     빙그레)맑은하늘도라지차500ML        300          4      0.863          1      0.995      0.995\n",
            "       신어산음료일화광천수500ML        300          4      0.863          1      0.995      0.995\n",
            "           썬키스트머스캣1-5L        300          4      0.913          1      0.995      0.852\n",
            "          썬키스트후레쉬알로에1L        300          7      0.925          1      0.995      0.967\n",
            "           썬키스트훼미리망고2L        300          6       0.65          1      0.927      0.927\n",
            "       썬키스트훼미리오렌지180ML        300          6       0.89          1      0.995      0.946\n",
            "         썬키스트훼미리제주감귤2L        300          3      0.333          1      0.913       0.83\n",
            "      웅진가야농장제주감귤농장1-5L        300          5      0.823          1      0.995       0.95\n",
            "          웅진고칼슘오렌지1-5L        300          6      0.898          1      0.995      0.943\n",
            "          웅진내사랑사과180ML        300          5      0.997          1      0.995      0.941\n",
            "           웅진아침햇살180ML        300          8      0.966          1      0.995      0.995\n",
            "           웅진아침햇살500ML        300          5      0.911          1      0.995      0.975\n",
            "        웅진자연은내사랑감귤1-5L        300          4      0.911          1      0.995      0.945\n",
            "         웅진자연은토마토500ML        300          3      0.847          1      0.995      0.951\n",
            "          웅진초록꿀매실180ML        300          3      0.821          1      0.995      0.995\n",
            "          웰치스그레이프500ml        300          6       0.93          1      0.995       0.88\n",
            "         유어스광동어성초500ml        300          7      0.923          1      0.995      0.979\n",
            "            일화)맥콜500ML        300          7      0.915          1      0.995      0.923\n",
            "           일화헛개꿀물175ML        300          5      0.906          1      0.995      0.967\n",
            "          쟈뎅로얄헤이즐넛1-1L        300          2      0.816          1      0.995      0.945\n",
            "        쟈뎅시그니처까페라떼1-1L        300          9      0.947          1      0.995      0.959\n",
            "     쟈뎅시그니처아메리카노블랙1-1L        300          7      0.907          1      0.995      0.977\n",
            "    쟈뎅시그니처아메리카노스위트1-1L        300         10          1      0.972      0.995      0.974\n",
            "      조지아고티카빈티지블랙390ML        300          3      0.885      0.667      0.863      0.777\n",
            "          지에스)맑은샘물1-5L        300          8      0.938          1      0.995      0.865\n",
            "           카카오닙스차500ML        300          1      0.712          1      0.995      0.995\n",
            "      칸타타더블드립원두커피900ML        300          4      0.976          1      0.995       0.87\n",
            "       코카)미닛메이드오렌지1-5L        300          4      0.861          1      0.995      0.995\n",
            "          코카스프라이트500ML        300          5      0.952          1      0.995      0.876\n",
            "         코카씨그램플레인350ML        300          5      0.914          1      0.995      0.934\n",
            "     코카콜라)더탄산애플민트450ML        300          5      0.996          1      0.995      0.836\n",
            "        코카콜라)코카콜라350ML        300          5      0.481        0.6      0.767       0.72\n",
            "코카콜라)코카콜라오리지날테이스트245ml        300          3      0.446      0.819      0.608      0.592\n",
            "        코카콜라)킨사이다185ML        300         11      0.978          1      0.995      0.986\n",
            "              코카콜라1-5L        300          8      0.958          1      0.995      0.957\n",
            "             코카콜라355ML        300          5          1      0.269      0.928      0.849\n",
            "         코카콜라오리지널300ML        300          2       0.82          1      0.995      0.995\n",
            "  코카콜라조지아고티카빈티지블랙270ML        300          9      0.888          1      0.995      0.965\n",
            "           코카쿠우포도300ML        300          2      0.781          1      0.995       0.92\n",
            "           코카환타오렌지1-5L        300         11      0.976          1      0.995      0.933\n",
            "          코카환타오렌지215ML        300          5      0.928          1      0.995      0.975\n",
            "          코카환타파인애플1-5L        300          5      0.903          1      0.995      0.931\n",
            "           코카환타포도250ML        300          5      0.876          1      0.995      0.967\n",
            "           팔도비락식혜500ML        300          3      0.845          1      0.995       0.94\n",
            "         팔도뽀로로보리차220ML        300          6      0.937          1      0.995      0.965\n",
            "       하이트진로)토닉워터300ML        300          9      0.962      0.889       0.94      0.838\n",
            "       해태)결명자서리태차350ML        300          9      0.935          1      0.995      0.976\n",
            "          해태갈배사이다238ML        300          1      0.806          1      0.995      0.995\n",
            "          해태갈배사이다355ML        300          1      0.928          1      0.995      0.995\n",
            "          해태레몬에이드350ML        300          8      0.926          1      0.995      0.917\n",
            "       해태써니텐블라스트포도1-5L        300          3      0.807          1      0.995      0.995\n",
            "  해태썬키스트자몽소다350ML(PET)        300          6      0.917          1      0.995      0.995\n",
            "  해태음료구론산바몬드오리지날액150ML        300         11      0.961          1      0.995      0.923\n",
            "        해태후레쉬포도50_1-5L        300          4      0.894          1      0.995      0.995\n",
            "Results saved to \u001b[1mruns/train/yolov5s_results2\u001b[0m\n",
            "CPU times: user 21.9 s, sys: 3.37 s, total: 25.2 s\n",
            "Wall time: 24min 21s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# train yolov5s on custom data for 100 epochs\n",
        "# time its performance\n",
        "%%time\n",
        "%cd /content/yolov5/\n",
        "!python train.py --img 640 --batch 16 --epochs 10 --data ./drink-1/data.yaml --weights /content/yolov5/runs/train/yolov5s_results2/weights/best.pt --name yolov5s_results  --cache"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UdMNsbAU52HF",
        "outputId": "a659c419-18be-43b4-f26d-984ffd361832"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/yolov5\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mweights=/content/yolov5/runs/train/yolov5s_results2/weights/best.pt, cfg=, data=./drink-1/data.yaml, hyp=data/hyps/hyp.scratch-low.yaml, epochs=10, 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=yolov5s_results, 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\n",
            "\u001b[34m\u001b[1mgithub: \u001b[0mup to date with https://github.com/ultralytics/yolov5 ✅\n",
            "YOLOv5 🚀 v7.0-72-g064365d Python-3.8.10 torch-1.13.1+cu116 CUDA:0 (NVIDIA A100-SXM4-40GB, 40536MiB)\n",
            "\n",
            "\u001b[34m\u001b[1mhyperparameters: \u001b[0mlr0=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\n",
            "\u001b[34m\u001b[1mClearML: \u001b[0mrun 'pip install clearml' to automatically track, visualize and remotely train YOLOv5 🚀 in ClearML\n",
            "\u001b[34m\u001b[1mComet: \u001b[0mrun 'pip install comet_ml' to automatically track and visualize YOLOv5 🚀 runs in Comet\n",
            "\u001b[34m\u001b[1mTensorBoard: \u001b[0mStart with 'tensorboard --logdir runs/train', view at http://localhost:6006/\n",
            "2023-02-04 10:10:25.360495: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
            "\n",
            "                 from  n    params  module                                  arguments                     \n",
            "  0                -1  1      3520  models.common.Conv                      [3, 32, 6, 2, 2]              \n",
            "  1                -1  1     18560  models.common.Conv                      [32, 64, 3, 2]                \n",
            "  2                -1  1     18816  models.common.C3                        [64, 64, 1]                   \n",
            "  3                -1  1     73984  models.common.Conv                      [64, 128, 3, 2]               \n",
            "  4                -1  2    115712  models.common.C3                        [128, 128, 2]                 \n",
            "  5                -1  1    295424  models.common.Conv                      [128, 256, 3, 2]              \n",
            "  6                -1  3    625152  models.common.C3                        [256, 256, 3]                 \n",
            "  7                -1  1   1180672  models.common.Conv                      [256, 512, 3, 2]              \n",
            "  8                -1  1   1182720  models.common.C3                        [512, 512, 1]                 \n",
            "  9                -1  1    656896  models.common.SPPF                      [512, 512, 5]                 \n",
            " 10                -1  1    131584  models.common.Conv                      [512, 256, 1, 1]              \n",
            " 11                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 12           [-1, 6]  1         0  models.common.Concat                    [1]                           \n",
            " 13                -1  1    361984  models.common.C3                        [512, 256, 1, False]          \n",
            " 14                -1  1     33024  models.common.Conv                      [256, 128, 1, 1]              \n",
            " 15                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 16           [-1, 4]  1         0  models.common.Concat                    [1]                           \n",
            " 17                -1  1     90880  models.common.C3                        [256, 128, 1, False]          \n",
            " 18                -1  1    147712  models.common.Conv                      [128, 128, 3, 2]              \n",
            " 19          [-1, 14]  1         0  models.common.Concat                    [1]                           \n",
            " 20                -1  1    296448  models.common.C3                        [256, 256, 1, False]          \n",
            " 21                -1  1    590336  models.common.Conv                      [256, 256, 3, 2]              \n",
            " 22          [-1, 10]  1         0  models.common.Concat                    [1]                           \n",
            " 23                -1  1   1182720  models.common.C3                        [512, 512, 1, False]          \n",
            " 24      [17, 20, 23]  1    280488  models.yolo.Detect                      [99, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512]]\n",
            "Model summary: 214 layers, 7286632 parameters, 7286632 gradients, 16.8 GFLOPs\n",
            "\n",
            "Transferred 349/349 items from /content/yolov5/runs/train/yolov5s_results2/weights/best.pt\n",
            "\u001b[34m\u001b[1mAMP: \u001b[0mchecks passed ✅\n",
            "\u001b[34m\u001b[1moptimizer:\u001b[0m SGD(lr=0.01) with parameter groups 57 weight(decay=0.0), 60 weight(decay=0.0005), 60 bias\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), CLAHE(p=0.01, clip_limit=(1, 4.0), tile_grid_size=(8, 8))\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mScanning /content/yolov5/drink-1/train/labels.cache... 3150 images, 0 backgrounds, 0 corrupt: 100% 3150/3150 [00:00<?, ?it/s]\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mCaching images (3.6GB ram): 100% 3150/3150 [00:02<00:00, 1146.59it/s]\n",
            "\u001b[34m\u001b[1mval: \u001b[0mScanning /content/yolov5/drink-1/valid/labels.cache... 300 images, 0 backgrounds, 0 corrupt: 100% 300/300 [00:00<?, ?it/s]\n",
            "\u001b[34m\u001b[1mval: \u001b[0mCaching images (0.3GB ram): 100% 300/300 [00:00<00:00, 386.71it/s]\n",
            "\n",
            "\u001b[34m\u001b[1mAutoAnchor: \u001b[0m3.74 anchors/target, 1.000 Best Possible Recall (BPR). Current anchors are a good fit to dataset ✅\n",
            "Plotting labels to runs/train/yolov5s_results3/labels.jpg... \n",
            "Image sizes 640 train, 640 val\n",
            "Using 8 dataloader workers\n",
            "Logging results to \u001b[1mruns/train/yolov5s_results3\u001b[0m\n",
            "Starting training for 10 epochs...\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "        0/9      3.63G      0.016    0.02285    0.01937        152        640: 100% 197/197 [00:22<00:00,  8.64it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.55it/s]\n",
            "                   all        300        536        0.9      0.948      0.977      0.838\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "        1/9      4.53G    0.02219    0.02534    0.02264        114        640: 100% 197/197 [00:19<00:00, 10.32it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.66it/s]\n",
            "                   all        300        536      0.822      0.931      0.956      0.859\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "        2/9      4.53G    0.02325    0.02675    0.02739        156        640: 100% 197/197 [00:18<00:00, 10.77it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.71it/s]\n",
            "                   all        300        536      0.838      0.934       0.96      0.791\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "        3/9      4.53G    0.02261    0.02696    0.03076        144        640: 100% 197/197 [00:18<00:00, 10.94it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.54it/s]\n",
            "                   all        300        536      0.826      0.911      0.956      0.832\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "        4/9      4.53G     0.0225    0.02702    0.03015        142        640: 100% 197/197 [00:18<00:00, 10.78it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.67it/s]\n",
            "                   all        300        536      0.826      0.888       0.95      0.805\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "        5/9      4.53G    0.02122    0.02672    0.02936        180        640: 100% 197/197 [00:18<00:00, 10.84it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.52it/s]\n",
            "                   all        300        536       0.86      0.949      0.968      0.794\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "        6/9      4.53G    0.01966    0.02629    0.02748        170        640: 100% 197/197 [00:18<00:00, 10.87it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.53it/s]\n",
            "                   all        300        536      0.911      0.919      0.969      0.836\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "        7/9      4.53G    0.01785    0.02461    0.02451        132        640: 100% 197/197 [00:18<00:00, 10.88it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.73it/s]\n",
            "                   all        300        536      0.876      0.949      0.978      0.875\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "        8/9      4.53G    0.01643     0.0239    0.02305        151        640: 100% 197/197 [00:17<00:00, 11.01it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.69it/s]\n",
            "                   all        300        536       0.92       0.95      0.977      0.883\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "        9/9      4.53G    0.01474    0.02258    0.02042        144        640: 100% 197/197 [00:17<00:00, 11.02it/s]\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:01<00:00,  5.77it/s]\n",
            "                   all        300        536      0.918      0.958      0.985      0.909\n",
            "\n",
            "10 epochs completed in 0.058 hours.\n",
            "Optimizer stripped from runs/train/yolov5s_results3/weights/last.pt, 15.0MB\n",
            "Optimizer stripped from runs/train/yolov5s_results3/weights/best.pt, 15.0MB\n",
            "\n",
            "Validating runs/train/yolov5s_results3/weights/best.pt...\n",
            "Fusing layers... \n",
            "Model summary: 157 layers, 7277128 parameters, 0 gradients, 16.6 GFLOPs\n",
            "                 Class     Images  Instances          P          R      mAP50   mAP50-95: 100% 10/10 [00:03<00:00,  3.25it/s]\n",
            "                   all        300        536      0.918      0.958      0.985      0.909\n",
            "          CJ)새싹보리410ML        300          2       0.86          1      0.995      0.995\n",
            "          CJ컨디션헛개100ML        300         16      0.976          1      0.995      0.939\n",
            "           가야복숭아농장1-5L        300          3          1      0.641      0.995       0.84\n",
            "           가야알로에농장1-5L        300          4      0.928          1      0.995      0.837\n",
            "           가야토마토농장1-5L        300          6      0.957          1      0.995      0.879\n",
            "          가야토마토농장180ML        300          9          1      0.981      0.995      0.846\n",
            "          광동)발효홍삼100ml        300          7      0.992          1      0.995       0.87\n",
            "      광동)비타500(병)100ML        300         11          1      0.852      0.995      0.836\n",
            "         광동)알찬콩두유200ML        300          5      0.927          1      0.995      0.931\n",
            "        광동)옥수수수염차500ML        300          1      0.771          1      0.995      0.995\n",
            "          광동)힘찬하루헛개차1L        300          6          1       0.78      0.904      0.826\n",
            "        광동힘찬하루헛개차500ML        300         10          1      0.824      0.995      0.878\n",
            "           남양엣홈오렌지1-5L        300          5      0.932          1      0.995      0.905\n",
            "       델몬트허니애플시나몬280ML        300          6      0.946          1      0.995      0.937\n",
            "    동서TOP심플리스무스블랙240ML        300          4      0.923          1      0.995      0.933\n",
            "동서TOP심플리스무스스위트아메리카노240ML        300          2       0.84          1      0.995      0.995\n",
            "      동아오츠카데미소다자몽250ML        300          6       0.94          1      0.995      0.995\n",
            "    동아오츠카마신다미네랄워터500ML        300          9      0.966          1      0.995       0.88\n",
            "         동아포카리스웨트500ML        300          1      0.807          1      0.995      0.995\n",
            "        동원시스템새벽헛개520ML        300         10      0.936          1      0.995      0.925\n",
            "         레쓰비그란데라떼500ML        300          7      0.994          1      0.995      0.955\n",
            "          롯데)델몬트듬뿍사과1L        300          7      0.981          1      0.995      0.917\n",
            "   롯데)술술풀리는하루작두콩차500ML        300          6       0.94          1      0.995      0.995\n",
            "롯데)칸타타콘트라베이스(스위트블랙)500ML        300          6          1      0.697      0.995      0.825\n",
            "  롯데)콜드비타민플러스(포도)240ML        300          8       0.97          1      0.995      0.953\n",
            "        롯데)파머스주스바오렌지1L        300          8      0.963          1      0.995      0.918\n",
            "           롯데고려홍삼180ML        300          6      0.943          1      0.995      0.918\n",
            "            롯데레쓰비175ML        300          9          1      0.964      0.995      0.947\n",
            "           롯데마운틴듀355ML        300          5      0.908          1      0.995       0.94\n",
            "             롯데밀키스1-5L        300          6      0.948          1      0.995      0.945\n",
            "            롯데밀키스500ML        300          1      0.798          1      0.995      0.995\n",
            "      롯데사랑초톡톡스파클링330ML        300          7      0.975          1      0.995       0.91\n",
            "           롯데아이시스200ML        300          3       0.89          1      0.995      0.995\n",
            "         롯데옥수수수염차500ML        300          7      0.972      0.857      0.856      0.813\n",
            "       롯데잘빠진하루우엉차500ML        300          6      0.896          1      0.995      0.921\n",
            "       롯데칸타타아메리카노200ML        300          8      0.988          1      0.995      0.954\n",
            "     롯데칸타타카페모카(병)250ML        300          1      0.737          1      0.995      0.995\n",
            "   롯데클라우드클리어제로(캔)350ML        300          4      0.935          1      0.995      0.895\n",
            "       롯데핑크퐁오렌지망고235ML        300          2      0.853          1      0.995      0.895\n",
            "          롯데핫식스자몽250ML        300          8      0.854          1      0.954      0.941\n",
            "      루비키스트루비레드자몽1-89L        300          4      0.929          1      0.995      0.902\n",
            "      미닛메이드오리지널오렌지1-5L        300          1      0.764          1      0.995      0.895\n",
            "     빙그레)맑은하늘도라지차500ML        300          4      0.933          1      0.995      0.902\n",
            "       신어산음료일화광천수500ML        300          4      0.917          1      0.995      0.945\n",
            "           썬키스트머스캣1-5L        300          4      0.904          1      0.995      0.864\n",
            "          썬키스트후레쉬알로에1L        300          7      0.954          1      0.995      0.857\n",
            "           썬키스트훼미리망고2L        300          6      0.665          1      0.995      0.939\n",
            "       썬키스트훼미리오렌지180ML        300          6      0.929          1      0.995       0.94\n",
            "         썬키스트훼미리제주감귤2L        300          3      0.531          1      0.913      0.821\n",
            "      웅진가야농장제주감귤농장1-5L        300          5      0.973          1      0.995      0.905\n",
            "          웅진고칼슘오렌지1-5L        300          6      0.926          1      0.995      0.912\n",
            "          웅진내사랑사과180ML        300          5          1      0.841      0.995      0.905\n",
            "           웅진아침햇살180ML        300          8      0.958          1      0.995       0.96\n",
            "           웅진아침햇살500ML        300          5      0.922          1      0.995      0.955\n",
            "        웅진자연은내사랑감귤1-5L        300          4      0.933          1      0.995      0.902\n",
            "         웅진자연은토마토500ML        300          3      0.882          1      0.995      0.907\n",
            "          웅진초록꿀매실180ML        300          3      0.882          1      0.995      0.962\n",
            "          웰치스그레이프500ml        300          6      0.972          1      0.995      0.816\n",
            "         유어스광동어성초500ml        300          7      0.948          1      0.995      0.922\n",
            "            일화)맥콜500ML        300          7      0.945          1      0.995      0.914\n",
            "           일화헛개꿀물175ML        300          5      0.933          1      0.995      0.905\n",
            "          쟈뎅로얄헤이즐넛1-1L        300          2      0.708          1      0.995      0.895\n",
            "        쟈뎅시그니처까페라떼1-1L        300          9          1      0.997      0.995      0.951\n",
            "     쟈뎅시그니처아메리카노블랙1-1L        300          7      0.942          1      0.995      0.882\n",
            "    쟈뎅시그니처아메리카노스위트1-1L        300         10          1      0.969      0.995      0.954\n",
            "      조지아고티카빈티지블랙390ML        300          3      0.801      0.667      0.806      0.731\n",
            "          지에스)맑은샘물1-5L        300          8      0.971          1      0.995      0.827\n",
            "           카카오닙스차500ML        300          1      0.769          1      0.995      0.895\n",
            "      칸타타더블드립원두커피900ML        300          4          1      0.774      0.995      0.777\n",
            "       코카)미닛메이드오렌지1-5L        300          4      0.923          1      0.995      0.952\n",
            "          코카스프라이트500ML        300          5          1      0.879      0.995      0.871\n",
            "         코카씨그램플레인350ML        300          5      0.943          1      0.995      0.897\n",
            "     코카콜라)더탄산애플민트450ML        300          5          1      0.888      0.995      0.836\n",
            "        코카콜라)코카콜라350ML        300          5      0.859        0.6      0.928      0.875\n",
            "코카콜라)코카콜라오리지날테이스트245ml        300          3      0.578      0.927      0.641      0.641\n",
            "        코카콜라)킨사이다185ML        300         11          1      0.951      0.995      0.955\n",
            "              코카콜라1-5L        300          8      0.966          1      0.995      0.851\n",
            "             코카콜라355ML        300          5          1          0      0.995      0.915\n",
            "         코카콜라오리지널300ML        300          2       0.85          1      0.995      0.995\n",
            "  코카콜라조지아고티카빈티지블랙270ML        300          9      0.951          1      0.995      0.917\n",
            "           코카쿠우포도300ML        300          2      0.867          1      0.995      0.895\n",
            "           코카환타오렌지1-5L        300         11      0.969          1      0.995      0.929\n",
            "          코카환타오렌지215ML        300          5      0.938          1      0.995      0.955\n",
            "          코카환타파인애플1-5L        300          5      0.933          1      0.995       0.95\n",
            "           코카환타포도250ML        300          5      0.975          1      0.995      0.975\n",
            "           팔도비락식혜500ML        300          3      0.896          1      0.995      0.863\n",
            "         팔도뽀로로보리차220ML        300          6      0.949          1      0.995      0.901\n",
            "       하이트진로)토닉워터300ML        300          9          1      0.838      0.995      0.848\n",
            "       해태)결명자서리태차350ML        300          9      0.957          1      0.995      0.928\n",
            "          해태갈배사이다238ML        300          1      0.667          1      0.995      0.995\n",
            "          해태갈배사이다355ML        300          1      0.952          1      0.995      0.895\n",
            "          해태레몬에이드350ML        300          8      0.946          1      0.995      0.869\n",
            "       해태써니텐블라스트포도1-5L        300          3      0.881          1      0.995      0.907\n",
            "  해태썬키스트자몽소다350ML(PET)        300          6      0.939          1      0.995      0.976\n",
            "  해태음료구론산바몬드오리지날액150ML        300         11      0.985          1      0.995       0.91\n",
            "        해태후레쉬포도50_1-5L        300          4      0.972          1      0.995      0.958\n",
            "Results saved to \u001b[1mruns/train/yolov5s_results3\u001b[0m\n",
            "CPU times: user 3.46 s, sys: 546 ms, total: 4.01 s\n",
            "Wall time: 4min 4s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "%cd /content/yolov5/\n",
        "!python export.py  --weights /content/yolov5/runs/train/yolov5s_results3/weights/best.pt --include tflite"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cfUCrNZW5qkU",
        "outputId": "4b8e0a29-1b50-4795-bd01-58ccc744382d"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/yolov5\n",
            "\u001b[34m\u001b[1mexport: \u001b[0mdata=data/coco128.yaml, weights=['/content/yolov5/runs/train/yolov5s_results3/weights/best.pt'], imgsz=[640, 640], batch_size=1, device=cpu, half=False, inplace=False, keras=False, optimize=False, int8=False, dynamic=False, simplify=False, opset=17, verbose=False, workspace=4, nms=False, agnostic_nms=False, topk_per_class=100, topk_all=100, iou_thres=0.45, conf_thres=0.25, include=['tflite']\n",
            "YOLOv5 🚀 v7.0-72-g064365d Python-3.8.10 torch-1.13.1+cu116 CPU\n",
            "\n",
            "Fusing layers... \n",
            "Model summary: 157 layers, 7277128 parameters, 0 gradients, 16.6 GFLOPs\n",
            "\n",
            "\u001b[34m\u001b[1mPyTorch:\u001b[0m starting from /content/yolov5/runs/train/yolov5s_results3/weights/best.pt with output shape (1, 25200, 104) (14.3 MB)\n",
            "2023-02-04 10:14:47.929746: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
            "\n",
            "\u001b[34m\u001b[1mTensorFlow SavedModel:\u001b[0m starting export with tensorflow 2.9.2...\n",
            "\n",
            "                 from  n    params  module                                  arguments                     \n",
            "2023-02-04 10:14:50.449889: E tensorflow/stream_executor/cuda/cuda_driver.cc:271] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected\n",
            "2023-02-04 10:14:50.449943: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: 7c1ade23ffb9\n",
            "2023-02-04 10:14:50.449960: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: 7c1ade23ffb9\n",
            "2023-02-04 10:14:50.450019: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:200] libcuda reported version is: 510.47.3\n",
            "2023-02-04 10:14:50.450063: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:204] kernel reported version is: 510.47.3\n",
            "2023-02-04 10:14:50.450078: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:310] kernel version seems to match DSO: 510.47.3\n",
            "2023-02-04 10:14:50.450477: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F AVX512_VNNI FMA\n",
            "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
            "  0                -1  1      3520  models.common.Conv                      [3, 32, 6, 2, 2]              \n",
            "  1                -1  1     18560  models.common.Conv                      [32, 64, 3, 2]                \n",
            "  2                -1  1     18816  models.common.C3                        [64, 64, 1]                   \n",
            "  3                -1  1     73984  models.common.Conv                      [64, 128, 3, 2]               \n",
            "  4                -1  1    115712  models.common.C3                        [128, 128, 2]                 \n",
            "  5                -1  1    295424  models.common.Conv                      [128, 256, 3, 2]              \n",
            "  6                -1  1    625152  models.common.C3                        [256, 256, 3]                 \n",
            "  7                -1  1   1180672  models.common.Conv                      [256, 512, 3, 2]              \n",
            "  8                -1  1   1182720  models.common.C3                        [512, 512, 1]                 \n",
            "  9                -1  1    656896  models.common.SPPF                      [512, 512, 5]                 \n",
            " 10                -1  1    131584  models.common.Conv                      [512, 256, 1, 1]              \n",
            " 11                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 12           [-1, 6]  1         0  models.common.Concat                    [1]                           \n",
            " 13                -1  1    361984  models.common.C3                        [512, 256, 1, False]          \n",
            " 14                -1  1     33024  models.common.Conv                      [256, 128, 1, 1]              \n",
            " 15                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 16           [-1, 4]  1         0  models.common.Concat                    [1]                           \n",
            " 17                -1  1     90880  models.common.C3                        [256, 128, 1, False]          \n",
            " 18                -1  1    147712  models.common.Conv                      [128, 128, 3, 2]              \n",
            " 19          [-1, 14]  1         0  models.common.Concat                    [1]                           \n",
            " 20                -1  1    296448  models.common.C3                        [256, 256, 1, False]          \n",
            " 21                -1  1    590336  models.common.Conv                      [256, 256, 3, 2]              \n",
            " 22          [-1, 10]  1         0  models.common.Concat                    [1]                           \n",
            " 23                -1  1   1182720  models.common.C3                        [512, 512, 1, False]          \n",
            " 24      [17, 20, 23]  1    280488  models.yolo.Detect                      [99, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512], [640, 640]]\n",
            "Model: \"model\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_1 (InputLayer)           [(1, 640, 640, 3)]   0           []                               \n",
            "                                                                                                  \n",
            " tf_conv (TFConv)               (1, 320, 320, 32)    3488        ['input_1[0][0]']                \n",
            "                                                                                                  \n",
            " tf_conv_1 (TFConv)             (1, 160, 160, 64)    18496       ['tf_conv[0][0]']                \n",
            "                                                                                                  \n",
            " tfc3 (TFC3)                    (1, 160, 160, 64)    18624       ['tf_conv_1[0][0]']              \n",
            "                                                                                                  \n",
            " tf_conv_7 (TFConv)             (1, 80, 80, 128)     73856       ['tfc3[0][0]']                   \n",
            "                                                                                                  \n",
            " tfc3_1 (TFC3)                  (1, 80, 80, 128)     115200      ['tf_conv_7[0][0]']              \n",
            "                                                                                                  \n",
            " tf_conv_15 (TFConv)            (1, 40, 40, 256)     295168      ['tfc3_1[0][0]']                 \n",
            "                                                                                                  \n",
            " tfc3_2 (TFC3)                  (1, 40, 40, 256)     623872      ['tf_conv_15[0][0]']             \n",
            "                                                                                                  \n",
            " tf_conv_25 (TFConv)            (1, 20, 20, 512)     1180160     ['tfc3_2[0][0]']                 \n",
            "                                                                                                  \n",
            " tfc3_3 (TFC3)                  (1, 20, 20, 512)     1181184     ['tf_conv_25[0][0]']             \n",
            "                                                                                                  \n",
            " tfsppf (TFSPPF)                (1, 20, 20, 512)     656128      ['tfc3_3[0][0]']                 \n",
            "                                                                                                  \n",
            " tf_conv_33 (TFConv)            (1, 20, 20, 256)     131328      ['tfsppf[0][0]']                 \n",
            "                                                                                                  \n",
            " tf_upsample (TFUpsample)       (1, 40, 40, 256)     0           ['tf_conv_33[0][0]']             \n",
            "                                                                                                  \n",
            " tf_concat (TFConcat)           (1, 40, 40, 512)     0           ['tf_upsample[0][0]',            \n",
            "                                                                  'tfc3_2[0][0]']                 \n",
            "                                                                                                  \n",
            " tfc3_4 (TFC3)                  (1, 40, 40, 256)     361216      ['tf_concat[0][0]']              \n",
            "                                                                                                  \n",
            " tf_conv_39 (TFConv)            (1, 40, 40, 128)     32896       ['tfc3_4[0][0]']                 \n",
            "                                                                                                  \n",
            " tf_upsample_1 (TFUpsample)     (1, 80, 80, 128)     0           ['tf_conv_39[0][0]']             \n",
            "                                                                                                  \n",
            " tf_concat_1 (TFConcat)         (1, 80, 80, 256)     0           ['tf_upsample_1[0][0]',          \n",
            "                                                                  'tfc3_1[0][0]']                 \n",
            "                                                                                                  \n",
            " tfc3_5 (TFC3)                  (1, 80, 80, 128)     90496       ['tf_concat_1[0][0]']            \n",
            "                                                                                                  \n",
            " tf_conv_45 (TFConv)            (1, 40, 40, 128)     147584      ['tfc3_5[0][0]']                 \n",
            "                                                                                                  \n",
            " tf_concat_2 (TFConcat)         (1, 40, 40, 256)     0           ['tf_conv_45[0][0]',             \n",
            "                                                                  'tf_conv_39[0][0]']             \n",
            "                                                                                                  \n",
            " tfc3_6 (TFC3)                  (1, 40, 40, 256)     295680      ['tf_concat_2[0][0]']            \n",
            "                                                                                                  \n",
            " tf_conv_51 (TFConv)            (1, 20, 20, 256)     590080      ['tfc3_6[0][0]']                 \n",
            "                                                                                                  \n",
            " tf_concat_3 (TFConcat)         (1, 20, 20, 512)     0           ['tf_conv_51[0][0]',             \n",
            "                                                                  'tf_conv_33[0][0]']             \n",
            "                                                                                                  \n",
            " tfc3_7 (TFC3)                  (1, 20, 20, 512)     1181184     ['tf_concat_3[0][0]']            \n",
            "                                                                                                  \n",
            " tf_detect (TFDetect)           ((1, 25200, 104),    280488      ['tfc3_5[0][0]',                 \n",
            "                                )                                 'tfc3_6[0][0]',                 \n",
            "                                                                  'tfc3_7[0][0]']                 \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 7,277,128\n",
            "Trainable params: 0\n",
            "Non-trainable params: 7,277,128\n",
            "__________________________________________________________________________________________________\n",
            "2023-02-04 10:14:53.354191: I tensorflow/core/grappler/devices.cc:66] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 0\n",
            "2023-02-04 10:14:53.354413: I tensorflow/core/grappler/clusters/single_machine.cc:358] Starting new session\n",
            "\u001b[34m\u001b[1mTensorFlow SavedModel:\u001b[0m export success ✅ 8.2s, saved as /content/yolov5/runs/train/yolov5s_results3/weights/best_saved_model (28.0 MB)\n",
            "\n",
            "\u001b[34m\u001b[1mTensorFlow Lite:\u001b[0m starting export with tensorflow 2.9.2...\n",
            "WARNING:absl:Found untraced functions such as tf_conv_2_layer_call_fn, tf_conv_2_layer_call_and_return_conditional_losses, tf_conv_3_layer_call_fn, tf_conv_3_layer_call_and_return_conditional_losses, tf_conv_4_layer_call_fn while saving (showing 5 of 268). These functions will not be directly callable after loading.\n",
            "2023-02-04 10:15:47.936385: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:362] Ignored output_format.\n",
            "2023-02-04 10:15:47.936433: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:365] Ignored drop_control_dependency.\n",
            "2023-02-04 10:15:47.937323: I tensorflow/cc/saved_model/reader.cc:43] Reading SavedModel from: /tmp/tmpawg7_j09\n",
            "2023-02-04 10:15:47.974082: I tensorflow/cc/saved_model/reader.cc:81] Reading meta graph with tags { serve }\n",
            "2023-02-04 10:15:47.974151: I tensorflow/cc/saved_model/reader.cc:122] Reading SavedModel debug info (if present) from: /tmp/tmpawg7_j09\n",
            "2023-02-04 10:15:48.080553: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:354] MLIR V1 optimization pass is not enabled\n",
            "2023-02-04 10:15:48.096653: I tensorflow/cc/saved_model/loader.cc:228] Restoring SavedModel bundle.\n",
            "2023-02-04 10:15:48.371019: I tensorflow/cc/saved_model/loader.cc:212] Running initialization op on SavedModel bundle at path: /tmp/tmpawg7_j09\n",
            "2023-02-04 10:15:48.526656: I tensorflow/cc/saved_model/loader.cc:301] SavedModel load for tags { serve }; Status: success: OK. Took 589338 microseconds.\n",
            "2023-02-04 10:15:48.926795: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:263] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n",
            "2023-02-04 10:15:49.411495: I tensorflow/compiler/mlir/lite/flatbuffer_export.cc:1972] Estimated count of arithmetic ops: 18.339 G  ops, equivalently 9.169 G  MACs\n",
            "\n",
            "Estimated count of arithmetic ops: 18.339 G  ops, equivalently 9.169 G  MACs\n",
            "\u001b[34m\u001b[1mTensorFlow Lite:\u001b[0m export success ✅ 53.9s, saved as /content/yolov5/runs/train/yolov5s_results3/weights/best-fp16.tflite (14.0 MB)\n",
            "\n",
            "Export complete (63.4s)\n",
            "Results saved to \u001b[1m/content/yolov5/runs/train/yolov5s_results3/weights\u001b[0m\n",
            "Detect:          python detect.py --weights /content/yolov5/runs/train/yolov5s_results3/weights/best-fp16.tflite \n",
            "Validate:        python val.py --weights /content/yolov5/runs/train/yolov5s_results3/weights/best-fp16.tflite \n",
            "PyTorch Hub:     model = torch.hub.load('ultralytics/yolov5', 'custom', '/content/yolov5/runs/train/yolov5s_results3/weights/best-fp16.tflite')  \n",
            "Visualize:       https://netron.app\n",
            "CPU times: user 426 ms, sys: 85.2 ms, total: 512 ms\n",
            "Wall time: 1min 7s\n"
          ]
        }
      ]
    }
  ]
}