Smart Dressing.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e813d9af",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Setup complete. Using torch 1.9.1+cpu (CPU)\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "from IPython.display import Image, clear_output  # to display images\n",
    "\n",
    "clear_output()\n",
    "print(f\"Setup complete. Using torch {torch.__version__} ({torch.cuda.get_device_properties(0).name if torch.cuda.is_available() else 'CPU'})\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0658fd29",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os \n",
    "import random\n",
    "import shutil\n",
    "from sklearn.model_selection import train_test_split\n",
    "import xml.etree.ElementTree as ET\n",
    "from xml.dom import minidom\n",
    "from tqdm import tqdm\n",
    "from PIL import Image, ImageDraw\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "random.seed(108)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9aac381e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Le volume dans le lecteur C n'a pas de nom.\n",
      " Le num‚ro de s‚rie du volume est D0A2-8479\n",
      "\n",
      " R‚pertoire de C:\\Users\\Ben Zahra\\Desktop\\yolov5\n",
      "\n",
      "21/10/2021  02:59    <DIR>          .\n",
      "21/10/2021  02:59    <DIR>          ..\n",
      "20/10/2021  06:26             3ÿ892 .dockerignore\n",
      "20/10/2021  06:26                77 .gitattributes\n",
      "20/10/2021  06:26    <DIR>          .github\n",
      "20/10/2021  06:26             4ÿ141 .gitignore\n",
      "21/10/2021  02:27    <DIR>          .ipynb_checkpoints\n",
      "21/10/2021  01:50    <DIR>          ClothingPascalVOC\n",
      "20/10/2021  06:26             5ÿ064 CONTRIBUTING.md\n",
      "20/10/2021  06:26    <DIR>          data\n",
      "20/10/2021  06:26            15ÿ737 detect.py\n",
      "20/10/2021  06:26             2ÿ151 Dockerfile\n",
      "20/10/2021  06:26            16ÿ843 export.py\n",
      "20/10/2021  06:26             6ÿ380 hubconf.py\n",
      "20/10/2021  06:26            35ÿ799 LICENSE\n",
      "21/10/2021  01:12    <DIR>          models\n",
      "20/10/2021  06:26            15ÿ056 README.md\n",
      "20/10/2021  06:26               928 requirements.txt\n",
      "21/10/2021  01:12    <DIR>          runs\n",
      "20/10/2021  06:26            32ÿ814 train.py\n",
      "21/10/2021  02:18            44ÿ383 tutorial.ipynb\n",
      "21/10/2021  02:59            10ÿ193 Untitled.ipynb\n",
      "21/10/2021  01:12    <DIR>          utils\n",
      "20/10/2021  06:26            17ÿ452 val.py\n",
      "21/10/2021  01:12        14ÿ698ÿ491 yolov5s.pt\n",
      "              16 fichier(s)       14ÿ909ÿ401 octets\n",
      "               9 R‚p(s)  121ÿ093ÿ918ÿ720 octets libres\n"
     ]
    }
   ],
   "source": [
    "ls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c53e9524",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ben Zahra\\Desktop\\yolov5\\ClothingPascalVOC\n"
     ]
    }
   ],
   "source": [
    "cd ClothingPascalVOC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "86585ad3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Le volume dans le lecteur C n'a pas de nom.\n",
      " Le num‚ro de s‚rie du volume est D0A2-8479\n",
      "\n",
      " R‚pertoire de C:\\Users\\Ben Zahra\\Desktop\\yolov5\\ClothingPascalVOC\n",
      "\n",
      "21/10/2021  01:50    <DIR>          .\n",
      "21/10/2021  01:50    <DIR>          ..\n",
      "21/10/2021  02:58    <DIR>          annotations\n",
      "21/10/2021  01:49    <DIR>          images\n",
      "               0 fichier(s)                0 octets\n",
      "               4 R‚p(s)  121ÿ093ÿ918ÿ720 octets libres\n"
     ]
    }
   ],
   "source": [
    "ls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "31e784b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to get the data from XML Annotation\n",
    "def extract_info_from_xml(xml_file):\n",
    "    root = ET.parse(xml_file).getroot()\n",
    "    \n",
    "    # Initialise the info dict \n",
    "    info_dict = {}\n",
    "    info_dict['bboxes'] = []\n",
    "\n",
    "    # Parse the XML Tree\n",
    "    for elem in root:\n",
    "        # Get the file name \n",
    "        if elem.tag == \"filename\":\n",
    "            info_dict['filename'] = elem.text\n",
    "            \n",
    "        # Get the image size\n",
    "        elif elem.tag == \"size\":\n",
    "            image_size = []\n",
    "            for subelem in elem:\n",
    "                image_size.append(int(subelem.text))\n",
    "            \n",
    "            info_dict['image_size'] = tuple(image_size)\n",
    "        \n",
    "        # Get details of the bounding box \n",
    "        elif elem.tag == \"object\":\n",
    "            bbox = {}\n",
    "            for subelem in elem:\n",
    "                if subelem.tag == \"name\":\n",
    "                    bbox[\"class\"] = subelem.text\n",
    "                    \n",
    "                elif subelem.tag == \"bndbox\":\n",
    "                    for subsubelem in subelem:\n",
    "                        bbox[subsubelem.tag] = int(subsubelem.text)            \n",
    "            info_dict['bboxes'].append(bbox)\n",
    "    \n",
    "    return info_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "bb2d85ea",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'bboxes': [{'class': 'Trousers', 'xmin': 46, 'ymin': 8, 'xmax': 155, 'ymax': 229}], 'filename': 'Clothing6.png', 'image_size': (200, 260, 3)}\n"
     ]
    }
   ],
   "source": [
    "print(extract_info_from_xml('annotations/Clothing6.xml'))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "2ee73281",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dictionary that maps class names to IDs\n",
    "class_name_to_id_mapping = {\"Skirts\": 0,\n",
    "                           \"Shirts\": 1,\n",
    "                           \"Tops\": 2,\n",
    "                           \"sunglasses\": 3,\n",
    "                           \"Trousers\":4,\n",
    "                           \"Jeans\":5,\n",
    "                           \"Tshirts\":6,\n",
    "                           \"Shoes\":7,\n",
    "                           \"Jackets\":8}\n",
    "\n",
    "# Convert the info dict to the required yolo format and write it to disk\n",
    "def convert_to_yolov5(info_dict):\n",
    "    print_buffer = []\n",
    "    \n",
    "    # For each bounding box\n",
    "    for b in info_dict[\"bboxes\"]:\n",
    "        try:\n",
    "            class_id = class_name_to_id_mapping[b[\"class\"]]\n",
    "        except KeyError:\n",
    "            print(\"Invalid Class. Must be one from \", class_name_to_id_mapping.keys())\n",
    "        \n",
    "        # Transform the bbox co-ordinates as per the format required by YOLO v5\n",
    "        b_center_x = (b[\"xmin\"] + b[\"xmax\"]) / 2 \n",
    "        b_center_y = (b[\"ymin\"] + b[\"ymax\"]) / 2\n",
    "        b_width    = (b[\"xmax\"] - b[\"xmin\"])\n",
    "        b_height   = (b[\"ymax\"] - b[\"ymin\"])\n",
    "        \n",
    "        # Normalise the co-ordinates by the dimensions of the image\n",
    "        image_w, image_h, image_c = info_dict[\"image_size\"]  \n",
    "        b_center_x /= image_w \n",
    "        b_center_y /= image_h \n",
    "        b_width    /= image_w \n",
    "        b_height   /= image_h \n",
    "        \n",
    "        #Write the bbox details to the file \n",
    "        print_buffer.append(\"{} {:.3f} {:.3f} {:.3f} {:.3f}\".format(class_id, b_center_x, b_center_y, b_width, b_height))\n",
    "        \n",
    "    # Name of the file which we have to save \n",
    "    save_file_name = os.path.join(\"annotations\", info_dict[\"filename\"].replace(\"png\", \"txt\"))\n",
    "    \n",
    "    # Save the annotation to disk\n",
    "    print(\"\\n\".join(print_buffer), file= open(save_file_name, \"w\"))\n",
    "  \n",
    "  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "20d4f5e6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████| 504/504 [00:00<00:00, 3635.68it/s]\n"
     ]
    }
   ],
   "source": [
    "# Get the annotations\n",
    "annotations = [os.path.join('annotations', x) for x in os.listdir('annotations') if x[-3:] == \"xml\"]\n",
    "annotations.sort()\n",
    "\n",
    "# Convert and save the annotations\n",
    "for ann in tqdm(annotations):\n",
    "    info_dict = extract_info_from_xml(ann)\n",
    "    convert_to_yolov5(info_dict)\n",
    "annotations = [os.path.join('annotations', x) for x in os.listdir('annotations') if x[-3:] == \"txt\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c75e60ef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "random.seed(0)\n",
    "\n",
    "class_id_to_name_mapping = dict(zip(class_name_to_id_mapping.values(), class_name_to_id_mapping.keys()))\n",
    "\n",
    "def plot_bounding_box(image, annotation_list):\n",
    "    annotations = np.array(annotation_list)\n",
    "    w, h = image.size\n",
    "    \n",
    "    plotted_image = ImageDraw.Draw(image)\n",
    "\n",
    "    transformed_annotations = np.copy(annotations)\n",
    "    transformed_annotations[:,[1,3]] = annotations[:,[1,3]] * w\n",
    "    transformed_annotations[:,[2,4]] = annotations[:,[2,4]] * h \n",
    "    \n",
    "    transformed_annotations[:,1] = transformed_annotations[:,1] - (transformed_annotations[:,3] / 2)\n",
    "    transformed_annotations[:,2] = transformed_annotations[:,2] - (transformed_annotations[:,4] / 2)\n",
    "    transformed_annotations[:,3] = transformed_annotations[:,1] + transformed_annotations[:,3]\n",
    "    transformed_annotations[:,4] = transformed_annotations[:,2] + transformed_annotations[:,4]\n",
    "    \n",
    "    for ann in transformed_annotations:\n",
    "        obj_cls, x0, y0, x1, y1 = ann\n",
    "        plotted_image.rectangle(((x0,y0), (x1,y1)))\n",
    "        \n",
    "        plotted_image.text((x0, y0 - 10), class_id_to_name_mapping[(int(obj_cls))])\n",
    "    \n",
    "    plt.imshow(np.array(image))\n",
    "    plt.show()\n",
    "\n",
    "# Get any random annotation file \n",
    "annotation_file = random.choice(annotations)\n",
    "with open(annotation_file, \"r\") as file:\n",
    "    annotation_list = file.read().split(\"\\n\")[:-1]\n",
    "    annotation_list = [x.split(\" \") for x in annotation_list]\n",
    "    annotation_list = [[float(y) for y in x ] for x in annotation_list]\n",
    "\n",
    "#Get the corresponding image file\n",
    "image_file = annotation_file.replace(\"annotations\", \"images\").replace(\"txt\", \"png\")\n",
    "assert os.path.exists(image_file)\n",
    "\n",
    "#Load the image\n",
    "image = Image.open(image_file)\n",
    "\n",
    "#Plot the Bounding Box\n",
    "plot_bounding_box(image, annotation_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "87f41826",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Read images and annotations\n",
    "images = [os.path.join('images', x) for x in os.listdir('images')]\n",
    "annotations = [os.path.join('annotations', x) for x in os.listdir('annotations') if x[-3:] == \"txt\"]\n",
    "\n",
    "images.sort()\n",
    "annotations.sort()\n",
    "\n",
    "# Split the dataset into train-valid-test splits \n",
    "train_images, val_images, train_annotations, val_annotations = train_test_split(images, annotations, test_size = 0.2, random_state = 1)\n",
    "val_images, test_images, val_annotations, test_annotations = train_test_split(val_images, val_annotations, test_size = 0.5, random_state = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "6b8f3562",
   "metadata": {},
   "outputs": [],
   "source": [
    "!mkdir images\\train images\\val images\\test annotations\\train annotations\\val annotations\\test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "6d2eedf2",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Utility function to move images \n",
    "def move_files_to_folder(list_of_files, destination_folder):\n",
    "    for f in list_of_files:\n",
    "        try:\n",
    "            shutil.move(f, destination_folder)\n",
    "        except:\n",
    "            print(f)\n",
    "            assert False\n",
    "\n",
    "# Move the splits into their folders\n",
    "move_files_to_folder(train_images, 'images/train')\n",
    "move_files_to_folder(val_images, 'images/val/')\n",
    "move_files_to_folder(test_images, 'images/test/')\n",
    "move_files_to_folder(train_annotations, 'annotations/train/')\n",
    "move_files_to_folder(val_annotations, 'annotations/val/')\n",
    "move_files_to_folder(test_annotations, 'annotations/test/')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "b8b2ff80",
   "metadata": {},
   "outputs": [],
   "source": [
    "!mv annotations labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "15450b3a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ben Zahra\\Desktop\\yolov5\n"
     ]
    }
   ],
   "source": [
    "cd .."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "4992e90e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "^C\n"
     ]
    }
   ],
   "source": [
    "!python train.py --img 640 --cfg yolov5s.yaml --hyp hyp.scratch.yaml --batch 32 --epochs 100 --data clothing_data.yaml --weights yolov5s.pt --workers 24 --name yolo_clothing_det"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "eaf7448a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[34m\u001b[1mdetect: \u001b[0mweights=['runs/train/yolo_clothing_det/weights/best.pt'], source=0, imgsz=[640, 640], conf_thres=0.25, iou_thres=0.45, max_det=1000, device=, view_img=False, save_txt=False, save_conf=False, save_crop=False, nosave=False, classes=None, agnostic_nms=False, augment=False, visualize=False, update=False, project=runs\\detect, name=yolo_clothing_det, exist_ok=False, line_thickness=3, hide_labels=False, hide_conf=False, half=False, dnn=False\n",
      "1/1: 0...  success (inf frames 640x480 at 30.00 FPS)\n",
      "\n",
      "0: 480x640 Done. (0.266s)\n",
      "0: 480x640 Done. (0.282s)\n",
      "0: 480x640 Done. (0.247s)\n",
      "0: 480x640 Done. (0.274s)\n",
      "0: 480x640 Done. (0.258s)\n",
      "0: 480x640 Done. (0.268s)\n",
      "0: 480x640 Done. (0.262s)\n",
      "0: 480x640 Done. (0.228s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 1 Tshirts, Done. (0.238s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 1 Tshirts, Done. (0.236s)\n",
      "0: 480x640 1 Tshirts, 1 Jackets, Done. (0.245s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.228s)\n",
      "0: 480x640 Done. (0.228s)\n",
      "0: 480x640 Done. (0.226s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.229s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.228s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 1 Shoes, Done. (0.234s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 1 Shoes, Done. (0.241s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 1 Shoes, Done. (0.244s)\n",
      "0: 480x640 Done. (0.255s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.270s)\n",
      "0: 480x640 1 Tshirts, Done. (0.243s)\n",
      "0: 480x640 1 Tshirts, Done. (0.235s)\n",
      "0: 480x640 1 Tshirts, Done. (0.232s)\n",
      "0: 480x640 1 Tshirts, Done. (0.235s)\n",
      "0: 480x640 1 Tshirts, Done. (0.234s)\n",
      "0: 480x640 1 Tshirts, Done. (0.246s)\n",
      "0: 480x640 1 Tshirts, Done. (0.244s)\n",
      "0: 480x640 1 Tshirts, Done. (0.232s)\n",
      "0: 480x640 1 Tshirts, Done. (0.242s)\n",
      "0: 480x640 1 Tshirts, Done. (0.238s)\n",
      "0: 480x640 1 Tshirts, Done. (0.236s)\n",
      "0: 480x640 1 Tshirts, Done. (0.244s)\n",
      "0: 480x640 1 Tshirts, Done. (0.233s)\n",
      "0: 480x640 1 Tshirts, Done. (0.243s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.233s)\n",
      "0: 480x640 1 Tshirts, Done. (0.243s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.235s)\n",
      "0: 480x640 1 Tshirts, Done. (0.236s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.238s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.245s)\n",
      "0: 480x640 1 Tshirts, Done. (0.230s)\n",
      "0: 480x640 1 Tshirts, Done. (0.238s)\n",
      "0: 480x640 1 Tshirts, Done. (0.239s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.249s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.246s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.237s)\n",
      "0: 480x640 1 Tshirts, Done. (0.233s)\n",
      "0: 480x640 1 Tshirts, Done. (0.245s)\n",
      "0: 480x640 1 Tshirts, Done. (0.233s)\n",
      "0: 480x640 1 Tshirts, Done. (0.242s)\n",
      "0: 480x640 1 Tshirts, Done. (0.233s)\n",
      "0: 480x640 1 Tshirts, Done. (0.245s)\n",
      "0: 480x640 1 Tshirts, Done. (0.245s)\n",
      "0: 480x640 1 Tshirts, Done. (0.236s)\n",
      "0: 480x640 1 Tshirts, Done. (0.240s)\n",
      "0: 480x640 1 Tshirts, Done. (0.243s)\n",
      "0: 480x640 1 Tshirts, Done. (0.238s)\n",
      "0: 480x640 1 Tshirts, Done. (0.237s)\n",
      "0: 480x640 1 Tshirts, Done. (0.239s)\n",
      "0: 480x640 1 Tshirts, Done. (0.240s)\n",
      "0: 480x640 1 Tshirts, Done. (0.235s)\n",
      "0: 480x640 1 Tshirts, Done. (0.237s)\n",
      "0: 480x640 1 Tshirts, Done. (0.241s)\n",
      "0: 480x640 1 Tshirts, Done. (0.240s)\n",
      "0: 480x640 1 Tshirts, Done. (0.250s)\n",
      "0: 480x640 1 Tshirts, Done. (0.245s)\n",
      "0: 480x640 1 Tshirts, Done. (0.238s)\n",
      "0: 480x640 1 Tshirts, Done. (0.230s)\n",
      "0: 480x640 1 Tshirts, Done. (0.233s)\n",
      "0: 480x640 1 Tshirts, Done. (0.244s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.243s)\n",
      "0: 480x640 1 Tshirts, Done. (0.240s)\n",
      "0: 480x640 1 Tshirts, Done. (0.243s)\n",
      "0: 480x640 1 Tshirts, Done. (0.237s)\n",
      "0: 480x640 1 Tshirts, Done. (0.241s)\n",
      "0: 480x640 1 Tshirts, Done. (0.240s)\n",
      "0: 480x640 1 Tshirts, Done. (0.238s)\n",
      "0: 480x640 1 Tshirts, Done. (0.238s)\n",
      "0: 480x640 1 Tshirts, Done. (0.224s)\n",
      "0: 480x640 1 Tshirts, Done. (0.239s)\n",
      "0: 480x640 1 Tshirts, Done. (0.238s)\n",
      "0: 480x640 1 Tshirts, Done. (0.239s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.236s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.245s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.238s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.240s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.239s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.241s)\n",
      "0: 480x640 1 Tshirts, Done. (0.237s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 1 Tshirts, Done. (0.242s)\n",
      "0: 480x640 1 Tshirts, Done. (0.243s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.242s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.238s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.237s)\n",
      "0: 480x640 1 Tshirts, Done. (0.241s)\n",
      "0: 480x640 1 Tshirts, Done. (0.231s)\n",
      "0: 480x640 1 Tshirts, Done. (0.236s)\n",
      "0: 480x640 2 Tshirtss, Done. (0.243s)\n",
      "0: 480x640 1 Jackets, Done. (0.240s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 1 Tshirts, Done. (0.239s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.247s)"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "YOLOv5  2021-10-21 torch 1.9.1+cpu CPU\n",
      "\n",
      "Fusing layers... \n",
      "Model Summary: 213 layers, 7034398 parameters, 0 gradients, 15.9 GFLOPs\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "0: 480x640 Done. (0.251s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.230s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.250s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.224s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 1 Shoes, Done. (0.236s)\n",
      "0: 480x640 Done. (0.228s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 1 Shoes, Done. (0.242s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 1 Jackets, Done. (0.235s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 1 Jackets, Done. (0.242s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.224s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 1 Jackets, Done. (0.249s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.230s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.227s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 1 Shoes, Done. (0.233s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.227s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.247s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 1 Shoes, Done. (0.242s)\n",
      "0: 480x640 Done. (0.247s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 1 Tshirts, 1 Jackets, Done. (0.241s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 1 Tshirts, Done. (0.237s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 1 Tshirts, Done. (0.247s)\n",
      "0: 480x640 1 Tshirts, Done. (0.235s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 1 Shoes, Done. (0.238s)\n",
      "0: 480x640 1 Shoes, Done. (0.238s)\n",
      "0: 480x640 1 Shoes, Done. (0.230s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 1 Shoes, Done. (0.244s)\n",
      "0: 480x640 1 Shoes, Done. (0.229s)\n",
      "0: 480x640 1 Jackets, Done. (0.247s)\n",
      "0: 480x640 1 Shirts, Done. (0.233s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.223s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 1 Shoes, Done. (0.241s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.250s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 1 Shoes, Done. (0.239s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.226s)\n",
      "0: 480x640 1 Shoes, Done. (0.227s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 1 Shoes, Done. (0.249s)\n",
      "0: 480x640 1 Shoes, Done. (0.243s)\n",
      "0: 480x640 1 Shoes, Done. (0.237s)\n",
      "0: 480x640 1 Shoes, Done. (0.233s)\n",
      "0: 480x640 1 Shoes, Done. (0.237s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.250s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 1 Skirts, Done. (0.240s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 1 Shoes, Done. (0.245s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.230s)\n",
      "0: 480x640 Done. (0.226s)\n",
      "0: 480x640 1 Shoes, Done. (0.239s)\n",
      "0: 480x640 1 Shoes, Done. (0.240s)\n",
      "0: 480x640 1 Shoes, Done. (0.235s)\n",
      "0: 480x640 1 Shoes, Done. (0.249s)\n",
      "0: 480x640 1 Shoes, Done. (0.229s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 1 Shoes, Done. (0.237s)\n",
      "0: 480x640 1 sunglasses, Done. (0.245s)\n",
      "0: 480x640 1 sunglasses, Done. (0.239s)\n",
      "0: 480x640 1 sunglasses, Done. (0.234s)\n",
      "0: 480x640 1 sunglasses, Done. (0.248s)\n",
      "0: 480x640 1 sunglasses, Done. (0.236s)\n",
      "0: 480x640 1 sunglasses, Done. (0.243s)\n",
      "0: 480x640 1 sunglasses, Done. (0.241s)\n",
      "0: 480x640 1 sunglasses, Done. (0.234s)\n",
      "0: 480x640 1 sunglasses, Done. (0.236s)\n",
      "0: 480x640 Done. (0.228s)\n",
      "0: 480x640 1 sunglasses, Done. (0.240s)\n",
      "0: 480x640 1 sunglasses, Done. (0.242s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 1 sunglasses, Done. (0.235s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 1 sunglasses, Done. (0.228s)\n",
      "0: 480x640 1 sunglasses, Done. (0.240s)\n",
      "0: 480x640 1 sunglasses, Done. (0.239s)\n",
      "0: 480x640 1 Shoes, Done. (0.234s)\n",
      "0: 480x640 1 Jackets, Done. (0.237s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.230s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.225s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.229s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 1 Shoes, Done. (0.228s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 1 Shoes, Done. (0.235s)\n",
      "0: 480x640 1 Shoes, Done. (0.236s)\n",
      "0: 480x640 1 Shoes, Done. (0.259s)\n",
      "0: 480x640 1 Shoes, Done. (0.260s)\n",
      "0: 480x640 1 Shoes, Done. (0.240s)\n",
      "0: 480x640 1 Tshirts, Done. (0.235s)\n",
      "0: 480x640 Done. (0.247s)\n",
      "0: 480x640 Done. (0.250s)\n",
      "0: 480x640 1 Jackets, Done. (0.243s)\n",
      "0: 480x640 1 Shirts, Done. (0.238s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.253s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 1 Shoes, Done. (0.237s)\n",
      "0: 480x640 1 Shoes, Done. (0.235s)\n",
      "0: 480x640 1 Shoes, Done. (0.246s)\n",
      "0: 480x640 1 Shoes, Done. (0.222s)\n",
      "0: 480x640 1 Shoes, Done. (0.243s)\n",
      "0: 480x640 1 Shoes, Done. (0.243s)\n",
      "0: 480x640 1 Shoes, Done. (0.244s)\n",
      "0: 480x640 1 Shoes, Done. (0.236s)\n",
      "0: 480x640 1 Shoes, Done. (0.239s)\n",
      "0: 480x640 1 Shoes, Done. (0.232s)\n",
      "0: 480x640 1 Shoes, Done. (0.241s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 1 Shoes, Done. (0.247s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.229s)\n",
      "0: 480x640 1 Shoes, Done. (0.247s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 1 Shoes, Done. (0.237s)\n",
      "0: 480x640 1 Shoes, Done. (0.244s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 1 Shoes, Done. (0.240s)\n",
      "0: 480x640 1 Shoes, Done. (0.245s)\n",
      "0: 480x640 1 Shoes, Done. (0.238s)\n",
      "0: 480x640 1 Shoes, Done. (0.236s)\n",
      "0: 480x640 1 Shoes, Done. (0.241s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 1 Shoes, Done. (0.239s)\n",
      "0: 480x640 1 Shoes, Done. (0.242s)\n",
      "0: 480x640 1 Shoes, Done. (0.243s)\n",
      "0: 480x640 1 Shoes, Done. (0.231s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 1 Shoes, Done. (0.237s)\n",
      "0: 480x640 1 Shoes, Done. (0.240s)\n",
      "0: 480x640 1 Shoes, Done. (0.231s)\n",
      "0: 480x640 1 Shoes, Done. (0.234s)\n",
      "0: 480x640 1 Shoes, Done. (0.220s)\n",
      "0: 480x640 1 Shoes, Done. (0.238s)\n",
      "0: 480x640 1 Shoes, Done. (0.236s)\n",
      "0: 480x640 1 Shoes, Done. (0.246s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 Done. (0.225s)\n",
      "0: 480x640 1 sunglasses, Done. (0.245s)\n",
      "0: 480x640 1 sunglasses, Done. (0.238s)\n",
      "0: 480x640 1 sunglasses, Done. (0.239s)\n",
      "0: 480x640 1 sunglasses, Done. (0.236s)\n",
      "0: 480x640 1 sunglasses, Done. (0.228s)\n",
      "0: 480x640 1 sunglasses, Done. (0.240s)\n",
      "0: 480x640 1 sunglasses, Done. (0.246s)\n",
      "0: 480x640 1 sunglasses, Done. (0.238s)\n",
      "0: 480x640 1 sunglasses, Done. (0.246s)\n",
      "0: 480x640 1 sunglasses, Done. (0.243s)\n",
      "0: 480x640 1 sunglasses, Done. (0.239s)\n",
      "0: 480x640 1 sunglasses, Done. (0.246s)\n",
      "0: 480x640 1 sunglasses, Done. (0.243s)\n",
      "0: 480x640 1 sunglasses, Done. (0.228s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 1 sunglasses, Done. (0.240s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.254s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 1 sunglasses, Done. (0.243s)\n",
      "0: 480x640 1 sunglasses, Done. (0.238s)\n",
      "0: 480x640 1 Shoes, Done. (0.244s)\n",
      "0: 480x640 1 Shoes, Done. (0.241s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 1 Shoes, Done. (0.232s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 1 Shoes, Done. (0.240s)\n",
      "0: 480x640 1 Shoes, Done. (0.236s)\n",
      "0: 480x640 Done. (0.229s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.226s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 1 Shoes, Done. (0.240s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 1 Shirts, Done. (0.248s)\n",
      "0: 480x640 1 Shirts, Done. (0.237s)\n",
      "0: 480x640 1 Shirts, Done. (0.242s)\n",
      "0: 480x640 1 Shirts, Done. (0.225s)\n",
      "0: 480x640 1 Shirts, Done. (0.231s)\n",
      "0: 480x640 1 Shirts, Done. (0.231s)\n",
      "0: 480x640 1 Shirts, Done. (0.247s)\n",
      "0: 480x640 1 Shirts, Done. (0.245s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 1 Shirts, Done. (0.236s)\n",
      "0: 480x640 1 Shirts, Done. (0.232s)\n",
      "0: 480x640 1 Shirts, Done. (0.246s)\n",
      "0: 480x640 1 Shirts, Done. (0.225s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 1 Shirts, Done. (0.243s)\n",
      "0: 480x640 Done. (0.230s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.247s)\n",
      "0: 480x640 1 Tshirts, Done. (0.234s)\n",
      "0: 480x640 Done. (0.228s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.226s)\n",
      "0: 480x640 Done. (0.247s)\n",
      "0: 480x640 Done. (0.228s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.247s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.228s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.228s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.247s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.226s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.253s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.263s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.251s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.268s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.225s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 1 Tshirts, Done. (0.244s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.229s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 1 Shirts, Done. (0.238s)\n",
      "0: 480x640 1 Shirts, Done. (0.250s)\n",
      "0: 480x640 1 Shirts, Done. (0.224s)\n",
      "0: 480x640 1 Shirts, Done. (0.235s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 1 Shoes, Done. (0.245s)\n",
      "0: 480x640 1 Shoes, Done. (0.241s)\n",
      "0: 480x640 1 Shoes, Done. (0.228s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 2 Shoess, Done. (0.236s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.248s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.230s)\n",
      "0: 480x640 1 Shoes, Done. (0.236s)\n",
      "0: 480x640 1 Shoes, Done. (0.243s)\n",
      "0: 480x640 1 Shoes, Done. (0.242s)\n",
      "0: 480x640 1 Shoes, Done. (0.242s)\n",
      "0: 480x640 1 Shoes, Done. (0.248s)\n",
      "0: 480x640 1 Shoes, Done. (0.237s)\n",
      "0: 480x640 1 Shoes, Done. (0.242s)\n",
      "0: 480x640 Done. (0.227s)\n",
      "0: 480x640 1 Shoes, Done. (0.243s)\n",
      "0: 480x640 1 Shoes, Done. (0.237s)\n",
      "0: 480x640 1 Shoes, Done. (0.247s)\n",
      "0: 480x640 1 Shoes, Done. (0.237s)\n",
      "0: 480x640 1 Shoes, Done. (0.235s)\n",
      "0: 480x640 1 Shoes, Done. (0.240s)\n",
      "0: 480x640 1 Shoes, Done. (0.236s)\n",
      "0: 480x640 1 Shoes, Done. (0.242s)\n",
      "0: 480x640 1 Shoes, Done. (0.234s)\n",
      "0: 480x640 1 Shoes, Done. (0.240s)\n",
      "0: 480x640 1 Shoes, Done. (0.241s)\n",
      "0: 480x640 2 Shoess, Done. (0.241s)\n",
      "0: 480x640 1 Shoes, Done. (0.236s)\n",
      "0: 480x640 1 Shoes, Done. (0.251s)\n",
      "0: 480x640 1 Shoes, Done. (0.245s)\n",
      "0: 480x640 1 Shoes, Done. (0.238s)\n",
      "0: 480x640 1 Shoes, Done. (0.245s)\n",
      "0: 480x640 1 Shoes, Done. (0.239s)\n",
      "0: 480x640 1 Shoes, Done. (0.251s)\n",
      "0: 480x640 1 Shoes, Done. (0.229s)\n",
      "0: 480x640 1 Shoes, Done. (0.247s)\n",
      "0: 480x640 1 Shoes, Done. (0.246s)\n",
      "0: 480x640 Done. (0.250s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.234s)\n",
      "0: 480x640 Done. (0.247s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.229s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.248s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.225s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.232s)\n",
      "0: 480x640 Done. (0.227s)\n",
      "0: 480x640 Done. (0.247s)\n",
      "0: 480x640 Done. (0.224s)\n",
      "0: 480x640 Done. (0.225s)\n",
      "0: 480x640 Done. (0.239s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.241s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.245s)\n",
      "0: 480x640 Done. (0.231s)\n",
      "0: 480x640 Done. (0.240s)\n",
      "0: 480x640 Done. (0.250s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.246s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.230s)\n",
      "0: 480x640 Done. (0.249s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.248s)\n",
      "0: 480x640 Done. (0.235s)\n",
      "0: 480x640 Done. (0.230s)\n",
      "0: 480x640 Done. (0.243s)\n",
      "0: 480x640 Done. (0.236s)\n",
      "0: 480x640 Done. (0.242s)\n",
      "0: 480x640 Done. (0.233s)\n",
      "0: 480x640 Done. (0.238s)\n",
      "0: 480x640 Done. (0.248s)\n",
      "0: 480x640 Done. (0.251s)\n",
      "0: 480x640 Done. (0.237s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 Done. (0.244s)\n",
      "0: 480x640 1 Shirts, Done. (0.241s)\n",
      "0: 480x640 1 Shirts, Done. (0.240s)\n",
      "0: 480x640 1 Shirts, Done. (0.226s)\n",
      "0: 480x640 1 Shirts, Done. (0.241s)\n",
      "0: 480x640 1 Shirts, Done. (0.243s)\n",
      "0: 480x640 1 Shirts, Done. (0.239s)\n",
      "0: 480x640 1 Shirts, Done. (0.240s)\n",
      "0: 480x640 1 Shoes, Done. (0.233s)\n",
      "0: 480x640 1 Shoes, Done. (0.233s)\n",
      "0: 480x640 1 Shoes, Done. (0.238s)\n",
      "0: 480x640 1 Shoes, Done. (0.260s)\n"
     ]
    }
   ],
   "source": [
    "!python detect.py --source 0 --weights runs/train/yolo_clothing_det/weights/best.pt --conf 0.25 --name yolo_clothing_det"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "fd2a9fcb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[34m\u001b[1mdetect: \u001b[0mweights=['runs/train/yolo_clothing_det/weights/best.pt'], source=./ClothingPascalVOC/images/test/test2.jpg, imgsz=[640, 640], conf_thres=0.25, iou_thres=0.45, max_det=1000, device=, view_img=False, save_txt=False, save_conf=False, save_crop=False, nosave=False, classes=None, agnostic_nms=False, augment=False, visualize=False, update=False, project=runs\\detect, name=yolo_clothing_det, exist_ok=False, line_thickness=3, hide_labels=False, hide_conf=False, half=False, dnn=False\n",
      "image 1/1 C:\\Users\\Ben Zahra\\Desktop\\yolov5\\ClothingPascalVOC\\images\\test\\test2.jpg: 640x448 Done. (0.244s)\n",
      "Speed: 0.0ms pre-process, 244.0ms inference, 0.0ms NMS per image at shape (1, 3, 640, 640)\n",
      "Results saved to \u001b[1mruns\\detect\\yolo_clothing_det9\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "YOLOv5  2021-10-21 torch 1.9.1+cpu CPU\n",
      "\n",
      "Fusing layers... \n",
      "Model Summary: 213 layers, 7034398 parameters, 0 gradients, 15.9 GFLOPs\n"
     ]
    }
   ],
   "source": [
    "!python detect.py --source ./ClothingPascalVOC/images/test/test2.jpg --weights runs/train/yolo_clothing_det/weights/best.pt --name yolo_clothing_det"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c007a374",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting onnx\n",
      "  Downloading onnx-1.10.1-cp38-cp38-win_amd64.whl (11.4 MB)\n",
      "Collecting coremltools\n",
      "  Downloading coremltools-5.0.tar.gz (912 kB)\n",
      "Collecting onnx-simplifier\n",
      "  Downloading onnx-simplifier-0.3.6.tar.gz (13 kB)\n",
      "Requirement already satisfied: numpy>=1.14.5 in c:\\programdata\\anaconda3\\lib\\site-packages (from coremltools) (1.20.1)\n",
      "Requirement already satisfied: protobuf>=3.1.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from coremltools) (3.19.0)\n",
      "Requirement already satisfied: sympy in c:\\programdata\\anaconda3\\lib\\site-packages (from coremltools) (1.8)\n",
      "Requirement already satisfied: tqdm in c:\\programdata\\anaconda3\\lib\\site-packages (from coremltools) (4.59.0)\n",
      "Requirement already satisfied: packaging in c:\\programdata\\anaconda3\\lib\\site-packages (from coremltools) (20.9)\n",
      "Requirement already satisfied: six in c:\\programdata\\anaconda3\\lib\\site-packages (from onnx) (1.15.0)\n",
      "Requirement already satisfied: typing-extensions>=3.6.2.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from onnx) (3.7.4.3)\n",
      "Collecting onnxoptimizer>=0.2.5\n",
      "  Downloading onnxoptimizer-0.2.6-cp38-cp38-win_amd64.whl (228 kB)\n",
      "Collecting onnxruntime>=1.6.0\n",
      "  Downloading onnxruntime-1.9.0-cp38-cp38-win_amd64.whl (5.1 MB)\n",
      "Collecting flatbuffers\n",
      "  Downloading flatbuffers-2.0-py2.py3-none-any.whl (26 kB)\n",
      "Requirement already satisfied: pyparsing>=2.0.2 in c:\\programdata\\anaconda3\\lib\\site-packages (from packaging->coremltools) (2.4.7)\n",
      "Requirement already satisfied: mpmath>=0.19 in c:\\programdata\\anaconda3\\lib\\site-packages (from sympy->coremltools) (1.2.1)\n",
      "Building wheels for collected packages: coremltools, onnx-simplifier\n",
      "  Building wheel for coremltools (setup.py): started\n",
      "  Building wheel for coremltools (setup.py): finished with status 'done'\n",
      "  Created wheel for coremltools: filename=coremltools-5.0-py3-none-any.whl size=1211830 sha256=c9f7ae84f6da64a12b94dea8d420dd05fdc9c674a78180ec1a4f730bd2d73366\n",
      "  Stored in directory: c:\\users\\ben zahra\\appdata\\local\\pip\\cache\\wheels\\1b\\7b\\0e\\35635ebdee828854ff576c0e53c84051bb8a9fb164153ed659\n",
      "  Building wheel for onnx-simplifier (setup.py): started\n",
      "  Building wheel for onnx-simplifier (setup.py): finished with status 'done'\n",
      "  Created wheel for onnx-simplifier: filename=onnx_simplifier-0.3.6-py3-none-any.whl size=12860 sha256=7264ceb17900e0e8c63023da33e4fe4349d61664f71d200181e8321864c248e0\n",
      "  Stored in directory: c:\\users\\ben zahra\\appdata\\local\\pip\\cache\\wheels\\f7\\b0\\8c\\21208ab41ded14656f0165b0837b4689f7082fb2a000d1240d\n",
      "Successfully built coremltools onnx-simplifier\n",
      "Installing collected packages: onnx, flatbuffers, onnxruntime, onnxoptimizer, onnx-simplifier, coremltools\n",
      "Successfully installed coremltools-5.0 flatbuffers-2.0 onnx-1.10.1 onnx-simplifier-0.3.6 onnxoptimizer-0.2.6 onnxruntime-1.9.0\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install onnx coremltools onnx-simplifier\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "bcf37606",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[34m\u001b[1mexport: \u001b[0mdata=data\\coco128.yaml, weights=runs/train/yolo_clothing_det/weights/best.pt, imgsz=[640, 640], batch_size=1, device=cpu, half=False, inplace=False, train=True, optimize=False, int8=False, dynamic=False, simplify=False, opset=13, topk_per_class=100, topk_all=100, iou_thres=0.45, conf_thres=0.25, include=['torchscript', 'onnx']\n",
      "\n",
      "\u001b[34m\u001b[1mPyTorch:\u001b[0m starting from runs\\train\\yolo_clothing_det\\weights\\best.pt (14.5 MB)\n",
      "\n",
      "\u001b[34m\u001b[1mTorchScript:\u001b[0m starting export with torch 1.9.1+cpu...\n",
      "\u001b[34m\u001b[1mTorchScript:\u001b[0m export success, saved as runs\\train\\yolo_clothing_det\\weights\\best.torchscript.pt (28.4 MB)\n",
      "\n",
      "\u001b[34m\u001b[1mONNX:\u001b[0m starting export with onnx 1.10.1...\n",
      "\u001b[34m\u001b[1mONNX:\u001b[0m export success, saved as runs\\train\\yolo_clothing_det\\weights\\best.onnx (28.2 MB)\n",
      "\u001b[34m\u001b[1mONNX:\u001b[0m run --dynamic ONNX model inference with: 'python detect.py --weights runs\\train\\yolo_clothing_det\\weights\\best.onnx'\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "YOLOv5  2021-10-21 torch 1.9.1+cpu CPU\n",
      "\n",
      "Fusing layers... \n",
      "Model Summary: 213 layers, 7034398 parameters, 0 gradients, 15.9 GFLOPs\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Export complete (4.96s)\n",
      "Results saved to \u001b[1mC:\\Users\\Ben Zahra\\Desktop\\yolov5\\runs\\train\\yolo_clothing_det\\weights\u001b[0m\n",
      "Visualize with https://netron.app\n"
     ]
    }
   ],
   "source": [
    "!python export.py --train --weights runs/train/yolo_clothing_det/weights/best.pt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "be5337df",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Le volume dans le lecteur C n'a pas de nom.\n",
      " Le num‚ro de s‚rie du volume est D0A2-8479\n",
      "\n",
      " R‚pertoire de C:\\Users\\Ben Zahra\\Desktop\\yolov5\\runs\\train\\yolo_clothing_det\\weights\n",
      "\n",
      "22/10/2021  00:55    <DIR>          .\n",
      "22/10/2021  00:55    <DIR>          ..\n",
      "22/10/2021  00:55        28ÿ569ÿ647 best.onnx\n",
      "21/10/2021  07:19        14ÿ498ÿ037 best.pt\n",
      "22/10/2021  00:55        28ÿ627ÿ340 best.torchscript.pt\n",
      "21/10/2021  07:19        14ÿ498ÿ037 last.pt\n",
      "               4 fichier(s)       86ÿ193ÿ061 octets\n",
      "               2 R‚p(s)  119ÿ751ÿ671ÿ808 octets libres\n"
     ]
    }
   ],
   "source": [
    "ls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "ab973517",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Simplifying...\n",
      "Checking 0/3...\n",
      "Checking 1/3...\n",
      "Checking 2/3...\n",
      "Ok!\n"
     ]
    }
   ],
   "source": [
    "!python -m onnxsim runs/train/yolo_clothing_det/weights/best.onnx runs/train/yolo_clothing_det/weights/best-sim.onnx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "e4e5b67a",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "'sudo' n'est pas reconnu en tant que commande interne\n",
      "ou externe, un programme ex‚cutable ou un fichier de commandes.\n"
     ]
    }
   ],
   "source": [
    "!sudo apt install build-essential libopencv-dev cmake"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8dee9536",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}