SSD.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "SSD.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7JrvtacKVLG9"
      },
      "source": [
        "# SSD를 사용한 OD"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ouadwTUuVvEr"
      },
      "source": [
        "## Configure"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mclogGkLVxIc"
      },
      "source": [
        "class Config: # hyperparameter 는 task에 맞게 유동적으로 바꿔진 것 같음.\n",
        "    VOC_ROOT = '/workspace/hakerton/simple-ssd-for-beginners/VOCdevkit'\n",
        "    num_classes = 7\n",
        "    resume = None\n",
        "    lr = 0.0007\n",
        "    batch_size = 64 \n",
        "    momentum = 0.9\n",
        "    weight_decay = 5e-4\n",
        "    epoch = 100 \n",
        "    gamma = 0.2\n",
        "    lr_reduce_epoch = 30\n",
        "    save_folder = 'weights/'\n",
        "    basenet = 'vgg16_reducedfc.pth'\n",
        "    log_fn = 1\n",
        "    neg_radio = 3\n",
        "    min_size = 300\n",
        "    grids = (38, 19, 10, 5, 3, 1)\n",
        "    aspect_ratios = ((2,), (2, 3), (2, 3), (2, 3), (2,), (2,)) # ratio.\n",
        "    steps = [s / 300 for s in (8, 16, 32, 64, 100, 300)]\n",
        "    sizes = [s / 300 for s in (30, 60, 111, 162, 213, 264, 315)] # default box size의 기준.\n",
        "    anchor_num = [4, 6, 6, 6, 4, 4]\n",
        "    mean = (104, 117, 123)\n",
        "    variance = (0.1, 0.2)\n",
        "opt = Config()"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "730FFMwwVjOQ"
      },
      "source": [
        "## Utils"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Cz4mu0gsVmat"
      },
      "source": [
        "\n",
        "\n",
        "import numpy as np\n",
        "import torch\n",
        "import cv2\n",
        "\n",
        "\"\"\"\n",
        "    cx,cy,h,w (중앙좌표 및 너비 높이) -> x1,y1, x2, y2\n",
        "\"\"\"\n",
        "\n",
        "def point_form(boxes):\n",
        "\n",
        "    tl = boxes[:, :2] - boxes[:, 2:]/2\n",
        "    br = boxes[:, :2] + boxes[:, 2:]/2\n",
        "\n",
        "    return np.concatenate([tl, br], axis=1)\n",
        "\n",
        "\n",
        "def detection_collate(batch):\n",
        "\n",
        "    targets = []\n",
        "    imgs = []\n",
        "    for sample in batch:\n",
        "        imgs.append(sample[0])\n",
        "        targets.append(sample[1])\n",
        "    return torch.stack(imgs), np.array(targets)\n",
        "\n",
        "\n",
        "\n",
        "def bbox_iou(box_a, box_b):\n",
        "\n",
        "    m = box_a.shape[0]\n",
        "    n = box_b.shape[0]\n",
        "\n",
        "    tl = np.maximum(box_a[:, None, :2], box_b[None, :, :2])\n",
        "    br = np.minimum(box_a[:, None, 2:], box_b[None, :, 2:])\n",
        "\n",
        "    wh = np.maximum(br-tl, 0)\n",
        "    \n",
        "    inner = wh[:, :, 0]*wh[:, :, 1]\n",
        "\n",
        "    a = box_a[:, 2:] - box_a[:, :2]\n",
        "    b = box_b[:, 2:] - box_b[:, :2]\n",
        "\n",
        "    a = a[:, 0] * a[:, 1]\n",
        "    b = b[:, 0] * b[:, 1]\n",
        "\n",
        "    a = a[:, None]\n",
        "    b = b[None, :]\n",
        "\n",
        "    return inner / (a+b-inner)\n",
        "\n",
        "\n",
        "def nms(boxes, score, threshold=0.4):\n",
        "\n",
        "    sort_ids = np.argsort(score)\n",
        "    pick = []\n",
        "    while len(sort_ids) > 0:\n",
        "        i = sort_ids[-1]\n",
        "        pick.append(i)\n",
        "        if len(sort_ids) == 1:\n",
        "            break\n",
        "\n",
        "        sort_ids = sort_ids[:-1]\n",
        "        box = boxes[i].reshape(1, 4)\n",
        "        ious = bbox_iou(box, boxes[sort_ids]).reshape(-1)\n",
        "\n",
        "        sort_ids = np.delete(sort_ids, np.where(ious > threshold)[0])\n",
        "\n",
        "    return pick\n",
        "\n",
        "\n",
        "\n",
        "\"\"\"\n",
        "    nms 기반으로 detect\n",
        "\"\"\"\n",
        "def detect(locations, scores, nms_threshold, gt_threshold):\n",
        "\n",
        "    scores = scores[:, 1:] \n",
        "\n",
        "    keep_boxes = []\n",
        "    keep_confs = []\n",
        "    keep_labels = []\n",
        "\n",
        "    for i in range(scores.shape[1]):\n",
        "\n",
        "        mask = scores[:, i] >= gt_threshold\n",
        "        label_scores = scores[mask, i] \n",
        "        label_boxes = locations[mask]\n",
        "\n",
        "        if len(label_scores) == 0:\n",
        "            continue\n",
        "\n",
        "        pick = nms(label_boxes, label_scores, threshold=nms_threshold)\n",
        "        label_scores = label_scores[pick]\n",
        "        label_boxes = label_boxes[pick]\n",
        "        \n",
        "\n",
        "        keep_boxes.append(label_boxes.reshape(-1))\n",
        "        keep_confs.append(label_scores)\n",
        "        keep_labels.extend([i]*len(label_scores))\n",
        "\n",
        "    if len(keep_boxes) == 0:\n",
        "        return np.array([]), np.array([]), np.array([])\n",
        "        \n",
        "    \n",
        "    keep_boxes = np.concatenate(keep_boxes, axis=0).reshape(-1, 4)\n",
        "\n",
        "    keep_confs = np.concatenate(keep_confs, axis=0)\n",
        "    keep_labels = np.array(keep_labels).reshape(-1)\n",
        "\n",
        "    return keep_boxes, keep_confs, keep_labels\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "# draw rectangle\n",
        "def draw_rectangle(src_img, labels, conf, locations, label_map):\n",
        "\n",
        "    num_obj = len(labels)\n",
        "    COLORS = [(255, 0, 0), (0, 255, 0), (0, 0, 255)]\n",
        "    FONT = cv2.FONT_HERSHEY_SIMPLEX\n",
        "    img = src_img.copy()\n",
        "    for i in range(num_obj):\n",
        "        tl = tuple(locations[i][:2])\n",
        "        br = tuple(locations[i][2:])\n",
        "        \n",
        "        cv2.rectangle(img,\n",
        "                      tl,\n",
        "                      br,\n",
        "                      COLORS[i%3], 3)\n",
        "        cv2.putText(img, label_map[labels[i]], tl,\n",
        "                    FONT, 1, (255, 255, 255), 2)\n",
        "    \n",
        "    img = img[:, :, ::-1]\n",
        "\n",
        "    return img"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3nHJHDSYV-DT"
      },
      "source": [
        "## Model (SSD)\n",
        "### SSD는 backbone model에서부터 그 후 추가적인 conv layers들에서 여러번 prediction을 진행합니다. (default : backbone에서 2번, extra에서 4번)\n",
        "![ssd.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hLI2DpN9WAt2"
      },
      "source": [
        "# -*- coding: utf-8 -*-\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "import numpy as np\n",
        "import torchvision\n",
        "from torchvision import models\n",
        "import torch.nn.functional as F\n",
        "import torch.nn.init as init\n",
        "\n",
        "\n",
        "\n",
        "class L2Norm(nn.Module): # model에서의 L2Norm 사용\n",
        "\n",
        "    def __init__(self, n_channels, scale):\n",
        "        super(L2Norm, self).__init__()\n",
        "        self.n_channels = n_channels\n",
        "        self.gamma = scale or None\n",
        "        self.eps = 1e-10\n",
        "        self.weight = nn.Parameter(torch.Tensor(self.n_channels))\n",
        "        self.reset_parameters()\n",
        "\n",
        "    def reset_parameters(self):\n",
        "        init.constant_(self.weight, self.gamma)\n",
        "        \n",
        "\n",
        "    def forward(self, x):\n",
        "        norm = torch.sqrt(x.pow(2).sum(dim=1, keepdim=True)) + self.eps # L2 Norm + 1e-10 (0 으로 나누지 않기 위해)\n",
        "        x = torch.div(x, norm) # x/norm, L2 Norm을 기준으로 Normalization\n",
        "        x = self.weight.unsqueeze(0).unsqueeze(2).unsqueeze(3).expand_as(x) * x # (1,n_channels, 1, 1) -> (BS, n_channels, H, W)weight를로 바꿔줌\n",
        "        return x\n",
        "\n",
        "\"\"\"\n",
        "    VGG에서 추가적인 예측을 진행해야하기 때문에 list로 저장해줍니다.\n",
        "\"\"\"\n",
        "\n",
        "def VGG():\n",
        "    base_vgg = models.vgg16().features # MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
        "    base_vgg[16].ceil_mode = True # ceilmode 수정\n",
        "    vgg = []\n",
        "    for i in range(30): # 마지막 maxpool 제외하고 가져오기\n",
        "        vgg.append(base_vgg[i])\n",
        "\n",
        "    pool5 = nn.MaxPool2d(kernel_size=3, stride=1, padding=1)\n",
        "    conv6 = nn.Conv2d(512, 1024, kernel_size=3, padding=6, dilation=6)\n",
        "    conv7 = nn.Conv2d(1024, 1024, kernel_size=1)\n",
        "    vgg += [pool5, conv6,nn.ReLU(inplace=True), conv7, nn.ReLU(inplace=True)]\n",
        "    return vgg\n",
        "\n",
        "\"\"\"\n",
        "    본 코드에서는 짝수 레이어를 예측에 사용합니다.    \n",
        "\"\"\"\n",
        "\n",
        "def Extra():\n",
        "    layers = []\n",
        "    conv8_1 = nn.Conv2d(1024, 256, kernel_size=1, stride=1)\n",
        "    conv8_2 = nn.Conv2d(256, 512, kernel_size=3, stride=2, padding=1)\n",
        "    conv9_1 = nn.Conv2d(512, 128, kernel_size=1, stride=1)\n",
        "    conv9_2 = nn.Conv2d(128, 256, kernel_size=3, stride=2, padding=1)\n",
        "    conv10_1 = nn.Conv2d(256, 128, kernel_size=1, stride=1)\n",
        "    conv10_2 = nn.Conv2d(128, 256, kernel_size=3, stride=1)\n",
        "    conv11_1 = nn.Conv2d(256, 128, kernel_size=1)\n",
        "    conv11_2 = nn.Conv2d(128, 256, kernel_size=3, stride=1)\n",
        "\n",
        "    layers = [conv8_1, conv8_2, conv9_1, conv9_2, conv10_1, conv10_2, conv11_1, conv11_2]\n",
        "\n",
        "    return layers\n",
        "\n",
        "\"\"\"\n",
        "    예측에 사용되는 레이어들을 지정해줍시다.\n",
        "    이 때 bboxes[k]가 의미하는 바는 default box의 ratio에 따른 예측의 갯수 입니다. (ratio에 대해 두가지를 고려한다면, 원래의 default box, ratio당 가로 세로 고려 한 것, 1:1 좀 더 큰 박스, 이렇게 6개가 존재하게 됩니다.)\n",
        "\"\"\"\n",
        "\n",
        "def Feature_extractor(vgg, extral, bboxes, num_classes):\n",
        "\n",
        "    loc_layers = []\n",
        "    conf_layers = []\n",
        "    vgg_useful = [21, 33] # 사용할 vgg conv 선택. 21 : 중간의 512-512 conv, 33 : 가장 마지막 1024-1024 conv\n",
        "    \n",
        "    for k, v in enumerate(vgg_useful):\n",
        "        loc_layers += [nn.Conv2d(vgg[v].out_channels, # 512, 1024 -> 38, 19 (이미지 사이즈가 300일 겅우임.)\n",
        "                                 bboxes[k] * 4, kernel_size=3, padding=1)]  #  anchor * 4개의 값을 갖게 하기 위해서(좌표) -> bounding box의 개수 * 4\n",
        "        conf_layers += [nn.Conv2d(vgg[v].out_channels,\n",
        "                        bboxes[k] * num_classes, kernel_size=3, padding=1)] # class에 대한 정보들(num_classes) -> bounding box의 개수 * num_classes\n",
        "\n",
        "    for k, v in enumerate(extral[1::2], 2): #conv 짝수만 사용. 시작 인덱스 2. 위에서 0, 1번재 가져옴\n",
        "        loc_layers += [nn.Conv2d(v.out_channels, bboxes[k] # 512, 256, 256, 256 -> 10, 5, 3, 1 (feature map)\n",
        "                                 * 4, kernel_size=3, padding=1)]\n",
        "        conf_layers += [nn.Conv2d(v.out_channels, bboxes[k]\n",
        "                                  * num_classes, kernel_size=3, padding=1)]\n",
        "        \n",
        "    \n",
        "    \n",
        "    return loc_layers, conf_layers \n",
        "\n",
        "\n",
        "\"\"\"\n",
        "    위 메서드 들을 사용해서 SSD model을 선언.\n",
        "\"\"\"\n",
        "\n",
        "class SSD(nn.Module):\n",
        "\n",
        "    def __init__(self, num_classes, bboxes):\n",
        "        super(SSD, self).__init__()\n",
        "\n",
        "        self.num_classes = num_classes\n",
        "        self.bboxes = bboxes \n",
        "\n",
        "        self.vgg_list = VGG()\n",
        "        self.extra_list = Extra()\n",
        "\n",
        "        self.loc_layers_list, self.conf_layers_list = Feature_extractor(self.vgg_list, self.extra_list, self.bboxes, self.num_classes)\n",
        "\n",
        "        self.L2Norm = L2Norm(512, 20)\n",
        "\n",
        "\n",
        "        self.vgg = nn.ModuleList(self.vgg_list)\n",
        "        self.extras = nn.ModuleList(self.extra_list)\n",
        "        self.loc = nn.ModuleList(self.loc_layers_list)\n",
        "        self.conf = nn.ModuleList(self.conf_layers_list)\n",
        "\n",
        "\n",
        "    \"\"\"\n",
        "        SSD는 기본적으로 여러 피쳐 정보에서부터 bbox, class를 예측한다.\n",
        "        feature map의 크기가 크다면 작은 객체를 탐지하기 용이하고, feature map의 크기가 작으면 큰 객체를 탐지하기 용이하다고 한다.\n",
        "    \"\"\"\n",
        "    \n",
        "    def forward(self, x):\n",
        "\n",
        "        source = []\n",
        "        loc = []\n",
        "        conf = []\n",
        "\n",
        "        vgg_source = [22, 34] # vgg에서 source로 가져올 것들.\n",
        "        for i, v in enumerate(self.vgg):\n",
        "            x = v(x)\n",
        "\n",
        "            if i in vgg_source:\n",
        "                if i == 22: # vgg source로 사용할 것이라면 source 에 포함.\n",
        "                    s = self.L2Norm(x)\n",
        "                else:\n",
        "                    s = x\n",
        "                source.append(s)\n",
        "\n",
        "        for i, v in enumerate(self.extras):\n",
        "            x = F.relu(v(x), inplace=True)\n",
        "            if i % 2 == 1: #짝수 레이어들 사용한다고 했으니 output source에 넣어주자. 이미지 데이터 자체가 크기때문에 많은 레이어를 사용하는 것 같다.\n",
        "                source.append(x)\n",
        "\n",
        "\n",
        "        for s, l, c in zip(source, self.loc, self.conf): # source (BS, C, H, W)\n",
        "            # 내가 loc layer로 쓰려는 것들. source를 내놓은 layer의 output channel을 이 conv의 input channel과 같게 설정. 그리고 bbox 갯수에 맞게 n*4를 내놓음\n",
        "            loc.append(l(s).permute(0, 2, 3, 1).contiguous()) # source_n*(BS, H, W, C)\n",
        "            # 내가 conf layer로 쓰려는 것들. source를 내놓은 layer의 output channel을 이 conv의 input channel과 같게 설정. 그리고 num classes에 맞게 n*num_classes를 내놓음\n",
        "            conf.append(c(s).permute(0, 2, 3, 1).contiguous()) # source_n*(BS, H, W, C)\n",
        "        # source_n -> 6개. 6번예측함.\n",
        "        loc = torch.cat([o.view(o.size(0), -1) for o in loc], 1) # source_n*(BS,H,W,4*n) -> source_n*(BS,H*W*4n) -> (BS,sourc_n*H*HH4n) ex) 6*10*10*4*n\n",
        "        conf = torch.cat([o.view(o.size(0), -1) for o in conf], 1) # ..\n",
        "\n",
        "        loc = loc.view(loc.size(0), -1, 4) # BS, 6*H*W, 4n 이미지에서 source에 대한 여러 이미지 예측. 4n\n",
        "        conf = conf.view(conf.size(0), -1, self.num_classes)\n",
        "\n",
        "        return loc, conf\n",
        "\n",
        "\n",
        "\n",
        "if __name__ == '__main__':\n",
        "\n",
        "    x = torch.randn(1, 3, 300, 300)\n",
        "    ssd = SSD(21, [4,6,6,6,4,4])\n",
        "\n",
        "    y = ssd(x)\n",
        "    print(y[0].shape, y[1].shape)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pUXZH1TwWEaV"
      },
      "source": [
        "## Multibox Encoder (Matching)\n",
        "### SSD 는 scaling [0,1] 된 default box를 토대로 offset을 적용하여 예측합니다.\n",
        "### 이를위해 default box를 생성하고 data를 훈련에 적합한 상태로 만들며 이를 다시 원래 상태로 되돌려 놓는 것이 필요합니다.\n",
        "![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t0_kzxXEWH0A"
      },
      "source": [
        "import itertools\n",
        "import numpy as np\n",
        "\n",
        "class MultiBoxEncoder(object):\n",
        "\n",
        "    def __init__(self, opt):\n",
        "        self.variance = opt.variance # (0.1, 0.2)\n",
        "\n",
        "        default_boxes = list()\n",
        "        for k in range(len(opt.grids)):\n",
        "\n",
        "            for v, u in itertools.product(range(opt.grids[k]), repeat=2): # range (5) -> 0,0 0,1 0,2 0,3 0,4 ...\n",
        "                cx = (u + 0.5) * opt.steps[k] # 어떤 feature map에서의 좌표 (x,y)일 때 각 grid 좌표의 center를 (x+0.5, y+0.5)\n",
        "                cy = (v + 0.5) * opt.steps[k] # 마찬가지. 연산 수행 후 전체 부분으로 표현해주기위해 scaling\n",
        "                 # cx, cy 최소 최대를 0, 1 사이로 두도록\n",
        "                s = opt.sizes[k]\n",
        "                default_boxes.append((cx, cy, s, s))\n",
        "\n",
        "                s = np.sqrt(opt.sizes[k] * opt.sizes[k + 1]) # ratio가 1일 때 좀 더 큰 박스\n",
        "                default_boxes.append((cx, cy, s, s)) # 각 좌표에서의 해당 grid 표현 (default box). anchor box와 결이 같은듯.\n",
        "\n",
        "                s = opt.sizes[k]\n",
        "                for ar in opt.aspect_ratios[k]: # ratio : s*sqrt(ar), s/sqrt(ar) ...  original, ratio적용 * 2, 1:1\n",
        "                    default_boxes.append(\n",
        "                        (cx, cy, s * np.sqrt(ar), s / np.sqrt(ar)))\n",
        "                    default_boxes.append(\n",
        "                        (cx, cy, s / np.sqrt(ar), s * np.sqrt(ar))) # ratio에 대해서도 고려. center 좌표에 대한 default boxes를 표현해준다\n",
        "\n",
        "        default_boxes = np.clip(default_boxes, a_min=0, a_max=1) # 밖으로 벗어나는 친구들은 무시해주자.\n",
        "        self.default_boxes = np.array(default_boxes) # 디폴트박스 생성 완료!!\n",
        "\n",
        "    def encode(self, boxes, labels, threshold=0.5): # 내가 생각한 default boxes들과 groud truth 사이의 iou를 파악하고 iou threshold를 넘는 것만 사용하자. (훈련 시에만.)\n",
        "    # 훈련이 용이하게 ground truth 들을 default boxes들과 매칭 시켜놓자.\n",
        "        if len(boxes) == 0:\n",
        "            return (\n",
        "                np.zeros(self.default_boxes.shape, dtype=np.float32),\n",
        "                np.zeros(self.default_boxes.shape[:1], dtype=np.int32))\n",
        "\n",
        "        iou = bbox_iou(point_form(self.default_boxes), boxes) # x1, y1, x2, y2 꼴로 만들어서 bbox_iou 계산하자.\n",
        "\n",
        "        gt_idx = iou.argmax(axis=1) # iou가 제일 큰거..\n",
        "        iou = iou.max(axis=1) # 제일 큰 iou\n",
        "        boxes = boxes[gt_idx]\n",
        "        labels = labels[gt_idx]\n",
        "\n",
        "        loc = np.hstack(( # gt 중앙 좌표 / 2 - default box 중앙 좌표 / variance[0] * (default boxes h,w / 2)\n",
        "            ((boxes[:, :2] + boxes[:, 2:]) / 2 - self.default_boxes[:, :2]) /\n",
        "            (self.variance[0] * self.default_boxes[:, 2:]),\n",
        "            np.log((boxes[:, 2:] - boxes[:, :2]) / self.default_boxes[:, 2:]) /\n",
        "            self.variance[1])) # box h,w / (default box h,w /2) / variance[1]\n",
        "    # box loss에 들어가는 형태로 encode 시켜줍니다.\n",
        "\n",
        "        conf = 1 + labels # 0번째 index는 배경이다.\n",
        "        conf[iou < threshold] = 0 # threshold를 넘지 못하면 background로 가정하자.\n",
        "       \n",
        "\n",
        "        return loc.astype(np.float32), conf.astype(np.int32)\n",
        " \n",
        "    def decode(self, loc): # 평가 할 때 box output이 center와 offset으로 나와서 이를 적용해주는 과정. width, height는 log가 씌워진 상태로 prediction이 나오기 때문에 이것도 바꿔줌.\n",
        "\n",
        "        boxes = np.hstack((\n",
        "            self.default_boxes[:, :2] + # offset 적용해주기\n",
        "            loc[:, :2] * self.variance[0] * self.default_boxes[:, 2:],\n",
        "            self.default_boxes[:, 2:] * np.exp(loc[:, 2:] * self.variance[1])))\n",
        "        boxes[:, :2] -= boxes[:, 2:] / 2\n",
        "        boxes[:, 2:] += boxes[:, :2] # 원래의 상태로 되돌려 줌.\n",
        "\n",
        "        return boxes\n",
        "\n",
        "\n",
        "if __name__ == '__main__':\n",
        "    from config import opt\n",
        "    mb = MultiBoxEncoder(opt)\n",
        "\n",
        "    print(mb.default_boxes[:10])\n",
        "    "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jHvrl8jPWN54"
      },
      "source": [
        "## SSD Loss\n",
        "### 아래 사진을 보시면 아시다시피 원래 데이터 형태를 바꿔서 loss에 적용하게 됩니다.\n",
        "![image.png](data:image/png;base64,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)\n",
        "![image.png](data:image/png;base64,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)![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LvE4XTSrWQdj"
      },
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "\n",
        "def hard_negtives(logits, labels, pos, neg_radio): # default box negative : positive = neg_ratio : 1 (3 : 1)\n",
        "\n",
        "    num_batch, num_anchors, num_classes = logits.shape\n",
        "    logits = logits.view(-1, num_classes)\n",
        "    labels = labels.view(-1)\n",
        "\n",
        "    losses = F.cross_entropy(logits, labels, reduction='none') # Jacovian Matrix loss를 만들어서 hard negative 시행.\n",
        "\n",
        "    losses = losses.view(num_batch, num_anchors) # batch 갯수, anchor 갯수(image에서 object 수)로 view\n",
        "\n",
        "    losses[pos] = 0 # Positive (물체가 있는 경우 이 메서드 에서는 positive인 경우는 필요 없다.)\n",
        "\n",
        "    loss_idx = losses.argsort(1, descending=True) # batch, n 에서 n 기준 sort index 가져오기.\n",
        "    rank = loss_idx.argsort(1) #(batch, n) # index 를 sorting한 index 가져오기.\n",
        "\n",
        "    num_pos = pos.long().sum(1, keepdim=True) # positive의 총 개수. (pos는 true false 형태의 (Batch, anchor_num)으로 들어온다.)\n",
        "    # keepdim True로 했기 때문에 (BS, num_anchors) 형태는 그대로지만 전체 값이 positive의 갯수로 지정된다.\n",
        "    num_neg = torch.clamp(neg_radio*num_pos, max=pos.shape[1]-1) # positive의 3배. pos shape은 anchor_num을 의미한다. 그니까 예측된 anchor(한 사진에서의 예측한 bb들.)의 수에 맞게 3배로 늘려줌.\n",
        "    # 만약 positive의 갯수에 3배 한 수가 한 사진에서의 object 예측 수를 넘으면 object 예측 수에 맞춘다.\n",
        "    neg = rank < num_neg.expand_as(rank) # 마스크 생성. (loss가 높은 것 부터 지정해줌.)\n",
        "    return neg\n",
        "    \n",
        "class MultiBoxLoss(nn.Module):\n",
        "\n",
        "    def __init__(self, num_classes=7, neg_radio=3):\n",
        "        super(MultiBoxLoss, self).__init__()\n",
        "        self.num_classes = num_classes\n",
        "        self.neg_radio = neg_radio\n",
        "    \n",
        "    def forward(self, pred_loc, pred_label, gt_loc, gt_label):\n",
        "        '''\n",
        "        pred_loc : (batch, anchor_num, 4)\n",
        "        pred_label : (batch, anchor_num, num_classes)\n",
        "        gt_loc : (batch, anchor_num, 4)\n",
        "        gt_label : (batch, anchor_num)\n",
        "        '''\n",
        "\n",
        "        num_batch = pred_loc.shape[0]\n",
        "\n",
        "        pos_idx = gt_label > 0\n",
        "        pos_loc_idx = pos_idx.unsqueeze(2).expand_as(pred_loc)\n",
        "        pred_loc_pos = pred_loc[pos_loc_idx].view(-1, 4)\n",
        "        gt_loc_pos = gt_loc[pos_loc_idx].view(-1, 4)\n",
        "\n",
        "        loc_loss = F.smooth_l1_loss(pred_loc_pos, gt_loc_pos, reduction='sum')\n",
        "\n",
        "        logits = pred_label.detach()\n",
        "        labels = gt_label.detach()\n",
        "        neg_idx = hard_negtives(logits, labels, pos_idx, self.neg_radio) #neg (batch, n)\n",
        "\n",
        "        pos_cls_mask = pos_idx.unsqueeze(2).expand_as(pred_label) # positive와 negative에 대해 mask 생성 (비율을 맞춰주기 위함.)\n",
        "        neg_cls_mask = neg_idx.unsqueeze(2).expand_as(pred_label)\n",
        "\n",
        "        conf_p = pred_label[(pos_cls_mask+neg_cls_mask).gt(0)].view(-1, self.num_classes)\n",
        "        target = gt_label[(pos_idx+neg_idx).gt(0)]\n",
        "\n",
        "        cls_loss = F.cross_entropy(conf_p, target, reduction='sum')\n",
        "        N = pos_idx.long().sum()\n",
        "\n",
        "        loc_loss /= N\n",
        "        cls_loss /= N\n",
        "\n",
        "\n",
        "        return loc_loss, cls_loss\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vjAiu77xVdwu"
      },
      "source": [
        "## Augmentation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Hyq5PxDNVVAb"
      },
      "source": [
        "import cv2\n",
        "import numpy as np\n",
        "import random\n",
        "\n",
        "\n",
        "def _crop(image, boxes, labels):\n",
        "    height, width, _ = image.shape\n",
        "\n",
        "    if len(boxes) == 0:\n",
        "        return image, boxes, labels\n",
        "\n",
        "    while True:\n",
        "        mode = random.choice((\n",
        "            None,\n",
        "            (0.1, None), # min_iou 선택\n",
        "            (0.3, None),\n",
        "            (0.7, None),\n",
        "            (0.9, None),\n",
        "            (None, None), # -inf, inf : min_iou, max_iou\n",
        "        )) # random으로 mode = None, mode = (0.1, None) ... 선택\n",
        "\n",
        "        if mode is None: # single None 나오면 크롭없이 반환\n",
        "            return image, boxes, labels\n",
        "\n",
        "        min_iou, max_iou = mode\n",
        "        if min_iou is None:\n",
        "            min_iou = float('-inf')\n",
        "        if max_iou is None:\n",
        "            max_iou = float('inf')\n",
        "\n",
        "        for _ in range(50):\n",
        "            w = random.randrange(int(0.3 * width), width)\n",
        "            h = random.randrange(int(0.3 * height), height)\n",
        "\n",
        "            if h / w < 0.5 or 2 < h / w:\n",
        "                continue # 이미지 크기의 비율은 2:1, 1:2 가 넘지 않게\n",
        "\n",
        "            l = random.randrange(width - w)\n",
        "            t = random.randrange(height - h)\n",
        "            roi = np.array((l, t, l + w, t + h))\n",
        "\n",
        "            iou = bbox_iou(boxes, roi[np.newaxis])\n",
        "            if not (min_iou <= iou.min() and iou.max() <= max_iou):\n",
        "                continue\n",
        "\n",
        "            image = image[roi[1]:roi[3], roi[0]:roi[2]]\n",
        "\n",
        "            centers = (boxes[:, :2] + boxes[:, 2:]) / 2\n",
        "            mask = np.logical_and(roi[:2] < centers, centers < roi[2:]) \\\n",
        "                     .all(axis=1)\n",
        "            boxes = boxes[mask].copy()\n",
        "            labels = labels[mask]\n",
        "\n",
        "            boxes[:, :2] = np.maximum(boxes[:, :2], roi[:2])\n",
        "            boxes[:, :2] -= roi[:2]\n",
        "            boxes[:, 2:] = np.minimum(boxes[:, 2:], roi[2:])\n",
        "            boxes[:, 2:] -= roi[:2]\n",
        "\n",
        "            return image, boxes, labels\n",
        "\n",
        "\n",
        "def _distort(image):\n",
        "    def _convert(image, alpha=1, beta=0):\n",
        "        tmp = image.astype(float) * alpha + beta\n",
        "        tmp[tmp < 0] = 0\n",
        "        tmp[tmp > 255] = 255\n",
        "        image[:] = tmp\n",
        "\n",
        "    image = image.copy()\n",
        "\n",
        "    if random.randrange(2):\n",
        "        _convert(image, beta=random.uniform(-32, 32))\n",
        "\n",
        "    if random.randrange(2):\n",
        "        _convert(image, alpha=random.uniform(0.5, 1.5))\n",
        "\n",
        "    image = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)\n",
        "\n",
        "    if random.randrange(2):\n",
        "        tmp = image[:, :, 0].astype(int) + random.randint(-18, 18)\n",
        "        tmp %= 180\n",
        "        image[:, :, 0] = tmp\n",
        "\n",
        "    if random.randrange(2):\n",
        "        _convert(image[:, :, 1], alpha=random.uniform(0.5, 1.5))\n",
        "\n",
        "    image = cv2.cvtColor(image, cv2.COLOR_HSV2BGR)\n",
        "\n",
        "    return image\n",
        "\n",
        "\n",
        "def _expand(image, boxes, fill):\n",
        "    if random.randrange(2):\n",
        "        return image, boxes\n",
        "\n",
        "    height, width, depth = image.shape\n",
        "    ratio = random.uniform(1, 4)\n",
        "    left = random.randint(0, int(width * ratio) - width)\n",
        "    top = random.randint(0, int(height * ratio) - height)\n",
        "\n",
        "    expand_image = np.empty(\n",
        "        (int(height * ratio), int(width * ratio), depth),\n",
        "        dtype=image.dtype)\n",
        "    expand_image[:, :] = fill\n",
        "    expand_image[top:top + height, left:left + width] = image\n",
        "    image = expand_image\n",
        "\n",
        "    boxes = boxes.copy()\n",
        "    boxes[:, :2] += (left, top)\n",
        "    boxes[:, 2:] += (left, top)\n",
        "\n",
        "    return image, boxes\n",
        "\n",
        "\n",
        "def _mirror(image, boxes):\n",
        "    _, width, _ = image.shape\n",
        "    if random.randrange(2):\n",
        "        image = image[:, ::-1]\n",
        "        boxes = boxes.copy()\n",
        "        boxes[:, 0::2] = width - boxes[:, 2::-2]\n",
        "    return image, boxes\n",
        "\n",
        "\n",
        "def preproc_for_test(image, insize, mean):\n",
        "    image = cv2.resize(image, (insize, insize))\n",
        "    image = image.astype(np.float32)\n",
        "    image -= mean\n",
        "    return image.transpose(2, 0 , 1)\n",
        "\n",
        "\n",
        "def preproc_for_train(image, boxes, labels, insize, mean): # transformation\n",
        "    if len(boxes) == 0:\n",
        "        boxes = np.empty((0, 4))\n",
        "\n",
        "    image, boxes, labels = _crop(image, boxes, labels)\n",
        "    image = _distort(image)\n",
        "    image, boxes = _expand(image, boxes, mean)\n",
        "    image, boxes = _mirror(image, boxes)\n",
        "\n",
        "    height, width, _ = image.shape\n",
        "    image = preproc_for_test(image, insize, mean)\n",
        "    boxes = boxes.copy()\n",
        "    boxes[:, 0::2] /= width\n",
        "    boxes[:, 1::2] /= height\n",
        "\n",
        "    return image, boxes, labels\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ouylJZSIVSZ6"
      },
      "source": [
        "# Dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "U821o9BtKP6q"
      },
      "source": [
        "import torch\n",
        "import torch.utils.data as data\n",
        "import xml.etree.ElementTree as ET\n",
        "import numpy as np\n",
        "import glob\n",
        "import os\n",
        "import cv2\n",
        "\n",
        "\"\"\"\n",
        "    bbox 좌표 데이터는 0~ 1사이의 값이다.\n",
        "\"\"\"\n",
        "\n",
        "PERSON_LABELS = (\n",
        "        'Person',\n",
        "        'stroller',\n",
        "        'blinder',\n",
        "        'wheelchair',\n",
        "        'child',\n",
        "        'merchant',\n",
        "    )\n",
        "\n",
        "# UHD (가로: 3840, 세로: 2160) 해상도\n",
        "\"\"\"\n",
        "    file name 에 대한 List를 가져와서 이를 토대로 img, ano 에 대한 정보 모두를 가져옵니다.\n",
        "\"\"\"\n",
        "\n",
        "class CCTVDetection(data.Dataset):\n",
        "\n",
        "\n",
        "    def __init__(self,is_train=True):\n",
        "\n",
        "        self.is_train = is_train\n",
        "        self.opt = opt\n",
        "        self.ids = []\n",
        "        root = '/content/drive/MyDrive/ImageTrackingStudy'\n",
        "        if self.is_train:\n",
        "            img_file = os.path.join(root,'train_img/*')\n",
        "            ano_file = os.path.join(root,'train_label/*')\n",
        "            file_list = glob.glob(img_file)\n",
        "            file_list_img = [file for file in file_list if file.endswith(\".jpg\")]\n",
        "\n",
        "            for i in file_list_img:\n",
        "                # file_name = i[14:-4]\n",
        "                file_name = os.path.split(i)[1].split('.')[0]\n",
        "                img = f\"{root}/train_img/{file_name}.jpg\"\n",
        "                ano = f\"{root}/train_label/{file_name}.txt\"\n",
        "                # ano = os.path.join(\"dataset/train_label\",file_name + '.txt')\n",
        "                # img = os.path.join(\"dataset/train_img\",file_name + '.jpg')\n",
        "                \n",
        "                if os.path.isfile(ano) and os.path.isfile(img):\n",
        "                    self.ids.append((img, ano))\n",
        "        else:\n",
        "            img_file = os.path.join(root,'test_img/*')\n",
        "            file_list = glob.glob(img_file)\n",
        "            file_list_img = [file for file in file_list if file.endswith(\".jpg\")]\n",
        "            for i in file_list_img:\n",
        "                self.ids.append((i))\n",
        "            \n",
        "            \n",
        "    def __getitem__(self, index):\n",
        "\n",
        "        if self.is_train:\n",
        "            img_path, ano_path = self.ids[index] # img, ano path\n",
        "            image = cv2.imread(img_path, cv2.IMREAD_COLOR)\n",
        "            boxes, labels = self.get_annotations(ano_path) # txt 파일에서 가져옴.\n",
        "\n",
        "            if self.is_train:\n",
        "                image, boxes, labels = preproc_for_train(image, boxes, labels, opt.min_size, opt.mean)\n",
        "                image = torch.from_numpy(image)\n",
        "\n",
        "            target = np.concatenate([boxes, labels.reshape(-1,1)], axis=1)\n",
        "\n",
        "            return image, target\n",
        "        else:\n",
        "            img_path = self.ids[index]\n",
        "            image = cv2.imread(img_path, cv2.IMREAD_COLOR)\n",
        "            return image\n",
        "        # 이미지 당 여러개의 boxes가 존재하고 이에 대한 여러 labels가 존재한다.\n",
        "        # 나중에 dataloader를 통해 batch를 형성하게 되면 (BS, N, 4) 이렇게 나온다.\n",
        "    def get_annotations(self, path):\n",
        "        \n",
        "        f = open(path, 'r')\n",
        "        det = f.readlines()\n",
        "        boxes = []\n",
        "        labels = []\n",
        "        for d in det:\n",
        "            obj = d.split(' ')\n",
        "            label = int(obj[0]) # label\n",
        "            box = [float(obj[1]),float(obj[2]),float(obj[1])+float(obj[3]),float(obj[2])+float(obj[4])] # x1, y1, x2, y2\n",
        "            boxes.append(box)\n",
        "            labels.append(label)\n",
        "        return np.array(boxes), np.array(labels)\n",
        "            \n",
        "\n",
        "        \n",
        "\n",
        "    def __len__(self):\n",
        "        return len(self.ids)\n",
        "        \n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SarG1vX7V6Jm"
      },
      "source": [
        "# Train"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "CX9NuZ5vhdup",
        "outputId": "0c618796-b16e-4c7f-9b44-b3aadae1196d"
      },
      "source": [
        "import numpy as np\n",
        "import torch\n",
        "import torch.nn.functional as F\n",
        "import os\n",
        "\n",
        "\n",
        "# device = torch.device('cuda')\n",
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
        "\n",
        "def adjust_learning_rate(optimizer, gamma, step):\n",
        "    \"\"\"Sets the learning rate to the initial LR decayed by 10 at every\n",
        "        specified step\n",
        "    \"\"\"\n",
        "\n",
        "    lr = opt.lr * (gamma ** (step))\n",
        "    print('change learning rate, now learning rate is :', lr)\n",
        "    for param_group in optimizer.param_groups:\n",
        "        param_group['lr'] = lr\n",
        "        \n",
        "def weights_init_normal(m):\n",
        "    classname = m.__class__.__name__\n",
        "    if classname.find(\"Conv\") != -1:\n",
        "        torch.nn.init.normal_(m.weight.data, 0.0, 0.02)\n",
        "    elif classname.find(\"BatchNorm2d\") != -1:\n",
        "        torch.nn.init.normal_(m.weight.data, 1.0, 0.02)\n",
        "        torch.nn.init.constant_(m.bias.data, 0.0)\n",
        "\n",
        "\n",
        "if __name__ == '__main__':\n",
        "\n",
        "    print('now runing on device : ', device)\n",
        "    if not os.path.exists(opt.save_folder):\n",
        "        os.mkdir(opt.save_folder)\n",
        "    \n",
        "    # SSD 모델 선언\n",
        "\n",
        "    model = SSD(opt.num_classes, opt.anchor_num) \n",
        "    # 전에 진행한 적이 있으면 checkpoint가 있는지 확인 해 주자.\n",
        "    if opt.resume:\n",
        "        print('loading checkpoint...')\n",
        "        model.load_state_dict(torch.load(opt.resume))\n",
        "    else: # 아니면 vgg weights를 pretrain 된 weights를 가져옴.\n",
        "        vgg_weights = torch.load(opt.save_folder + opt.basenet)\n",
        "        print('Loading base network...')\n",
        "        model.vgg.load_state_dict(vgg_weights)\n",
        "\n",
        "     \n",
        "    model.to(device)\n",
        "    model.train()\n",
        "    # Multibox Encoder\n",
        "    mb = MultiBoxEncoder(opt)\n",
        "    # Datset\n",
        "    dataset = CCTVDetection(True)\n",
        "    # Data 들의 크기가 너무 커서 (UHD) 불러오는데 시간이 많이 걸리는 것 같다.\n",
        "    dataloader = torch.utils.data.DataLoader(dataset, batch_size=opt.batch_size, collate_fn=detection_collate, num_workers=0)\n",
        "    \n",
        "    criterion = MultiBoxLoss(opt.num_classes, opt.neg_radio).to(device)\n",
        "    optimizer = torch.optim.SGD(model.parameters(), lr=opt.lr, momentum=opt.momentum,\n",
        "                          weight_decay=opt.weight_decay)\n",
        "\n",
        "    print('start training........')\n",
        "    for e in range(opt.epoch):\n",
        "        if e % opt.lr_reduce_epoch == 0:\n",
        "            adjust_learning_rate(optimizer, opt.gamma, e//opt.lr_reduce_epoch)\n",
        "        total_loc_loss = 0\n",
        "        total_cls_loss = 0\n",
        "        total_loss = 0\n",
        "        for i , (img, boxes) in enumerate(dataloader):\n",
        "            img = img.to(device)\n",
        "            gt_boxes = []\n",
        "            gt_labels = []\n",
        "            for box in boxes:\n",
        "                labels = box[:, 4]\n",
        "                box = box[:, :-1]\n",
        "                # train 형태에 맞게 encode (cx, cy, .. )\n",
        "                # default box와 미리 매칭 시켜주자. (iou 기준.)\n",
        "                match_loc, match_label = mb.encode(box, labels)\n",
        "            \n",
        "                gt_boxes.append(match_loc)\n",
        "                gt_labels.append(match_label)\n",
        "            \n",
        "            gt_boxes = torch.FloatTensor(gt_boxes).to(device)\n",
        "            gt_labels = torch.LongTensor(gt_labels).to(device)\n",
        "\n",
        "\n",
        "            p_loc, p_label = model(img)\n",
        "            \n",
        "            loc_loss, cls_loss = criterion(p_loc, p_label, gt_boxes, gt_labels)\n",
        "\n",
        "            loss = loc_loss + cls_loss\n",
        "\n",
        "            optimizer.zero_grad()\n",
        "            loss.backward()\n",
        "            optimizer.step()\n",
        "            \n",
        "            total_loc_loss += loc_loss.item()\n",
        "\n",
        "            total_cls_loss += cls_loss.item()\n",
        "            total_loss += loss.item()\n",
        "            if i % opt.log_fn == 0:\n",
        "                avg_loc = total_loc_loss / (i+1)\n",
        "                avg_cls = total_cls_loss / (i+1)\n",
        "                avg_loss = total_loss / (i+1)\n",
        "                print('epoch[{}] | batch_idx[{}] | loc_loss [{:.2f}] | cls_loss [{:.2f}] | avg_loss [{:.2f}]'.format(e, i, avg_loc, avg_cls, avg_loss))\n",
        "\n",
        "        if (e+1) % 1 == 0:\n",
        "            torch.save(model.state_dict(), os.path.join(opt.save_folder, 'loss-{:.2f}.pth'.format(total_loss)))\n",
        "\n",
        "\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "    VOC_ROOT = '/workspace/hakerton/simple-ssd-for-beginners/VOCdevkit'\n",
            "\n",
            "    num_classes = 7\n",
            "\n",
            "    resume = None\n",
            "\n",
            "    lr = 0.0005\n",
            "\n",
            "    batch_size = 64 \n",
            "\n",
            "    momentum = 0.9\n",
            "\n",
            "    weight_decay = 5e-4\n",
            "\n",
            "    epoch = 100 \n",
            "\n",
            "    gamma = 0.2\n",
            "\n",
            "    lr_reduce_epoch = 30\n",
            "\n",
            "    save_folder = '/content/drive/MyDrive/ImageTrackingStudy/weights/'\n",
            "\n",
            "    basenet = 'vgg16_reducedfc.pth'\n",
            "\n",
            "    log_fn = 1\n",
            "\n",
            "    neg_radio = 3\n",
            "\n",
            "    min_size = 300\n",
            "\n",
            "    grids = (38, 19, 10, 5, 3, 1)\n",
            "\n",
            "    aspect_ratios = ((2,), (2, 3), (2, 3), (2, 3), (2,), (2,)) # ratio.\n",
            "\n",
            "    steps = [s / 300 for s in (8, 16, 32, 64, 100, 300)]\n",
            "\n",
            "    sizes = [s / 300 for s in (30, 60, 111, 162, 213, 264, 315)] # default box size의 기준. np.sqrt(sizes[k] * sizes[k+1])\n",
            "\n",
            "    anchor_num = [4, 6, 6, 6, 4, 4]\n",
            "\n",
            "    mean = (104, 117, 123)\n",
            "\n",
            "    variance = (0.1, 0.2)\n",
            "\n",
            "opt = Config()\n",
            "\n",
            "now runing on device :  cuda\n",
            "Loading base network...\n",
            "start training........\n",
            "change learning rate, now learning rate is : 0.0005\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/content/lib/utils.py:25: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
            "  return torch.stack(imgs), np.array(targets)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "epoch[0] | batch_idx[0] | loc_loss [3.08] | cls_loss [12.39] | avg_loss [15.47]\n",
            "epoch[0] | batch_idx[1] | loc_loss [3.06] | cls_loss [11.79] | avg_loss [14.84]\n",
            "epoch[0] | batch_idx[2] | loc_loss [3.08] | cls_loss [11.22] | avg_loss [14.30]\n",
            "epoch[0] | batch_idx[3] | loc_loss [2.99] | cls_loss [10.74] | avg_loss [13.73]\n",
            "epoch[0] | batch_idx[4] | loc_loss [2.95] | cls_loss [10.35] | avg_loss [13.30]\n",
            "epoch[0] | batch_idx[5] | loc_loss [2.90] | cls_loss [9.96] | avg_loss [12.86]\n",
            "epoch[0] | batch_idx[6] | loc_loss [2.86] | cls_loss [9.69] | avg_loss [12.55]\n",
            "epoch[0] | batch_idx[7] | loc_loss [2.84] | cls_loss [9.49] | avg_loss [12.33]\n",
            "epoch[0] | batch_idx[8] | loc_loss [2.80] | cls_loss [9.30] | avg_loss [12.10]\n",
            "epoch[0] | batch_idx[9] | loc_loss [2.77] | cls_loss [9.16] | avg_loss [11.93]\n",
            "epoch[0] | batch_idx[10] | loc_loss [2.75] | cls_loss [9.02] | avg_loss [11.77]\n",
            "epoch[0] | batch_idx[11] | loc_loss [2.72] | cls_loss [8.90] | avg_loss [11.62]\n",
            "epoch[0] | batch_idx[12] | loc_loss [2.69] | cls_loss [8.77] | avg_loss [11.46]\n",
            "epoch[0] | batch_idx[13] | loc_loss [2.67] | cls_loss [8.70] | avg_loss [11.37]\n",
            "epoch[0] | batch_idx[14] | loc_loss [2.65] | cls_loss [8.67] | avg_loss [11.31]\n",
            "epoch[0] | batch_idx[15] | loc_loss [2.64] | cls_loss [8.65] | avg_loss [11.29]\n",
            "epoch[0] | batch_idx[16] | loc_loss [2.62] | cls_loss [8.58] | avg_loss [11.21]\n",
            "epoch[0] | batch_idx[17] | loc_loss [2.62] | cls_loss [8.54] | avg_loss [11.16]\n",
            "epoch[0] | batch_idx[18] | loc_loss [2.59] | cls_loss [8.50] | avg_loss [11.09]\n",
            "epoch[0] | batch_idx[19] | loc_loss [2.58] | cls_loss [8.46] | avg_loss [11.04]\n",
            "epoch[0] | batch_idx[20] | loc_loss [2.56] | cls_loss [8.44] | avg_loss [11.00]\n",
            "epoch[0] | batch_idx[21] | loc_loss [2.56] | cls_loss [8.42] | avg_loss [10.98]\n",
            "epoch[0] | batch_idx[22] | loc_loss [2.55] | cls_loss [8.39] | avg_loss [10.93]\n",
            "epoch[0] | batch_idx[23] | loc_loss [2.54] | cls_loss [8.36] | avg_loss [10.90]\n",
            "epoch[0] | batch_idx[24] | loc_loss [2.54] | cls_loss [8.33] | avg_loss [10.86]\n",
            "epoch[0] | batch_idx[25] | loc_loss [2.53] | cls_loss [8.30] | avg_loss [10.83]\n",
            "epoch[0] | batch_idx[26] | loc_loss [2.52] | cls_loss [8.26] | avg_loss [10.78]\n",
            "epoch[0] | batch_idx[27] | loc_loss [2.52] | cls_loss [8.23] | avg_loss [10.75]\n",
            "epoch[0] | batch_idx[28] | loc_loss [2.51] | cls_loss [8.20] | avg_loss [10.72]\n",
            "epoch[0] | batch_idx[29] | loc_loss [2.50] | cls_loss [8.17] | avg_loss [10.67]\n",
            "epoch[0] | batch_idx[30] | loc_loss [2.50] | cls_loss [8.15] | avg_loss [10.64]\n",
            "epoch[0] | batch_idx[31] | loc_loss [2.50] | cls_loss [8.13] | avg_loss [10.63]\n",
            "epoch[0] | batch_idx[32] | loc_loss [2.50] | cls_loss [8.12] | avg_loss [10.62]\n",
            "epoch[0] | batch_idx[33] | loc_loss [2.50] | cls_loss [8.10] | avg_loss [10.60]\n",
            "epoch[0] | batch_idx[34] | loc_loss [2.49] | cls_loss [8.09] | avg_loss [10.58]\n",
            "epoch[0] | batch_idx[35] | loc_loss [2.49] | cls_loss [8.08] | avg_loss [10.56]\n",
            "epoch[0] | batch_idx[36] | loc_loss [2.48] | cls_loss [8.06] | avg_loss [10.55]\n",
            "epoch[0] | batch_idx[37] | loc_loss [2.48] | cls_loss [8.04] | avg_loss [10.52]\n",
            "epoch[0] | batch_idx[38] | loc_loss [2.48] | cls_loss [8.03] | avg_loss [10.51]\n",
            "epoch[0] | batch_idx[39] | loc_loss [2.48] | cls_loss [8.00] | avg_loss [10.48]\n",
            "epoch[0] | batch_idx[40] | loc_loss [2.47] | cls_loss [7.98] | avg_loss [10.45]\n",
            "epoch[0] | batch_idx[41] | loc_loss [2.47] | cls_loss [7.96] | avg_loss [10.43]\n",
            "epoch[0] | batch_idx[42] | loc_loss [2.47] | cls_loss [7.94] | avg_loss [10.41]\n",
            "epoch[0] | batch_idx[43] | loc_loss [2.46] | cls_loss [7.93] | avg_loss [10.38]\n",
            "epoch[0] | batch_idx[44] | loc_loss [2.45] | cls_loss [7.91] | avg_loss [10.36]\n",
            "epoch[0] | batch_idx[45] | loc_loss [2.45] | cls_loss [7.91] | avg_loss [10.36]\n",
            "epoch[0] | batch_idx[46] | loc_loss [2.44] | cls_loss [7.89] | avg_loss [10.33]\n",
            "epoch[0] | batch_idx[47] | loc_loss [2.44] | cls_loss [7.87] | avg_loss [10.31]\n",
            "epoch[0] | batch_idx[48] | loc_loss [2.43] | cls_loss [7.86] | avg_loss [10.29]\n",
            "epoch[0] | batch_idx[49] | loc_loss [2.42] | cls_loss [7.84] | avg_loss [10.27]\n",
            "epoch[0] | batch_idx[50] | loc_loss [2.42] | cls_loss [7.84] | avg_loss [10.26]\n",
            "epoch[0] | batch_idx[51] | loc_loss [2.42] | cls_loss [7.82] | avg_loss [10.24]\n",
            "epoch[0] | batch_idx[52] | loc_loss [2.41] | cls_loss [7.81] | avg_loss [10.22]\n",
            "epoch[0] | batch_idx[53] | loc_loss [2.40] | cls_loss [7.79] | avg_loss [10.20]\n",
            "epoch[0] | batch_idx[54] | loc_loss [2.40] | cls_loss [7.78] | avg_loss [10.18]\n",
            "epoch[0] | batch_idx[55] | loc_loss [2.40] | cls_loss [7.77] | avg_loss [10.16]\n",
            "epoch[0] | batch_idx[56] | loc_loss [2.39] | cls_loss [7.75] | avg_loss [10.14]\n",
            "epoch[0] | batch_idx[57] | loc_loss [2.39] | cls_loss [7.73] | avg_loss [10.12]\n",
            "epoch[0] | batch_idx[58] | loc_loss [2.38] | cls_loss [7.72] | avg_loss [10.10]\n",
            "epoch[0] | batch_idx[59] | loc_loss [2.37] | cls_loss [7.71] | avg_loss [10.08]\n",
            "epoch[0] | batch_idx[60] | loc_loss [2.37] | cls_loss [7.70] | avg_loss [10.07]\n",
            "epoch[0] | batch_idx[61] | loc_loss [2.37] | cls_loss [7.69] | avg_loss [10.06]\n",
            "epoch[0] | batch_idx[62] | loc_loss [2.36] | cls_loss [7.69] | avg_loss [10.05]\n",
            "epoch[0] | batch_idx[63] | loc_loss [2.36] | cls_loss [7.68] | avg_loss [10.04]\n",
            "epoch[0] | batch_idx[64] | loc_loss [2.35] | cls_loss [7.67] | avg_loss [10.03]\n",
            "epoch[0] | batch_idx[65] | loc_loss [2.35] | cls_loss [7.67] | avg_loss [10.02]\n",
            "epoch[0] | batch_idx[66] | loc_loss [2.34] | cls_loss [7.65] | avg_loss [9.99]\n",
            "epoch[0] | batch_idx[67] | loc_loss [2.33] | cls_loss [7.64] | avg_loss [9.98]\n",
            "epoch[0] | batch_idx[68] | loc_loss [2.33] | cls_loss [7.64] | avg_loss [9.97]\n",
            "epoch[0] | batch_idx[69] | loc_loss [2.33] | cls_loss [7.63] | avg_loss [9.96]\n",
            "epoch[0] | batch_idx[70] | loc_loss [2.32] | cls_loss [7.62] | avg_loss [9.95]\n",
            "epoch[0] | batch_idx[71] | loc_loss [2.32] | cls_loss [7.61] | avg_loss [9.93]\n",
            "epoch[0] | batch_idx[72] | loc_loss [2.31] | cls_loss [7.60] | avg_loss [9.92]\n",
            "epoch[0] | batch_idx[73] | loc_loss [2.31] | cls_loss [7.60] | avg_loss [9.91]\n",
            "epoch[0] | batch_idx[74] | loc_loss [2.30] | cls_loss [7.59] | avg_loss [9.89]\n",
            "epoch[0] | batch_idx[75] | loc_loss [2.30] | cls_loss [7.58] | avg_loss [9.88]\n",
            "epoch[0] | batch_idx[76] | loc_loss [2.29] | cls_loss [7.57] | avg_loss [9.86]\n",
            "epoch[0] | batch_idx[77] | loc_loss [2.29] | cls_loss [7.56] | avg_loss [9.86]\n",
            "epoch[0] | batch_idx[78] | loc_loss [2.29] | cls_loss [7.56] | avg_loss [9.85]\n",
            "epoch[0] | batch_idx[79] | loc_loss [2.28] | cls_loss [7.55] | avg_loss [9.83]\n",
            "epoch[0] | batch_idx[80] | loc_loss [2.28] | cls_loss [7.54] | avg_loss [9.81]\n",
            "epoch[0] | batch_idx[81] | loc_loss [2.27] | cls_loss [7.52] | avg_loss [9.79]\n",
            "epoch[0] | batch_idx[82] | loc_loss [2.27] | cls_loss [7.51] | avg_loss [9.78]\n",
            "epoch[0] | batch_idx[83] | loc_loss [2.26] | cls_loss [7.50] | avg_loss [9.77]\n",
            "epoch[0] | batch_idx[84] | loc_loss [2.26] | cls_loss [7.50] | avg_loss [9.76]\n",
            "epoch[0] | batch_idx[85] | loc_loss [2.26] | cls_loss [7.49] | avg_loss [9.75]\n",
            "epoch[0] | batch_idx[86] | loc_loss [2.25] | cls_loss [7.48] | avg_loss [9.74]\n",
            "epoch[0] | batch_idx[87] | loc_loss [2.25] | cls_loss [7.47] | avg_loss [9.72]\n",
            "epoch[0] | batch_idx[88] | loc_loss [2.24] | cls_loss [7.46] | avg_loss [9.70]\n",
            "epoch[0] | batch_idx[89] | loc_loss [2.24] | cls_loss [7.45] | avg_loss [9.69]\n",
            "epoch[0] | batch_idx[90] | loc_loss [2.23] | cls_loss [7.44] | avg_loss [9.68]\n",
            "epoch[0] | batch_idx[91] | loc_loss [2.23] | cls_loss [7.43] | avg_loss [9.67]\n",
            "epoch[0] | batch_idx[92] | loc_loss [2.23] | cls_loss [7.42] | avg_loss [9.65]\n",
            "epoch[0] | batch_idx[93] | loc_loss [2.22] | cls_loss [7.42] | avg_loss [9.64]\n",
            "epoch[0] | batch_idx[94] | loc_loss [2.22] | cls_loss [7.41] | avg_loss [9.63]\n",
            "epoch[0] | batch_idx[95] | loc_loss [2.21] | cls_loss [7.40] | avg_loss [9.62]\n",
            "epoch[0] | batch_idx[96] | loc_loss [2.21] | cls_loss [7.39] | avg_loss [9.60]\n",
            "epoch[0] | batch_idx[97] | loc_loss [2.21] | cls_loss [7.39] | avg_loss [9.59]\n",
            "epoch[0] | batch_idx[98] | loc_loss [2.20] | cls_loss [7.38] | avg_loss [9.58]\n",
            "epoch[0] | batch_idx[99] | loc_loss [2.20] | cls_loss [7.37] | avg_loss [9.57]\n",
            "epoch[0] | batch_idx[100] | loc_loss [2.20] | cls_loss [7.36] | avg_loss [9.56]\n",
            "epoch[0] | batch_idx[101] | loc_loss [2.19] | cls_loss [7.35] | avg_loss [9.54]\n",
            "epoch[0] | batch_idx[102] | loc_loss [2.19] | cls_loss [7.35] | avg_loss [9.53]\n",
            "epoch[0] | batch_idx[103] | loc_loss [2.18] | cls_loss [7.34] | avg_loss [9.52]\n",
            "epoch[0] | batch_idx[104] | loc_loss [2.18] | cls_loss [7.33] | avg_loss [9.51]\n",
            "epoch[0] | batch_idx[105] | loc_loss [2.17] | cls_loss [7.33] | avg_loss [9.50]\n",
            "epoch[0] | batch_idx[106] | loc_loss [2.17] | cls_loss [7.32] | avg_loss [9.49]\n",
            "epoch[0] | batch_idx[107] | loc_loss [2.16] | cls_loss [7.31] | avg_loss [9.48]\n",
            "epoch[0] | batch_idx[108] | loc_loss [2.16] | cls_loss [7.30] | avg_loss [9.46]\n",
            "epoch[0] | batch_idx[109] | loc_loss [2.15] | cls_loss [7.30] | avg_loss [9.45]\n",
            "epoch[0] | batch_idx[110] | loc_loss [2.15] | cls_loss [7.29] | avg_loss [9.45]\n",
            "epoch[0] | batch_idx[111] | loc_loss [2.15] | cls_loss [7.29] | avg_loss [9.44]\n",
            "epoch[0] | batch_idx[112] | loc_loss [2.15] | cls_loss [7.28] | avg_loss [9.42]\n",
            "epoch[0] | batch_idx[113] | loc_loss [2.14] | cls_loss [7.27] | avg_loss [9.41]\n",
            "epoch[0] | batch_idx[114] | loc_loss [2.14] | cls_loss [7.27] | avg_loss [9.40]\n",
            "epoch[0] | batch_idx[115] | loc_loss [2.13] | cls_loss [7.26] | avg_loss [9.39]\n",
            "epoch[0] | batch_idx[116] | loc_loss [2.13] | cls_loss [7.25] | avg_loss [9.39]\n",
            "epoch[0] | batch_idx[117] | loc_loss [2.13] | cls_loss [7.25] | avg_loss [9.38]\n",
            "epoch[0] | batch_idx[118] | loc_loss [2.13] | cls_loss [7.24] | avg_loss [9.37]\n",
            "epoch[0] | batch_idx[119] | loc_loss [2.12] | cls_loss [7.24] | avg_loss [9.36]\n",
            "epoch[0] | batch_idx[120] | loc_loss [2.12] | cls_loss [7.23] | avg_loss [9.34]\n",
            "epoch[0] | batch_idx[121] | loc_loss [2.11] | cls_loss [7.22] | avg_loss [9.33]\n",
            "epoch[0] | batch_idx[122] | loc_loss [2.11] | cls_loss [7.22] | avg_loss [9.33]\n",
            "epoch[0] | batch_idx[123] | loc_loss [2.11] | cls_loss [7.21] | avg_loss [9.32]\n",
            "epoch[0] | batch_idx[124] | loc_loss [2.10] | cls_loss [7.20] | avg_loss [9.31]\n",
            "epoch[0] | batch_idx[125] | loc_loss [2.10] | cls_loss [7.20] | avg_loss [9.30]\n",
            "epoch[0] | batch_idx[126] | loc_loss [2.11] | cls_loss [7.19] | avg_loss [9.30]\n",
            "epoch[0] | batch_idx[127] | loc_loss [2.11] | cls_loss [7.19] | avg_loss [9.29]\n",
            "epoch[0] | batch_idx[128] | loc_loss [2.11] | cls_loss [7.18] | avg_loss [9.29]\n",
            "epoch[0] | batch_idx[129] | loc_loss [2.11] | cls_loss [7.18] | avg_loss [9.29]\n",
            "epoch[0] | batch_idx[130] | loc_loss [2.11] | cls_loss [7.17] | avg_loss [9.28]\n",
            "epoch[0] | batch_idx[131] | loc_loss [2.10] | cls_loss [7.17] | avg_loss [9.27]\n",
            "epoch[0] | batch_idx[132] | loc_loss [2.10] | cls_loss [7.17] | avg_loss [9.27]\n",
            "epoch[0] | batch_idx[133] | loc_loss [2.10] | cls_loss [7.17] | avg_loss [9.27]\n",
            "epoch[0] | batch_idx[134] | loc_loss [2.10] | cls_loss [7.16] | avg_loss [9.26]\n",
            "epoch[0] | batch_idx[135] | loc_loss [2.10] | cls_loss [7.16] | avg_loss [9.25]\n",
            "epoch[0] | batch_idx[136] | loc_loss [2.10] | cls_loss [7.15] | avg_loss [9.25]\n",
            "epoch[0] | batch_idx[137] | loc_loss [2.09] | cls_loss [7.15] | avg_loss [9.24]\n",
            "epoch[0] | batch_idx[138] | loc_loss [2.09] | cls_loss [7.14] | avg_loss [9.24]\n",
            "epoch[0] | batch_idx[139] | loc_loss [2.09] | cls_loss [7.14] | avg_loss [9.23]\n",
            "epoch[0] | batch_idx[140] | loc_loss [2.09] | cls_loss [7.13] | avg_loss [9.22]\n",
            "epoch[0] | batch_idx[141] | loc_loss [2.09] | cls_loss [7.13] | avg_loss [9.22]\n",
            "epoch[0] | batch_idx[142] | loc_loss [2.09] | cls_loss [7.12] | avg_loss [9.21]\n",
            "epoch[0] | batch_idx[143] | loc_loss [2.09] | cls_loss [7.12] | avg_loss [9.21]\n",
            "epoch[0] | batch_idx[144] | loc_loss [2.08] | cls_loss [7.11] | avg_loss [9.20]\n",
            "epoch[0] | batch_idx[145] | loc_loss [2.08] | cls_loss [7.11] | avg_loss [9.19]\n",
            "epoch[0] | batch_idx[146] | loc_loss [2.08] | cls_loss [7.10] | avg_loss [9.18]\n",
            "epoch[0] | batch_idx[147] | loc_loss [2.08] | cls_loss [7.10] | avg_loss [9.17]\n",
            "epoch[0] | batch_idx[148] | loc_loss [2.08] | cls_loss [7.09] | avg_loss [9.17]\n",
            "epoch[0] | batch_idx[149] | loc_loss [2.08] | cls_loss [7.08] | avg_loss [9.16]\n",
            "epoch[0] | batch_idx[150] | loc_loss [2.08] | cls_loss [7.08] | avg_loss [9.15]\n",
            "epoch[0] | batch_idx[151] | loc_loss [2.07] | cls_loss [7.07] | avg_loss [9.15]\n",
            "epoch[0] | batch_idx[152] | loc_loss [2.07] | cls_loss [7.07] | avg_loss [9.14]\n",
            "epoch[0] | batch_idx[153] | loc_loss [2.07] | cls_loss [7.06] | avg_loss [9.13]\n",
            "epoch[0] | batch_idx[154] | loc_loss [2.07] | cls_loss [7.06] | avg_loss [9.12]\n",
            "epoch[0] | batch_idx[155] | loc_loss [2.06] | cls_loss [7.05] | avg_loss [9.12]\n",
            "epoch[0] | batch_idx[156] | loc_loss [2.06] | cls_loss [7.05] | avg_loss [9.11]\n",
            "epoch[0] | batch_idx[157] | loc_loss [2.06] | cls_loss [7.04] | avg_loss [9.11]\n",
            "epoch[0] | batch_idx[158] | loc_loss [2.06] | cls_loss [7.04] | avg_loss [9.10]\n",
            "epoch[0] | batch_idx[159] | loc_loss [2.06] | cls_loss [7.03] | avg_loss [9.09]\n",
            "epoch[0] | batch_idx[160] | loc_loss [2.06] | cls_loss [7.03] | avg_loss [9.08]\n",
            "epoch[0] | batch_idx[161] | loc_loss [2.05] | cls_loss [7.02] | avg_loss [9.07]\n",
            "epoch[0] | batch_idx[162] | loc_loss [2.05] | cls_loss [7.01] | avg_loss [9.07]\n",
            "epoch[0] | batch_idx[163] | loc_loss [2.05] | cls_loss [7.01] | avg_loss [9.06]\n",
            "epoch[0] | batch_idx[164] | loc_loss [2.04] | cls_loss [7.00] | avg_loss [9.05]\n",
            "epoch[0] | batch_idx[165] | loc_loss [2.04] | cls_loss [7.00] | avg_loss [9.04]\n",
            "epoch[0] | batch_idx[166] | loc_loss [2.04] | cls_loss [6.99] | avg_loss [9.03]\n",
            "epoch[0] | batch_idx[167] | loc_loss [2.03] | cls_loss [6.99] | avg_loss [9.02]\n",
            "epoch[0] | batch_idx[168] | loc_loss [2.03] | cls_loss [6.98] | avg_loss [9.01]\n",
            "epoch[0] | batch_idx[169] | loc_loss [2.03] | cls_loss [6.97] | avg_loss [9.00]\n",
            "epoch[0] | batch_idx[170] | loc_loss [2.03] | cls_loss [6.97] | avg_loss [8.99]\n",
            "epoch[0] | batch_idx[171] | loc_loss [2.02] | cls_loss [6.97] | avg_loss [8.99]\n",
            "epoch[0] | batch_idx[172] | loc_loss [2.02] | cls_loss [6.98] | avg_loss [9.00]\n",
            "epoch[0] | batch_idx[173] | loc_loss [2.02] | cls_loss [6.99] | avg_loss [9.01]\n",
            "epoch[0] | batch_idx[174] | loc_loss [2.01] | cls_loss [7.00] | avg_loss [9.02]\n",
            "epoch[0] | batch_idx[175] | loc_loss [2.01] | cls_loss [7.01] | avg_loss [9.02]\n",
            "epoch[0] | batch_idx[176] | loc_loss [2.01] | cls_loss [7.01] | avg_loss [9.02]\n",
            "epoch[0] | batch_idx[177] | loc_loss [2.01] | cls_loss [7.01] | avg_loss [9.01]\n",
            "epoch[0] | batch_idx[178] | loc_loss [2.01] | cls_loss [7.01] | avg_loss [9.01]\n",
            "epoch[0] | batch_idx[179] | loc_loss [2.01] | cls_loss [7.01] | avg_loss [9.01]\n",
            "epoch[0] | batch_idx[180] | loc_loss [2.00] | cls_loss [7.01] | avg_loss [9.01]\n",
            "epoch[0] | batch_idx[181] | loc_loss [2.00] | cls_loss [7.01] | avg_loss [9.01]\n",
            "epoch[0] | batch_idx[182] | loc_loss [2.00] | cls_loss [7.01] | avg_loss [9.01]\n",
            "epoch[0] | batch_idx[183] | loc_loss [2.00] | cls_loss [7.01] | avg_loss [9.01]\n",
            "epoch[0] | batch_idx[184] | loc_loss [2.00] | cls_loss [7.00] | avg_loss [9.00]\n",
            "epoch[0] | batch_idx[185] | loc_loss [2.00] | cls_loss [7.00] | avg_loss [9.00]\n",
            "epoch[0] | batch_idx[186] | loc_loss [2.00] | cls_loss [7.00] | avg_loss [9.00]\n",
            "epoch[0] | batch_idx[187] | loc_loss [2.00] | cls_loss [7.00] | avg_loss [8.99]\n",
            "epoch[0] | batch_idx[188] | loc_loss [2.00] | cls_loss [6.99] | avg_loss [8.99]\n",
            "epoch[0] | batch_idx[189] | loc_loss [2.00] | cls_loss [6.99] | avg_loss [8.99]\n",
            "epoch[0] | batch_idx[190] | loc_loss [1.99] | cls_loss [6.99] | avg_loss [8.98]\n",
            "epoch[0] | batch_idx[191] | loc_loss [1.99] | cls_loss [6.99] | avg_loss [8.98]\n",
            "epoch[0] | batch_idx[192] | loc_loss [1.99] | cls_loss [6.99] | avg_loss [8.97]\n",
            "epoch[0] | batch_idx[193] | loc_loss [1.98] | cls_loss [6.99] | avg_loss [8.97]\n",
            "epoch[0] | batch_idx[194] | loc_loss [1.98] | cls_loss [6.99] | avg_loss [8.96]\n",
            "epoch[0] | batch_idx[195] | loc_loss [1.98] | cls_loss [6.98] | avg_loss [8.96]\n",
            "epoch[0] | batch_idx[196] | loc_loss [1.97] | cls_loss [6.98] | avg_loss [8.95]\n",
            "epoch[0] | batch_idx[197] | loc_loss [1.97] | cls_loss [6.97] | avg_loss [8.94]\n",
            "epoch[0] | batch_idx[198] | loc_loss [1.97] | cls_loss [6.97] | avg_loss [8.94]\n",
            "epoch[0] | batch_idx[199] | loc_loss [1.97] | cls_loss [6.97] | avg_loss [8.94]\n",
            "epoch[0] | batch_idx[200] | loc_loss [1.96] | cls_loss [6.97] | avg_loss [8.93]\n",
            "epoch[0] | batch_idx[201] | loc_loss [1.96] | cls_loss [6.96] | avg_loss [8.93]\n",
            "epoch[0] | batch_idx[202] | loc_loss [1.96] | cls_loss [6.96] | avg_loss [8.92]\n",
            "epoch[0] | batch_idx[203] | loc_loss [1.96] | cls_loss [6.96] | avg_loss [8.91]\n",
            "epoch[0] | batch_idx[204] | loc_loss [1.95] | cls_loss [6.95] | avg_loss [8.91]\n",
            "epoch[0] | batch_idx[205] | loc_loss [1.95] | cls_loss [6.95] | avg_loss [8.90]\n",
            "epoch[0] | batch_idx[206] | loc_loss [1.95] | cls_loss [6.94] | avg_loss [8.89]\n",
            "epoch[0] | batch_idx[207] | loc_loss [1.95] | cls_loss [6.94] | avg_loss [8.89]\n",
            "epoch[0] | batch_idx[208] | loc_loss [1.95] | cls_loss [6.94] | avg_loss [8.89]\n",
            "epoch[0] | batch_idx[209] | loc_loss [1.95] | cls_loss [6.94] | avg_loss [8.89]\n",
            "epoch[0] | batch_idx[210] | loc_loss [1.94] | cls_loss [6.94] | avg_loss [8.88]\n",
            "epoch[0] | batch_idx[211] | loc_loss [1.94] | cls_loss [6.93] | avg_loss [8.88]\n",
            "epoch[0] | batch_idx[212] | loc_loss [1.94] | cls_loss [6.93] | avg_loss [8.87]\n",
            "epoch[0] | batch_idx[213] | loc_loss [1.94] | cls_loss [6.93] | avg_loss [8.87]\n",
            "epoch[0] | batch_idx[214] | loc_loss [1.94] | cls_loss [6.92] | avg_loss [8.86]\n",
            "epoch[0] | batch_idx[215] | loc_loss [1.93] | cls_loss [6.92] | avg_loss [8.85]\n",
            "epoch[0] | batch_idx[216] | loc_loss [1.93] | cls_loss [6.92] | avg_loss [8.85]\n",
            "epoch[0] | batch_idx[217] | loc_loss [1.93] | cls_loss [6.91] | avg_loss [8.84]\n",
            "epoch[0] | batch_idx[218] | loc_loss [1.93] | cls_loss [6.91] | avg_loss [8.84]\n",
            "epoch[0] | batch_idx[219] | loc_loss [1.93] | cls_loss [6.90] | avg_loss [8.83]\n",
            "epoch[0] | batch_idx[220] | loc_loss [1.93] | cls_loss [6.90] | avg_loss [8.83]\n",
            "epoch[0] | batch_idx[221] | loc_loss [1.93] | cls_loss [6.90] | avg_loss [8.83]\n",
            "epoch[0] | batch_idx[222] | loc_loss [1.93] | cls_loss [6.90] | avg_loss [8.82]\n",
            "epoch[0] | batch_idx[223] | loc_loss [1.93] | cls_loss [6.89] | avg_loss [8.82]\n",
            "epoch[0] | batch_idx[224] | loc_loss [1.93] | cls_loss [6.89] | avg_loss [8.82]\n",
            "epoch[0] | batch_idx[225] | loc_loss [1.93] | cls_loss [6.88] | avg_loss [8.81]\n",
            "epoch[0] | batch_idx[226] | loc_loss [1.92] | cls_loss [6.88] | avg_loss [8.80]\n",
            "epoch[0] | batch_idx[227] | loc_loss [1.92] | cls_loss [6.88] | avg_loss [8.80]\n",
            "epoch[0] | batch_idx[228] | loc_loss [1.92] | cls_loss [6.87] | avg_loss [8.79]\n",
            "epoch[0] | batch_idx[229] | loc_loss [1.92] | cls_loss [6.87] | avg_loss [8.79]\n",
            "epoch[0] | batch_idx[230] | loc_loss [1.92] | cls_loss [6.86] | avg_loss [8.78]\n",
            "epoch[0] | batch_idx[231] | loc_loss [1.92] | cls_loss [6.85] | avg_loss [8.77]\n",
            "epoch[0] | batch_idx[232] | loc_loss [1.92] | cls_loss [6.85] | avg_loss [8.76]\n",
            "epoch[0] | batch_idx[233] | loc_loss [1.92] | cls_loss [6.84] | avg_loss [8.76]\n",
            "epoch[0] | batch_idx[234] | loc_loss [1.92] | cls_loss [6.83] | avg_loss [8.75]\n",
            "epoch[0] | batch_idx[235] | loc_loss [1.92] | cls_loss [6.83] | avg_loss [8.74]\n",
            "epoch[0] | batch_idx[236] | loc_loss [1.91] | cls_loss [6.82] | avg_loss [8.73]\n",
            "epoch[0] | batch_idx[237] | loc_loss [1.91] | cls_loss [6.81] | avg_loss [8.73]\n",
            "epoch[0] | batch_idx[238] | loc_loss [1.91] | cls_loss [6.80] | avg_loss [8.71]\n",
            "epoch[0] | batch_idx[239] | loc_loss [1.91] | cls_loss [6.80] | avg_loss [8.70]\n",
            "epoch[0] | batch_idx[240] | loc_loss [1.91] | cls_loss [6.79] | avg_loss [8.69]\n",
            "epoch[0] | batch_idx[241] | loc_loss [1.90] | cls_loss [6.78] | avg_loss [8.68]\n",
            "epoch[0] | batch_idx[242] | loc_loss [1.90] | cls_loss [6.77] | avg_loss [8.67]\n",
            "epoch[0] | batch_idx[243] | loc_loss [1.90] | cls_loss [6.76] | avg_loss [8.66]\n",
            "epoch[0] | batch_idx[244] | loc_loss [1.90] | cls_loss [6.75] | avg_loss [8.65]\n",
            "epoch[0] | batch_idx[245] | loc_loss [1.90] | cls_loss [6.75] | avg_loss [8.64]\n",
            "epoch[0] | batch_idx[246] | loc_loss [1.90] | cls_loss [6.74] | avg_loss [8.63]\n",
            "epoch[0] | batch_idx[247] | loc_loss [1.89] | cls_loss [6.73] | avg_loss [8.62]\n",
            "epoch[0] | batch_idx[248] | loc_loss [1.89] | cls_loss [6.72] | avg_loss [8.61]\n",
            "epoch[0] | batch_idx[249] | loc_loss [1.89] | cls_loss [6.71] | avg_loss [8.60]\n",
            "epoch[0] | batch_idx[250] | loc_loss [1.89] | cls_loss [6.70] | avg_loss [8.59]\n",
            "epoch[0] | batch_idx[251] | loc_loss [1.89] | cls_loss [6.69] | avg_loss [8.58]\n",
            "epoch[0] | batch_idx[252] | loc_loss [1.89] | cls_loss [6.68] | avg_loss [8.57]\n",
            "epoch[0] | batch_idx[253] | loc_loss [1.89] | cls_loss [6.67] | avg_loss [8.56]\n",
            "epoch[0] | batch_idx[254] | loc_loss [1.89] | cls_loss [6.66] | avg_loss [8.55]\n",
            "epoch[0] | batch_idx[255] | loc_loss [1.89] | cls_loss [6.65] | avg_loss [8.54]\n",
            "epoch[0] | batch_idx[256] | loc_loss [1.89] | cls_loss [6.64] | avg_loss [8.52]\n",
            "epoch[0] | batch_idx[257] | loc_loss [1.88] | cls_loss [6.63] | avg_loss [8.51]\n",
            "epoch[0] | batch_idx[258] | loc_loss [1.88] | cls_loss [6.62] | avg_loss [8.50]\n",
            "epoch[0] | batch_idx[259] | loc_loss [1.88] | cls_loss [6.61] | avg_loss [8.49]\n",
            "epoch[0] | batch_idx[260] | loc_loss [1.88] | cls_loss [6.59] | avg_loss [8.48]\n",
            "epoch[0] | batch_idx[261] | loc_loss [1.88] | cls_loss [6.58] | avg_loss [8.47]\n",
            "epoch[0] | batch_idx[262] | loc_loss [1.88] | cls_loss [6.58] | avg_loss [8.46]\n",
            "epoch[0] | batch_idx[263] | loc_loss [1.88] | cls_loss [6.57] | avg_loss [8.45]\n",
            "epoch[0] | batch_idx[264] | loc_loss [1.88] | cls_loss [6.56] | avg_loss [8.44]\n",
            "epoch[0] | batch_idx[265] | loc_loss [1.88] | cls_loss [6.55] | avg_loss [8.43]\n",
            "epoch[0] | batch_idx[266] | loc_loss [1.87] | cls_loss [6.55] | avg_loss [8.43]\n",
            "epoch[0] | batch_idx[267] | loc_loss [1.87] | cls_loss [6.55] | avg_loss [8.42]\n",
            "epoch[0] | batch_idx[268] | loc_loss [1.87] | cls_loss [6.54] | avg_loss [8.42]\n",
            "epoch[0] | batch_idx[269] | loc_loss [1.87] | cls_loss [6.54] | avg_loss [8.41]\n",
            "epoch[0] | batch_idx[270] | loc_loss [1.87] | cls_loss [6.53] | avg_loss [8.40]\n",
            "epoch[0] | batch_idx[271] | loc_loss [1.87] | cls_loss [6.52] | avg_loss [8.39]\n",
            "epoch[0] | batch_idx[272] | loc_loss [1.87] | cls_loss [6.52] | avg_loss [8.39]\n",
            "epoch[0] | batch_idx[273] | loc_loss [1.87] | cls_loss [6.51] | avg_loss [8.38]\n",
            "epoch[0] | batch_idx[274] | loc_loss [1.87] | cls_loss [6.50] | avg_loss [8.37]\n",
            "epoch[0] | batch_idx[275] | loc_loss [1.87] | cls_loss [6.49] | avg_loss [8.36]\n",
            "epoch[0] | batch_idx[276] | loc_loss [1.87] | cls_loss [6.49] | avg_loss [8.35]\n",
            "epoch[0] | batch_idx[277] | loc_loss [1.87] | cls_loss [6.48] | avg_loss [8.35]\n",
            "epoch[0] | batch_idx[278] | loc_loss [1.87] | cls_loss [6.47] | avg_loss [8.34]\n",
            "epoch[0] | batch_idx[279] | loc_loss [1.87] | cls_loss [6.46] | avg_loss [8.33]\n",
            "epoch[0] | batch_idx[280] | loc_loss [1.87] | cls_loss [6.45] | avg_loss [8.32]\n",
            "epoch[0] | batch_idx[281] | loc_loss [1.87] | cls_loss [6.44] | avg_loss [8.31]\n",
            "epoch[0] | batch_idx[282] | loc_loss [1.87] | cls_loss [6.44] | avg_loss [8.31]\n",
            "epoch[0] | batch_idx[283] | loc_loss [1.87] | cls_loss [6.43] | avg_loss [8.30]\n",
            "epoch[0] | batch_idx[284] | loc_loss [1.86] | cls_loss [6.42] | avg_loss [8.29]\n",
            "epoch[0] | batch_idx[285] | loc_loss [1.86] | cls_loss [6.41] | avg_loss [8.28]\n",
            "epoch[0] | batch_idx[286] | loc_loss [1.86] | cls_loss [6.41] | avg_loss [8.27]\n",
            "epoch[0] | batch_idx[287] | loc_loss [1.86] | cls_loss [6.40] | avg_loss [8.26]\n",
            "epoch[1] | batch_idx[0] | loc_loss [1.47] | cls_loss [4.22] | avg_loss [5.68]\n",
            "epoch[1] | batch_idx[1] | loc_loss [1.50] | cls_loss [4.02] | avg_loss [5.53]\n",
            "epoch[1] | batch_idx[2] | loc_loss [1.48] | cls_loss [4.04] | avg_loss [5.52]\n",
            "epoch[1] | batch_idx[3] | loc_loss [1.51] | cls_loss [4.09] | avg_loss [5.59]\n",
            "epoch[1] | batch_idx[4] | loc_loss [1.54] | cls_loss [4.09] | avg_loss [5.63]\n",
            "epoch[1] | batch_idx[5] | loc_loss [1.53] | cls_loss [4.06] | avg_loss [5.59]\n",
            "epoch[1] | batch_idx[6] | loc_loss [1.53] | cls_loss [4.10] | avg_loss [5.63]\n",
            "epoch[1] | batch_idx[7] | loc_loss [1.57] | cls_loss [4.11] | avg_loss [5.68]\n",
            "epoch[1] | batch_idx[8] | loc_loss [1.56] | cls_loss [4.02] | avg_loss [5.58]\n",
            "epoch[1] | batch_idx[9] | loc_loss [1.59] | cls_loss [4.03] | avg_loss [5.62]\n",
            "epoch[1] | batch_idx[10] | loc_loss [1.59] | cls_loss [4.05] | avg_loss [5.64]\n",
            "epoch[1] | batch_idx[11] | loc_loss [1.60] | cls_loss [4.02] | avg_loss [5.61]\n",
            "epoch[1] | batch_idx[12] | loc_loss [1.59] | cls_loss [4.00] | avg_loss [5.58]\n",
            "epoch[1] | batch_idx[13] | loc_loss [1.58] | cls_loss [4.00] | avg_loss [5.58]\n",
            "epoch[1] | batch_idx[14] | loc_loss [1.59] | cls_loss [4.02] | avg_loss [5.61]\n",
            "epoch[1] | batch_idx[15] | loc_loss [1.59] | cls_loss [4.05] | avg_loss [5.64]\n",
            "epoch[1] | batch_idx[16] | loc_loss [1.58] | cls_loss [4.01] | avg_loss [5.59]\n",
            "epoch[1] | batch_idx[17] | loc_loss [1.58] | cls_loss [4.00] | avg_loss [5.58]\n",
            "epoch[1] | batch_idx[18] | loc_loss [1.57] | cls_loss [3.97] | avg_loss [5.54]\n",
            "epoch[1] | batch_idx[19] | loc_loss [1.55] | cls_loss [3.94] | avg_loss [5.49]\n",
            "epoch[1] | batch_idx[20] | loc_loss [1.56] | cls_loss [3.93] | avg_loss [5.49]\n",
            "epoch[1] | batch_idx[21] | loc_loss [1.55] | cls_loss [3.91] | avg_loss [5.46]\n",
            "epoch[1] | batch_idx[22] | loc_loss [1.53] | cls_loss [3.87] | avg_loss [5.41]\n",
            "epoch[1] | batch_idx[23] | loc_loss [1.54] | cls_loss [3.86] | avg_loss [5.40]\n",
            "epoch[1] | batch_idx[24] | loc_loss [1.55] | cls_loss [3.84] | avg_loss [5.39]\n",
            "epoch[1] | batch_idx[25] | loc_loss [1.55] | cls_loss [3.83] | avg_loss [5.37]\n",
            "epoch[1] | batch_idx[26] | loc_loss [1.53] | cls_loss [3.81] | avg_loss [5.34]\n",
            "epoch[1] | batch_idx[27] | loc_loss [1.54] | cls_loss [3.80] | avg_loss [5.34]\n",
            "epoch[1] | batch_idx[28] | loc_loss [1.52] | cls_loss [3.79] | avg_loss [5.31]\n",
            "epoch[1] | batch_idx[29] | loc_loss [1.53] | cls_loss [3.77] | avg_loss [5.30]\n",
            "epoch[1] | batch_idx[30] | loc_loss [1.52] | cls_loss [3.76] | avg_loss [5.28]\n",
            "epoch[1] | batch_idx[31] | loc_loss [1.52] | cls_loss [3.76] | avg_loss [5.28]\n",
            "epoch[1] | batch_idx[32] | loc_loss [1.52] | cls_loss [3.75] | avg_loss [5.27]\n",
            "epoch[1] | batch_idx[33] | loc_loss [1.53] | cls_loss [3.74] | avg_loss [5.27]\n",
            "epoch[1] | batch_idx[34] | loc_loss [1.52] | cls_loss [3.72] | avg_loss [5.23]\n",
            "epoch[1] | batch_idx[35] | loc_loss [1.51] | cls_loss [3.70] | avg_loss [5.22]\n",
            "epoch[1] | batch_idx[36] | loc_loss [1.52] | cls_loss [3.69] | avg_loss [5.21]\n",
            "epoch[1] | batch_idx[37] | loc_loss [1.52] | cls_loss [3.69] | avg_loss [5.20]\n",
            "epoch[1] | batch_idx[38] | loc_loss [1.51] | cls_loss [3.68] | avg_loss [5.19]\n",
            "epoch[1] | batch_idx[39] | loc_loss [1.52] | cls_loss [3.66] | avg_loss [5.18]\n",
            "epoch[1] | batch_idx[40] | loc_loss [1.51] | cls_loss [3.64] | avg_loss [5.16]\n",
            "epoch[1] | batch_idx[41] | loc_loss [1.51] | cls_loss [3.63] | avg_loss [5.14]\n",
            "epoch[1] | batch_idx[42] | loc_loss [1.51] | cls_loss [3.62] | avg_loss [5.12]\n",
            "epoch[1] | batch_idx[43] | loc_loss [1.50] | cls_loss [3.60] | avg_loss [5.10]\n",
            "epoch[1] | batch_idx[44] | loc_loss [1.50] | cls_loss [3.59] | avg_loss [5.09]\n",
            "epoch[1] | batch_idx[45] | loc_loss [1.51] | cls_loss [3.60] | avg_loss [5.11]\n",
            "epoch[1] | batch_idx[46] | loc_loss [1.50] | cls_loss [3.59] | avg_loss [5.09]\n",
            "epoch[1] | batch_idx[47] | loc_loss [1.50] | cls_loss [3.57] | avg_loss [5.07]\n",
            "epoch[1] | batch_idx[48] | loc_loss [1.49] | cls_loss [3.55] | avg_loss [5.05]\n",
            "epoch[1] | batch_idx[49] | loc_loss [1.49] | cls_loss [3.54] | avg_loss [5.03]\n",
            "epoch[1] | batch_idx[50] | loc_loss [1.49] | cls_loss [3.53] | avg_loss [5.02]\n",
            "epoch[1] | batch_idx[51] | loc_loss [1.49] | cls_loss [3.51] | avg_loss [5.00]\n",
            "epoch[1] | batch_idx[52] | loc_loss [1.49] | cls_loss [3.50] | avg_loss [4.99]\n",
            "epoch[1] | batch_idx[53] | loc_loss [1.48] | cls_loss [3.48] | avg_loss [4.96]\n",
            "epoch[1] | batch_idx[54] | loc_loss [1.48] | cls_loss [3.47] | avg_loss [4.94]\n",
            "epoch[1] | batch_idx[55] | loc_loss [1.47] | cls_loss [3.45] | avg_loss [4.92]\n",
            "epoch[1] | batch_idx[56] | loc_loss [1.47] | cls_loss [3.44] | avg_loss [4.91]\n",
            "epoch[1] | batch_idx[57] | loc_loss [1.47] | cls_loss [3.43] | avg_loss [4.90]\n",
            "epoch[1] | batch_idx[58] | loc_loss [1.47] | cls_loss [3.42] | avg_loss [4.89]\n",
            "epoch[1] | batch_idx[59] | loc_loss [1.46] | cls_loss [3.41] | avg_loss [4.87]\n",
            "epoch[1] | batch_idx[60] | loc_loss [1.47] | cls_loss [3.40] | avg_loss [4.86]\n",
            "epoch[1] | batch_idx[61] | loc_loss [1.47] | cls_loss [3.39] | avg_loss [4.85]\n",
            "epoch[1] | batch_idx[62] | loc_loss [1.47] | cls_loss [3.38] | avg_loss [4.85]\n",
            "epoch[1] | batch_idx[63] | loc_loss [1.46] | cls_loss [3.38] | avg_loss [4.84]\n",
            "epoch[1] | batch_idx[64] | loc_loss [1.46] | cls_loss [3.37] | avg_loss [4.83]\n",
            "epoch[1] | batch_idx[65] | loc_loss [1.46] | cls_loss [3.36] | avg_loss [4.81]\n",
            "epoch[1] | batch_idx[66] | loc_loss [1.45] | cls_loss [3.34] | avg_loss [4.80]\n",
            "epoch[1] | batch_idx[67] | loc_loss [1.45] | cls_loss [3.34] | avg_loss [4.79]\n",
            "epoch[1] | batch_idx[68] | loc_loss [1.46] | cls_loss [3.32] | avg_loss [4.78]\n",
            "epoch[1] | batch_idx[69] | loc_loss [1.45] | cls_loss [3.31] | avg_loss [4.77]\n",
            "epoch[1] | batch_idx[70] | loc_loss [1.45] | cls_loss [3.30] | avg_loss [4.75]\n",
            "epoch[1] | batch_idx[71] | loc_loss [1.45] | cls_loss [3.29] | avg_loss [4.74]\n",
            "epoch[1] | batch_idx[72] | loc_loss [1.44] | cls_loss [3.28] | avg_loss [4.72]\n",
            "epoch[1] | batch_idx[73] | loc_loss [1.44] | cls_loss [3.27] | avg_loss [4.71]\n",
            "epoch[1] | batch_idx[74] | loc_loss [1.44] | cls_loss [3.26] | avg_loss [4.70]\n",
            "epoch[1] | batch_idx[75] | loc_loss [1.43] | cls_loss [3.25] | avg_loss [4.68]\n",
            "epoch[1] | batch_idx[76] | loc_loss [1.43] | cls_loss [3.24] | avg_loss [4.67]\n",
            "epoch[1] | batch_idx[77] | loc_loss [1.43] | cls_loss [3.23] | avg_loss [4.66]\n",
            "epoch[1] | batch_idx[78] | loc_loss [1.42] | cls_loss [3.22] | avg_loss [4.65]\n",
            "epoch[1] | batch_idx[79] | loc_loss [1.42] | cls_loss [3.22] | avg_loss [4.64]\n",
            "epoch[1] | batch_idx[80] | loc_loss [1.42] | cls_loss [3.21] | avg_loss [4.63]\n",
            "epoch[1] | batch_idx[81] | loc_loss [1.41] | cls_loss [3.20] | avg_loss [4.61]\n",
            "epoch[1] | batch_idx[82] | loc_loss [1.41] | cls_loss [3.19] | avg_loss [4.60]\n",
            "epoch[1] | batch_idx[83] | loc_loss [1.41] | cls_loss [3.18] | avg_loss [4.59]\n",
            "epoch[1] | batch_idx[84] | loc_loss [1.41] | cls_loss [3.17] | avg_loss [4.58]\n",
            "epoch[1] | batch_idx[85] | loc_loss [1.40] | cls_loss [3.17] | avg_loss [4.57]\n",
            "epoch[1] | batch_idx[86] | loc_loss [1.40] | cls_loss [3.16] | avg_loss [4.56]\n",
            "epoch[1] | batch_idx[87] | loc_loss [1.40] | cls_loss [3.15] | avg_loss [4.55]\n",
            "epoch[1] | batch_idx[88] | loc_loss [1.39] | cls_loss [3.15] | avg_loss [4.54]\n",
            "epoch[1] | batch_idx[89] | loc_loss [1.39] | cls_loss [3.14] | avg_loss [4.53]\n",
            "epoch[1] | batch_idx[90] | loc_loss [1.39] | cls_loss [3.13] | avg_loss [4.52]\n",
            "epoch[1] | batch_idx[91] | loc_loss [1.39] | cls_loss [3.12] | avg_loss [4.51]\n",
            "epoch[1] | batch_idx[92] | loc_loss [1.39] | cls_loss [3.12] | avg_loss [4.50]\n",
            "epoch[1] | batch_idx[93] | loc_loss [1.39] | cls_loss [3.11] | avg_loss [4.50]\n",
            "epoch[1] | batch_idx[94] | loc_loss [1.38] | cls_loss [3.11] | avg_loss [4.49]\n",
            "epoch[1] | batch_idx[95] | loc_loss [1.38] | cls_loss [3.11] | avg_loss [4.49]\n",
            "epoch[1] | batch_idx[96] | loc_loss [1.38] | cls_loss [3.10] | avg_loss [4.48]\n",
            "epoch[1] | batch_idx[97] | loc_loss [1.38] | cls_loss [3.09] | avg_loss [4.47]\n",
            "epoch[1] | batch_idx[98] | loc_loss [1.38] | cls_loss [3.09] | avg_loss [4.47]\n",
            "epoch[1] | batch_idx[99] | loc_loss [1.37] | cls_loss [3.09] | avg_loss [4.46]\n",
            "epoch[1] | batch_idx[100] | loc_loss [1.37] | cls_loss [3.08] | avg_loss [4.45]\n",
            "epoch[1] | batch_idx[101] | loc_loss [1.36] | cls_loss [3.08] | avg_loss [4.44]\n",
            "epoch[1] | batch_idx[102] | loc_loss [1.36] | cls_loss [3.07] | avg_loss [4.43]\n",
            "epoch[1] | batch_idx[103] | loc_loss [1.36] | cls_loss [3.07] | avg_loss [4.42]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "error",
          "ename": "KeyboardInterrupt",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-3-1a63a19fd3e2>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     76\u001b[0m         \u001b[0mtotal_cls_loss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     77\u001b[0m         \u001b[0mtotal_loss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 78\u001b[0;31m         \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mimg\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mboxes\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32min\u001b[0m \u001b[0menumerate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataloader\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     79\u001b[0m             \u001b[0mimg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mimg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     80\u001b[0m             \u001b[0mgt_boxes\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py\u001b[0m in \u001b[0;36m__next__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    515\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sampler_iter\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    516\u001b[0m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_reset\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 517\u001b[0;31m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_next_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    518\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_num_yielded\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    519\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dataset_kind\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0m_DatasetKind\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mIterable\u001b[0m \u001b[0;32mand\u001b[0m\u001b[0;31m \u001b[0m\u001b[0;31m\\\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py\u001b[0m in \u001b[0;36m_next_data\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    555\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_next_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    556\u001b[0m         \u001b[0mindex\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_next_index\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m  \u001b[0;31m# may raise StopIteration\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 557\u001b[0;31m         \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dataset_fetcher\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfetch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mindex\u001b[0m\u001b[0;34m)\u001b[0m  \u001b[0;31m# may raise StopIteration\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    558\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_pin_memory\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    559\u001b[0m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_utils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpin_memory\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpin_memory\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/utils/data/_utils/fetch.py\u001b[0m in \u001b[0;36mfetch\u001b[0;34m(self, possibly_batched_index)\u001b[0m\n\u001b[1;32m     42\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mfetch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     43\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mauto_collation\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 44\u001b[0;31m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0midx\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0midx\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     45\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     46\u001b[0m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/utils/data/_utils/fetch.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m     42\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mfetch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     43\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mauto_collation\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 44\u001b[0;31m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0midx\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0midx\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     45\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     46\u001b[0m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m<ipython-input-2-6cef9367771b>\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, index)\u001b[0m\n\u001b[1;32m     61\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_train\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     62\u001b[0m             \u001b[0mimg_path\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mano_path\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mids\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mindex\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;31m# img, ano path\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 63\u001b[0;31m             \u001b[0mimage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcv2\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mimread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimg_path\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcv2\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mIMREAD_COLOR\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     64\u001b[0m             \u001b[0mboxes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlabels\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_annotations\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mano_path\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# txt 파일에서 가져옴.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     65\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7LExxjs2WhgX"
      },
      "source": [
        "## Evaluation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cpmaptWwKKEB"
      },
      "source": [
        "import csv\n",
        "from dataset import CCTVDetection\n",
        "from dataset import PERSON_LABELS\n",
        "from config import opt\n",
        "import numpy as np\n",
        "from lib.model import SSD\n",
        "import torch\n",
        "import torch.nn.functional as F\n",
        "import os \n",
        "from lib.multibox_encoder import MultiBoxEncoder\n",
        "from lib.ssd_loss import MultiBoxLoss\n",
        "import cv2\n",
        "from lib.utils import nms\n",
        "from lib.augmentations import preproc_for_test\n",
        "import matplotlib.pyplot as plt\n",
        "from lib.utils import detect\n",
        "import tqdm\n",
        "import os\n",
        "import argparse\n",
        "\n",
        "def convert_to_xyxy(imgWidth, imgHeight, xywh):\n",
        "    px = float(xywh[0])\n",
        "    py = float(xywh[1])\n",
        "    pw = float(xywh[2])\n",
        "    ph = float(xywh[3])\n",
        "    \n",
        "    cpx = px + pw/2\n",
        "    cpy = py + ph/2\n",
        "\n",
        "    abspx = cpx / imgWidth\n",
        "    abspy = cpy / imgHeight\n",
        "    abspw = pw / imgWidth\n",
        "    absph = ph / imgHeight  \n",
        "\n",
        "    abspx = 0.999999 if abspx >= 1 else abspx\n",
        "    abspy = 0.999999 if abspy >= 1 else abspy\n",
        "    abspw = 0.999999 if abspw >= 1 else abspw\n",
        "    absph = 0.999999 if absph >= 1 else absph\n",
        "    abspx = 0.000001 if abspx < 0 else abspx\n",
        "    abspy = 0.000001 if abspy < 0 else abspy\n",
        "    abspw = 0.000001 if abspw < 0 else abspw\n",
        "    absph = 0.000001 if absph < 0 else absph\n",
        "            \n",
        "    return abspx, abspy, abspw, absph\n",
        "\n",
        "parser = argparse.ArgumentParser()\n",
        "\n",
        "parser.add_argument('--model', \n",
        "                    default='',\n",
        "                    type=str,\n",
        "                    help='model checkpoint used to eval CCTV dataset')\n",
        "args = parser.parse_args()\n",
        "\n",
        "checkpoint = args.model\n",
        "\n",
        "\n",
        "device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')\n",
        "\n",
        "if __name__ == '__main__': \n",
        "    \n",
        "    print('using {} to eval, use cpu may take an hour to complete !!'.format(device))\n",
        "    model = SSD(opt.num_classes, opt.anchor_num)\n",
        "    \n",
        "    print('loading checkpoint from {}'.format(checkpoint))\n",
        "    state_dict = torch.load(checkpoint, map_location=None if torch.cuda.is_available() else 'cpu')\n",
        "    model.load_state_dict(state_dict)\n",
        "    model.to(device)\n",
        "    print('model loaded')\n",
        "    \n",
        "    multibox_encoder = MultiBoxEncoder(opt)\n",
        "    test_dataset = CCTVDetection(False)\n",
        "\n",
        "    print('start detect.........')\n",
        "    \n",
        "    f = open('pred.csv','w', newline='')\n",
        "    wr = csv.writer(f)\n",
        "    wr.writerow([\"ImageID\",\"LabelName\",\"Conf\",\"XMin\",\"XMax\",\"YMin\",\"YMax\"])\n",
        "    \n",
        "    for i in tqdm.tqdm(range(len(test_dataset))):\n",
        "\n",
        "        src = test_dataset[i]\n",
        "        img_name = os.path.basename(test_dataset.ids[i][19:-4])\n",
        "        image = preproc_for_test(src, opt.min_size, opt.mean)\n",
        "        image = torch.from_numpy(image).to(device)\n",
        "        with torch.no_grad():\n",
        "            loc, conf = model(image.unsqueeze(0))\n",
        "        loc = loc[0]\n",
        "        conf = conf[0]\n",
        "        conf = F.softmax(conf, dim=1)\n",
        "        conf = conf.cpu().numpy()\n",
        "        loc = loc.cpu().numpy()\n",
        "\n",
        "        decode_loc = multibox_encoder.decode(loc)\n",
        "        gt_boxes, gt_confs, gt_labels = detect(decode_loc, conf, nms_threshold=0.5, gt_threshold=0.5)\n",
        "\n",
        "        #no object detected\n",
        "        if len(gt_boxes) == 0:\n",
        "            continue\n",
        "\n",
        "        h, w = src.shape[:2]\n",
        "        gt_boxes[:, 0] = gt_boxes[:, 0] * w\n",
        "        gt_boxes[:, 1] = gt_boxes[:, 1] * h\n",
        "        gt_boxes[:, 2] = gt_boxes[:, 2] * w\n",
        "        gt_boxes[:, 3] = gt_boxes[:, 3] * h\n",
        "        \n",
        "        for box, label, score in zip(gt_boxes, gt_labels, gt_confs):\n",
        "            box[0],box[1],box[2],box[3] = convert_to_xyxy(w,h,[box[0],box[1],box[2],box[3]])\n",
        "            wr.writerow([img_name,list(PERSON_LABELS)[label],\"{:.3f}\".format(score),\"{:.8f}\".format(box[0]),\"{:.8f}\".format(box[1]),\"{:.8f}\".format(box[2]),\"{:.8f}\".format(box[3])])\n",
        "\n",
        "    \n",
        "    f.close()\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hNcspfWEhpFy"
      },
      "source": [
        "import torch\n",
        "\n",
        "pos = torch.ones((32,10))\n",
        "num_pos = pos.long().sum(1, keepdim=True)\n",
        "num_neg = torch.clamp(3*num_pos, max=pos.shape[1]-1) #(batch, 1)"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TFvl-xBCi_p5",
        "outputId": "f765d67f-645a-452d-8f02-a3e796d06ea8"
      },
      "source": [
        "num_pos"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "tensor([[10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10],\n",
              "        [10]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NhJSjSp0jJVd"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}