Run_model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "nvcc: NVIDIA (R) Cuda compiler driver\n",
      "Copyright (c) 2005-2018 NVIDIA Corporation\n",
      "Built on Sat_Aug_25_21:08:01_CDT_2018\n",
      "Cuda compilation tools, release 10.0, V10.0.130\n",
      "gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0\n",
      "Copyright (C) 2017 Free Software Foundation, Inc.\n",
      "This is free software; see the source for copying conditions.  There is NO\n",
      "warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Check nvcc version\n",
    "!nvcc -V\n",
    "# Check GCC version\n",
    "!gcc --version"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/home/esbg-lab/Desktop/P3/mmsegmentation\n"
     ]
    }
   ],
   "source": [
    "# !rm -rf mmsegmentation\n",
    "# !git clone https://github.com/open-mmlab/mmsegmentation.git \n",
    "%cd mmsegmentation\n",
    "# !pip install -e ."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.5.1+cu101 True\n",
      "0.12.0\n"
     ]
    }
   ],
   "source": [
    "# Check Pytorch installation\n",
    "import torch, torchvision\n",
    "print(torch.__version__, torch.cuda.is_available())\n",
    "\n",
    "# Check MMSegmentation installation\n",
    "import mmseg\n",
    "print(mmseg.__version__)\n",
    "\n",
    "import os.path as osp\n",
    "import numpy as np\n",
    "from PIL import Image\n",
    "\n",
    "import mmcv\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.patches as mpatches\n",
    "\n",
    "from mmseg.apis import inference_segmentor, init_segmentor, show_result_pyplot\n",
    "from mmseg.core.evaluation import get_palette"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mkdir: cannot create directory 'checkpoints': File exists\n"
     ]
    }
   ],
   "source": [
    "!mkdir checkpoints\n",
    "# !wget https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_40k_voc12aug/pspnet_r101-d8_512x512_40k_voc12aug_20200613_161222-bc933b18.pth -P checkpoints"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "img = mmcv.imread('../data/my_dataset/img_dir/train/007f736aedbc4ca67989f8ca62f1bbeb447ad76698351fe387923963ee50e5ae_frame0000.png')\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.imshow(mmcv.bgr2rgb(img))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert dataset annotation to semantic segmentation map\n",
    "data_root = '../data/my_dataset'\n",
    "img_dir = 'img_dir/train'\n",
    "ann_dir = 'ann_dir/train'\n",
    "# define class and plaette for better visualization\n",
    "classes = ('background', 'cell', 'cilia')\n",
    "palette = [[128, 128, 128], [129, 127, 38], [120, 69, 125]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "img = Image.open('../data/my_dataset/ann_dir/train/007f736aedbc4ca67989f8ca62f1bbeb447ad76698351fe387923963ee50e5ae_frame0000.png')\n",
    "plt.figure(figsize=(8, 6))\n",
    "im = plt.imshow(np.array(img))\n",
    "\n",
    "# create a patch (proxy artist) for every color \n",
    "patches = [mpatches.Patch(color=np.array(palette[i])/255., \n",
    "                          label=classes[i]) for i in range(3)]\n",
    "# put those patched as legend-handles into the legend\n",
    "plt.legend(handles=patches, bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0., \n",
    "           fontsize='large')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# split train/val set randomly\n",
    "split_dir = 'splits'\n",
    "mmcv.mkdir_or_exist(osp.join(data_root, split_dir))\n",
    "filename_list = [osp.splitext(filename)[0] for filename in mmcv.scandir(\n",
    "    osp.join(data_root, ann_dir), suffix='.png')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(osp.join(data_root, split_dir, 'train.txt'), 'w') as f:\n",
    "  # select first 4/5 as train set\n",
    "  train_length = int(len(filename_list)*4/5)\n",
    "  f.writelines(line + '\\n' for line in filename_list[:train_length])\n",
    "with open(osp.join(data_root, split_dir, 'val.txt'), 'w') as f:\n",
    "  # select last 1/5 as train set\n",
    "  f.writelines(line + '\\n' for line in filename_list[train_length:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from mmseg.datasets.builder import DATASETS\n",
    "from mmseg.datasets.custom import CustomDataset\n",
    "\n",
    "@DATASETS.register_module()\n",
    "class CiliaDataset(CustomDataset):\n",
    "  CLASSES = classes\n",
    "  PALETTE = palette\n",
    "  def __init__(self, split, **kwargs):\n",
    "    super().__init__(img_suffix='.png', seg_map_suffix='.png', \n",
    "                     split=split, **kwargs)\n",
    "    assert osp.exists(self.img_dir) and self.split is not None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from mmcv import Config\n",
    "cfg = Config.fromfile('./configs/pspnet/pspnet_r101-d8_512x512_40k_voc12aug.py')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Config:\n",
      "norm_cfg = dict(type='BN', requires_grad=True)\n",
      "model = dict(\n",
      "    type='EncoderDecoder',\n",
      "    pretrained='open-mmlab://resnet101_v1c',\n",
      "    backbone=dict(\n",
      "        type='ResNetV1c',\n",
      "        depth=101,\n",
      "        num_stages=4,\n",
      "        out_indices=(0, 1, 2, 3),\n",
      "        dilations=(1, 1, 2, 4),\n",
      "        strides=(1, 2, 1, 1),\n",
      "        norm_cfg=dict(type='BN', requires_grad=True),\n",
      "        norm_eval=False,\n",
      "        style='pytorch',\n",
      "        contract_dilation=True),\n",
      "    decode_head=dict(\n",
      "        type='PSPHead',\n",
      "        in_channels=2048,\n",
      "        in_index=3,\n",
      "        channels=512,\n",
      "        pool_scales=(1, 2, 3, 6),\n",
      "        dropout_ratio=0.1,\n",
      "        num_classes=3,\n",
      "        norm_cfg=dict(type='BN', requires_grad=True),\n",
      "        align_corners=False,\n",
      "        loss_decode=dict(\n",
      "            type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),\n",
      "    auxiliary_head=dict(\n",
      "        type='FCNHead',\n",
      "        in_channels=1024,\n",
      "        in_index=2,\n",
      "        channels=256,\n",
      "        num_convs=1,\n",
      "        concat_input=False,\n",
      "        dropout_ratio=0.1,\n",
      "        num_classes=3,\n",
      "        norm_cfg=dict(type='BN', requires_grad=True),\n",
      "        align_corners=False,\n",
      "        loss_decode=dict(\n",
      "            type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),\n",
      "    train_cfg=dict(),\n",
      "    test_cfg=dict(mode='whole'))\n",
      "dataset_type = 'CiliaDataset'\n",
      "data_root = '../data/my_dataset'\n",
      "img_norm_cfg = dict(\n",
      "    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)\n",
      "crop_size = (256, 256)\n",
      "train_pipeline = [\n",
      "    dict(type='LoadImageFromFile'),\n",
      "    dict(type='LoadAnnotations'),\n",
      "    dict(type='Resize', img_scale=(320, 240), ratio_range=(0.5, 2.0)),\n",
      "    dict(type='RandomCrop', crop_size=(256, 256), cat_max_ratio=0.75),\n",
      "    dict(type='RandomFlip', flip_ratio=0.5),\n",
      "    dict(type='PhotoMetricDistortion'),\n",
      "    dict(\n",
      "        type='Normalize',\n",
      "        mean=[123.675, 116.28, 103.53],\n",
      "        std=[58.395, 57.12, 57.375],\n",
      "        to_rgb=True),\n",
      "    dict(type='Pad', size=(256, 256), pad_val=0, seg_pad_val=255),\n",
      "    dict(type='DefaultFormatBundle'),\n",
      "    dict(type='Collect', keys=['img', 'gt_semantic_seg'])\n",
      "]\n",
      "test_pipeline = [\n",
      "    dict(type='LoadImageFromFile'),\n",
      "    dict(\n",
      "        type='MultiScaleFlipAug',\n",
      "        img_scale=(320, 240),\n",
      "        flip=False,\n",
      "        transforms=[\n",
      "            dict(type='Resize', keep_ratio=True),\n",
      "            dict(type='RandomFlip'),\n",
      "            dict(\n",
      "                type='Normalize',\n",
      "                mean=[123.675, 116.28, 103.53],\n",
      "                std=[58.395, 57.12, 57.375],\n",
      "                to_rgb=True),\n",
      "            dict(type='ImageToTensor', keys=['img']),\n",
      "            dict(type='Collect', keys=['img'])\n",
      "        ])\n",
      "]\n",
      "data = dict(\n",
      "    samples_per_gpu=16,\n",
      "    workers_per_gpu=16,\n",
      "    train=dict(\n",
      "        type='CiliaDataset',\n",
      "        data_root='../data/my_dataset',\n",
      "        img_dir='img_dir/train',\n",
      "        ann_dir='ann_dir/train',\n",
      "        split='splits/train.txt',\n",
      "        pipeline=[\n",
      "            dict(type='LoadImageFromFile'),\n",
      "            dict(type='LoadAnnotations'),\n",
      "            dict(type='Resize', img_scale=(320, 240), ratio_range=(0.5, 2.0)),\n",
      "            dict(type='RandomCrop', crop_size=(256, 256), cat_max_ratio=0.75),\n",
      "            dict(type='RandomFlip', flip_ratio=0.5),\n",
      "            dict(type='PhotoMetricDistortion'),\n",
      "            dict(\n",
      "                type='Normalize',\n",
      "                mean=[123.675, 116.28, 103.53],\n",
      "                std=[58.395, 57.12, 57.375],\n",
      "                to_rgb=True),\n",
      "            dict(type='Pad', size=(256, 256), pad_val=0, seg_pad_val=255),\n",
      "            dict(type='DefaultFormatBundle'),\n",
      "            dict(type='Collect', keys=['img', 'gt_semantic_seg'])\n",
      "        ]),\n",
      "    val=dict(\n",
      "        type='CiliaDataset',\n",
      "        data_root='../data/my_dataset',\n",
      "        img_dir='img_dir/train',\n",
      "        ann_dir='ann_dir/train',\n",
      "        split='splits/val.txt',\n",
      "        pipeline=[\n",
      "            dict(type='LoadImageFromFile'),\n",
      "            dict(\n",
      "                type='MultiScaleFlipAug',\n",
      "                img_scale=(320, 240),\n",
      "                flip=False,\n",
      "                transforms=[\n",
      "                    dict(type='Resize', keep_ratio=True),\n",
      "                    dict(type='RandomFlip'),\n",
      "                    dict(\n",
      "                        type='Normalize',\n",
      "                        mean=[123.675, 116.28, 103.53],\n",
      "                        std=[58.395, 57.12, 57.375],\n",
      "                        to_rgb=True),\n",
      "                    dict(type='ImageToTensor', keys=['img']),\n",
      "                    dict(type='Collect', keys=['img'])\n",
      "                ])\n",
      "        ]),\n",
      "    test=dict(\n",
      "        type='CiliaDataset',\n",
      "        data_root='../data/my_dataset',\n",
      "        img_dir='img_dir/train',\n",
      "        ann_dir='ann_dir/train',\n",
      "        split='splits/val.txt',\n",
      "        pipeline=[\n",
      "            dict(type='LoadImageFromFile'),\n",
      "            dict(\n",
      "                type='MultiScaleFlipAug',\n",
      "                img_scale=(320, 240),\n",
      "                flip=False,\n",
      "                transforms=[\n",
      "                    dict(type='Resize', keep_ratio=True),\n",
      "                    dict(type='RandomFlip'),\n",
      "                    dict(\n",
      "                        type='Normalize',\n",
      "                        mean=[123.675, 116.28, 103.53],\n",
      "                        std=[58.395, 57.12, 57.375],\n",
      "                        to_rgb=True),\n",
      "                    dict(type='ImageToTensor', keys=['img']),\n",
      "                    dict(type='Collect', keys=['img'])\n",
      "                ])\n",
      "        ]))\n",
      "log_config = dict(\n",
      "    interval=10, hooks=[dict(type='TextLoggerHook', by_epoch=False)])\n",
      "dist_params = dict(backend='nccl')\n",
      "log_level = 'INFO'\n",
      "load_from = './checkpoints/pspnet_r101-d8_512x512_40k_voc12aug_20200613_161222-bc933b18.pth'\n",
      "resume_from = None\n",
      "workflow = [('train', 1)]\n",
      "cudnn_benchmark = True\n",
      "optimizer = dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0005)\n",
      "optimizer_config = dict()\n",
      "lr_config = dict(policy='poly', power=0.9, min_lr=0.0001, by_epoch=False)\n",
      "runner = dict(type='IterBasedRunner', max_iters=5000)\n",
      "checkpoint_config = dict(by_epoch=False, interval=200)\n",
      "evaluation = dict(interval=200, metric='mIoU')\n",
      "work_dir = './work_dirs/tutorial'\n",
      "seed = 0\n",
      "gpu_ids = range(0, 1)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from mmseg.apis import set_random_seed\n",
    "\n",
    "# Since we use ony one GPU, BN is used instead of SyncBN\n",
    "cfg.norm_cfg = dict(type='BN', requires_grad=True)\n",
    "cfg.model.backbone.norm_cfg = cfg.norm_cfg\n",
    "cfg.model.decode_head.norm_cfg = cfg.norm_cfg\n",
    "cfg.model.auxiliary_head.norm_cfg = cfg.norm_cfg\n",
    "# modify num classes of the model in decode/auxiliary head\n",
    "cfg.model.decode_head.num_classes = 3\n",
    "cfg.model.auxiliary_head.num_classes = 3\n",
    "\n",
    "# Modify dataset type and path\n",
    "cfg.dataset_type = 'CiliaDataset'\n",
    "cfg.data_root = data_root\n",
    "\n",
    "cfg.data.samples_per_gpu = 16\n",
    "cfg.data.workers_per_gpu=16\n",
    "\n",
    "cfg.img_norm_cfg = dict(\n",
    "    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)\n",
    "cfg.crop_size = (256, 256)\n",
    "cfg.train_pipeline = [\n",
    "    dict(type='LoadImageFromFile'),\n",
    "    dict(type='LoadAnnotations'),\n",
    "    dict(type='Resize', img_scale=(320, 240), ratio_range=(0.5, 2.0)),\n",
    "    dict(type='RandomCrop', crop_size=cfg.crop_size, cat_max_ratio=0.75),\n",
    "    dict(type='RandomFlip', flip_ratio=0.5),\n",
    "    dict(type='PhotoMetricDistortion'),\n",
    "    dict(type='Normalize', **cfg.img_norm_cfg),\n",
    "    dict(type='Pad', size=cfg.crop_size, pad_val=0, seg_pad_val=255),\n",
    "    dict(type='DefaultFormatBundle'),\n",
    "    dict(type='Collect', keys=['img', 'gt_semantic_seg']),\n",
    "]\n",
    "\n",
    "cfg.test_pipeline = [\n",
    "    dict(type='LoadImageFromFile'),\n",
    "    dict(\n",
    "        type='MultiScaleFlipAug',\n",
    "        img_scale=(320, 240),\n",
    "        # img_ratios=[0.5, 0.75, 1.0, 1.25, 1.5, 1.75],\n",
    "        flip=False,\n",
    "        transforms=[\n",
    "            dict(type='Resize', keep_ratio=True),\n",
    "            dict(type='RandomFlip'),\n",
    "            dict(type='Normalize', **cfg.img_norm_cfg),\n",
    "            dict(type='ImageToTensor', keys=['img']),\n",
    "            dict(type='Collect', keys=['img']),\n",
    "        ])\n",
    "]\n",
    "\n",
    "\n",
    "cfg.data.train.type = cfg.dataset_type\n",
    "cfg.data.train.data_root = cfg.data_root\n",
    "cfg.data.train.img_dir = img_dir\n",
    "cfg.data.train.ann_dir = ann_dir\n",
    "cfg.data.train.pipeline = cfg.train_pipeline\n",
    "cfg.data.train.split = 'splits/train.txt'\n",
    "\n",
    "cfg.data.val.type = cfg.dataset_type\n",
    "cfg.data.val.data_root = cfg.data_root\n",
    "cfg.data.val.img_dir = img_dir\n",
    "cfg.data.val.ann_dir = ann_dir\n",
    "cfg.data.val.pipeline = cfg.test_pipeline\n",
    "cfg.data.val.split = 'splits/val.txt'\n",
    "\n",
    "cfg.data.test.type = cfg.dataset_type\n",
    "cfg.data.test.data_root = cfg.data_root\n",
    "cfg.data.test.img_dir = img_dir\n",
    "cfg.data.test.ann_dir = ann_dir\n",
    "cfg.data.test.pipeline = cfg.test_pipeline\n",
    "cfg.data.test.split = 'splits/val.txt'\n",
    "\n",
    "# We can still use the pre-trained Mask RCNN model though we do not need to\n",
    "# use the mask branch\n",
    "cfg.load_from = './checkpoints/pspnet_r101-d8_512x512_40k_voc12aug_20200613_161222-bc933b18.pth'\n",
    "\n",
    "# Set up working dir to save files and logs.\n",
    "cfg.work_dir = './work_dirs/tutorial'\n",
    "\n",
    "cfg.runner.max_iters = 5000\n",
    "cfg.log_config.interval = 10\n",
    "cfg.evaluation.interval = 200\n",
    "cfg.checkpoint_config.interval = 200\n",
    "\n",
    "# Set seed to facitate reproducing the result\n",
    "cfg.seed = 0\n",
    "set_random_seed(0, deterministic=False)\n",
    "cfg.gpu_ids = range(1)\n",
    "\n",
    "# Let's have a look at the final config used for training\n",
    "print(f'Config:\\n{cfg.pretty_text}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/esbg-lab/anaconda3/envs/GT/lib/python3.7/site-packages/mmcv/utils/misc.py:304: UserWarning: \"flip_ratio\" is deprecated in `RandomFlip.__init__`, please use \"prob\" instead\n",
      "  f'\"{src_arg_name}\" is deprecated in '\n",
      "2021-04-20 09:49:34,554 - mmseg - INFO - Loaded 16880 images\n",
      "2021-04-20 09:49:35,163 - mmseg - WARNING - The model and loaded state dict do not match exactly\n",
      "\n",
      "unexpected key in source state_dict: fc.weight, fc.bias\n",
      "\n",
      "2021-04-20 09:49:37,132 - mmseg - INFO - Loaded 4220 images\n",
      "2021-04-20 09:49:37,133 - mmseg - INFO - load checkpoint from ./checkpoints/pspnet_r101-d8_512x512_40k_voc12aug_20200613_161222-bc933b18.pth\n",
      "2021-04-20 09:49:37,343 - mmseg - WARNING - The model and loaded state dict do not match exactly\n",
      "\n",
      "size mismatch for decode_head.conv_seg.weight: copying a param with shape torch.Size([21, 512, 1, 1]) from checkpoint, the shape in current model is torch.Size([3, 512, 1, 1]).\n",
      "size mismatch for decode_head.conv_seg.bias: copying a param with shape torch.Size([21]) from checkpoint, the shape in current model is torch.Size([3]).\n",
      "size mismatch for auxiliary_head.conv_seg.weight: copying a param with shape torch.Size([21, 256, 1, 1]) from checkpoint, the shape in current model is torch.Size([3, 256, 1, 1]).\n",
      "size mismatch for auxiliary_head.conv_seg.bias: copying a param with shape torch.Size([21]) from checkpoint, the shape in current model is torch.Size([3]).\n",
      "2021-04-20 09:49:37,357 - mmseg - INFO - Start running, host: esbg-lab@lambda-dual, work_dir: /home/esbg-lab/Desktop/P3/mmsegmentation/work_dirs/tutorial\n",
      "2021-04-20 09:49:37,358 - mmseg - INFO - workflow: [('train', 1)], max: 5000 iters\n",
      "2021-04-20 09:49:44,458 - mmseg - INFO - Iter [10/5000]\tlr: 9.984e-03, eta: 0:53:41, time: 0.646, data_time: 0.039, memory: 9591, decode.loss_seg: 0.8243, decode.acc_seg: 50.1814, aux.loss_seg: 0.3569, aux.acc_seg: 44.2363, loss: 1.1812\n",
      "2021-04-20 09:49:50,543 - mmseg - INFO - Iter [20/5000]\tlr: 9.966e-03, eta: 0:52:02, time: 0.608, data_time: 0.004, memory: 9591, decode.loss_seg: 0.5398, decode.acc_seg: 62.0643, aux.loss_seg: 0.2602, aux.acc_seg: 61.6119, loss: 0.8000\n",
      "2021-04-20 09:49:56,632 - mmseg - INFO - Iter [30/5000]\tlr: 9.948e-03, eta: 0:51:26, time: 0.609, data_time: 0.004, memory: 9591, decode.loss_seg: 0.4413, decode.acc_seg: 66.2990, aux.loss_seg: 0.2090, aux.acc_seg: 66.0047, loss: 0.6504\n",
      "2021-04-20 09:50:02,745 - mmseg - INFO - Iter [40/5000]\tlr: 9.930e-03, eta: 0:51:08, time: 0.611, data_time: 0.004, memory: 9591, decode.loss_seg: 0.3982, decode.acc_seg: 68.0896, aux.loss_seg: 0.1875, aux.acc_seg: 66.2173, loss: 0.5857\n",
      "2021-04-20 09:50:08,879 - mmseg - INFO - Iter [50/5000]\tlr: 9.913e-03, eta: 0:50:56, time: 0.613, data_time: 0.004, memory: 9591, decode.loss_seg: 0.3339, decode.acc_seg: 73.8084, aux.loss_seg: 0.1552, aux.acc_seg: 70.8189, loss: 0.4891\n",
      "2021-04-20 09:50:15,044 - mmseg - INFO - Iter [60/5000]\tlr: 9.895e-03, eta: 0:50:49, time: 0.617, data_time: 0.004, memory: 9591, decode.loss_seg: 0.3518, decode.acc_seg: 71.3033, aux.loss_seg: 0.1599, aux.acc_seg: 68.7850, loss: 0.5117\n",
      "2021-04-20 09:50:21,240 - mmseg - INFO - Iter [70/5000]\tlr: 9.877e-03, eta: 0:50:45, time: 0.620, data_time: 0.004, memory: 9591, decode.loss_seg: 0.3341, decode.acc_seg: 70.4792, aux.loss_seg: 0.1504, aux.acc_seg: 69.6656, loss: 0.4846\n",
      "2021-04-20 09:50:27,433 - mmseg - INFO - Iter [80/5000]\tlr: 9.859e-03, eta: 0:50:40, time: 0.619, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2787, decode.acc_seg: 74.2683, aux.loss_seg: 0.1231, aux.acc_seg: 74.0251, loss: 0.4019\n",
      "2021-04-20 09:50:33,646 - mmseg - INFO - Iter [90/5000]\tlr: 9.841e-03, eta: 0:50:35, time: 0.621, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2717, decode.acc_seg: 73.0005, aux.loss_seg: 0.1227, aux.acc_seg: 72.1481, loss: 0.3944\n",
      "2021-04-20 09:50:39,868 - mmseg - INFO - Iter [100/5000]\tlr: 9.823e-03, eta: 0:50:31, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2911, decode.acc_seg: 75.0019, aux.loss_seg: 0.1252, aux.acc_seg: 74.7282, loss: 0.4163\n",
      "2021-04-20 09:50:46,105 - mmseg - INFO - Iter [110/5000]\tlr: 9.806e-03, eta: 0:50:27, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2783, decode.acc_seg: 73.6282, aux.loss_seg: 0.1170, aux.acc_seg: 73.4654, loss: 0.3953\n",
      "2021-04-20 09:50:52,339 - mmseg - INFO - Iter [120/5000]\tlr: 9.788e-03, eta: 0:50:23, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.3028, decode.acc_seg: 71.6545, aux.loss_seg: 0.1270, aux.acc_seg: 71.4658, loss: 0.4298\n",
      "2021-04-20 09:50:58,585 - mmseg - INFO - Iter [130/5000]\tlr: 9.770e-03, eta: 0:50:18, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2598, decode.acc_seg: 73.0948, aux.loss_seg: 0.1111, aux.acc_seg: 72.5411, loss: 0.3708\n",
      "2021-04-20 09:51:04,844 - mmseg - INFO - Iter [140/5000]\tlr: 9.752e-03, eta: 0:50:14, time: 0.626, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2677, decode.acc_seg: 74.9339, aux.loss_seg: 0.1123, aux.acc_seg: 74.6000, loss: 0.3800\n",
      "2021-04-20 09:51:11,110 - mmseg - INFO - Iter [150/5000]\tlr: 9.734e-03, eta: 0:50:10, time: 0.627, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2682, decode.acc_seg: 76.1114, aux.loss_seg: 0.1169, aux.acc_seg: 75.4475, loss: 0.3852\n",
      "2021-04-20 09:51:17,379 - mmseg - INFO - Iter [160/5000]\tlr: 9.716e-03, eta: 0:50:06, time: 0.627, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2496, decode.acc_seg: 74.1640, aux.loss_seg: 0.1064, aux.acc_seg: 73.7233, loss: 0.3560\n",
      "2021-04-20 09:51:23,648 - mmseg - INFO - Iter [170/5000]\tlr: 9.698e-03, eta: 0:50:01, time: 0.627, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2590, decode.acc_seg: 76.0444, aux.loss_seg: 0.1114, aux.acc_seg: 75.6858, loss: 0.3704\n",
      "2021-04-20 09:51:29,932 - mmseg - INFO - Iter [180/5000]\tlr: 9.680e-03, eta: 0:49:57, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2172, decode.acc_seg: 77.4452, aux.loss_seg: 0.0935, aux.acc_seg: 77.1119, loss: 0.3107\n",
      "2021-04-20 09:51:36,216 - mmseg - INFO - Iter [190/5000]\tlr: 9.663e-03, eta: 0:49:52, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2395, decode.acc_seg: 75.9970, aux.loss_seg: 0.1065, aux.acc_seg: 74.9601, loss: 0.3459\n",
      "2021-04-20 09:51:42,494 - mmseg - INFO - Saving checkpoint at 200 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 37.0 task/s, elapsed: 114s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 09:53:54,983 - mmseg - INFO - per class results:\n",
      "2021-04-20 09:53:54,984 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 90.66 | 94.96 |\n",
      "| cell       | 83.35 | 92.23 |\n",
      "| cilia      | 46.17 | 59.94 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 09:53:54,984 - mmseg - INFO - Summary:\n",
      "2021-04-20 09:53:54,985 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 73.39 | 82.38 | 91.36 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 09:53:54,992 - mmseg - INFO - Iter(val) [200]\tmIoU: 0.7339, mAcc: 0.8238, aAcc: 0.9136, IoU.background: 0.9066, IoU.cell: 0.8335, IoU.cilia: 0.4617, Acc.background: 0.9496, Acc.cell: 0.9223, Acc.cilia: 0.5994\n",
      "2021-04-20 09:54:01,224 - mmseg - INFO - Iter [210/5000]\tlr: 9.627e-03, eta: 1:37:24, time: 13.801, data_time: 13.182, memory: 9591, decode.loss_seg: 0.2177, decode.acc_seg: 78.0530, aux.loss_seg: 0.0946, aux.acc_seg: 77.5114, loss: 0.3123\n",
      "2021-04-20 09:54:07,461 - mmseg - INFO - Iter [220/5000]\tlr: 9.609e-03, eta: 1:35:02, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2285, decode.acc_seg: 75.9228, aux.loss_seg: 0.0979, aux.acc_seg: 75.4000, loss: 0.3265\n",
      "2021-04-20 09:54:13,700 - mmseg - INFO - Iter [230/5000]\tlr: 9.591e-03, eta: 1:32:52, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2212, decode.acc_seg: 74.4847, aux.loss_seg: 0.0950, aux.acc_seg: 74.1652, loss: 0.3162\n",
      "2021-04-20 09:54:19,956 - mmseg - INFO - Iter [240/5000]\tlr: 9.573e-03, eta: 1:30:53, time: 0.626, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2095, decode.acc_seg: 77.2961, aux.loss_seg: 0.0913, aux.acc_seg: 76.6450, loss: 0.3009\n",
      "2021-04-20 09:54:26,223 - mmseg - INFO - Iter [250/5000]\tlr: 9.555e-03, eta: 1:29:03, time: 0.627, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1981, decode.acc_seg: 76.3169, aux.loss_seg: 0.0885, aux.acc_seg: 75.5452, loss: 0.2866\n",
      "2021-04-20 09:54:32,501 - mmseg - INFO - Iter [260/5000]\tlr: 9.537e-03, eta: 1:27:21, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1972, decode.acc_seg: 77.8264, aux.loss_seg: 0.0860, aux.acc_seg: 77.4145, loss: 0.2832\n",
      "2021-04-20 09:54:38,787 - mmseg - INFO - Iter [270/5000]\tlr: 9.519e-03, eta: 1:25:46, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1998, decode.acc_seg: 74.0863, aux.loss_seg: 0.0894, aux.acc_seg: 73.5279, loss: 0.2892\n",
      "2021-04-20 09:54:45,076 - mmseg - INFO - Iter [280/5000]\tlr: 9.501e-03, eta: 1:24:18, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2185, decode.acc_seg: 75.7197, aux.loss_seg: 0.0934, aux.acc_seg: 75.1219, loss: 0.3119\n",
      "2021-04-20 09:54:51,374 - mmseg - INFO - Iter [290/5000]\tlr: 9.483e-03, eta: 1:22:55, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1958, decode.acc_seg: 76.7380, aux.loss_seg: 0.0852, aux.acc_seg: 76.1747, loss: 0.2810\n",
      "2021-04-20 09:54:57,673 - mmseg - INFO - Iter [300/5000]\tlr: 9.466e-03, eta: 1:21:38, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1767, decode.acc_seg: 80.0791, aux.loss_seg: 0.0778, aux.acc_seg: 79.6315, loss: 0.2545\n",
      "2021-04-20 09:55:03,978 - mmseg - INFO - Iter [310/5000]\tlr: 9.448e-03, eta: 1:20:25, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2025, decode.acc_seg: 82.2534, aux.loss_seg: 0.0861, aux.acc_seg: 81.9602, loss: 0.2886\n",
      "2021-04-20 09:55:10,291 - mmseg - INFO - Iter [320/5000]\tlr: 9.430e-03, eta: 1:19:17, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2041, decode.acc_seg: 82.2501, aux.loss_seg: 0.0867, aux.acc_seg: 81.9633, loss: 0.2907\n",
      "2021-04-20 09:55:16,602 - mmseg - INFO - Iter [330/5000]\tlr: 9.412e-03, eta: 1:18:12, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1874, decode.acc_seg: 77.2185, aux.loss_seg: 0.0795, aux.acc_seg: 76.9491, loss: 0.2669\n",
      "2021-04-20 09:55:22,911 - mmseg - INFO - Iter [340/5000]\tlr: 9.394e-03, eta: 1:17:11, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1736, decode.acc_seg: 76.7826, aux.loss_seg: 0.0755, aux.acc_seg: 76.3610, loss: 0.2492\n",
      "2021-04-20 09:55:29,220 - mmseg - INFO - Iter [350/5000]\tlr: 9.376e-03, eta: 1:16:13, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1757, decode.acc_seg: 78.4792, aux.loss_seg: 0.0758, aux.acc_seg: 78.1470, loss: 0.2515\n",
      "2021-04-20 09:55:35,532 - mmseg - INFO - Iter [360/5000]\tlr: 9.358e-03, eta: 1:15:17, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1715, decode.acc_seg: 78.8765, aux.loss_seg: 0.0741, aux.acc_seg: 78.4597, loss: 0.2457\n",
      "2021-04-20 09:55:41,848 - mmseg - INFO - Iter [370/5000]\tlr: 9.340e-03, eta: 1:14:25, time: 0.632, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1812, decode.acc_seg: 77.5964, aux.loss_seg: 0.0777, aux.acc_seg: 77.2072, loss: 0.2589\n",
      "2021-04-20 09:55:48,157 - mmseg - INFO - Iter [380/5000]\tlr: 9.322e-03, eta: 1:13:35, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1684, decode.acc_seg: 78.5107, aux.loss_seg: 0.0724, aux.acc_seg: 78.1343, loss: 0.2408\n",
      "2021-04-20 09:55:54,467 - mmseg - INFO - Iter [390/5000]\tlr: 9.304e-03, eta: 1:12:47, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1685, decode.acc_seg: 76.0927, aux.loss_seg: 0.0726, aux.acc_seg: 75.7394, loss: 0.2411\n",
      "2021-04-20 09:56:00,767 - mmseg - INFO - Saving checkpoint at 400 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.6 task/s, elapsed: 115s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 09:58:14,935 - mmseg - INFO - per class results:\n",
      "2021-04-20 09:58:14,936 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 93.71 | 96.87 |\n",
      "| cell       | 87.78 | 95.43 |\n",
      "| cilia      | 53.84 | 63.08 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 09:58:14,937 - mmseg - INFO - Summary:\n",
      "2021-04-20 09:58:14,937 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 78.44 | 85.13 | 93.79 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 09:58:14,944 - mmseg - INFO - Iter(val) [400]\tmIoU: 0.7844, mAcc: 0.8513, aAcc: 0.9379, IoU.background: 0.9371, IoU.cell: 0.8778, IoU.cilia: 0.5384, Acc.background: 0.9687, Acc.cell: 0.9543, Acc.cilia: 0.6308\n",
      "2021-04-20 09:58:21,176 - mmseg - INFO - Iter [410/5000]\tlr: 9.268e-03, eta: 1:35:01, time: 13.983, data_time: 13.364, memory: 9591, decode.loss_seg: 0.1717, decode.acc_seg: 78.3204, aux.loss_seg: 0.0723, aux.acc_seg: 78.1532, loss: 0.2440\n",
      "2021-04-20 09:58:27,421 - mmseg - INFO - Iter [420/5000]\tlr: 9.250e-03, eta: 1:33:41, time: 0.625, data_time: 0.005, memory: 9591, decode.loss_seg: 0.1686, decode.acc_seg: 76.7104, aux.loss_seg: 0.0729, aux.acc_seg: 76.2956, loss: 0.2414\n",
      "2021-04-20 09:58:33,658 - mmseg - INFO - Iter [430/5000]\tlr: 9.232e-03, eta: 1:32:25, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1599, decode.acc_seg: 82.4630, aux.loss_seg: 0.0679, aux.acc_seg: 82.1441, loss: 0.2278\n",
      "2021-04-20 09:58:39,906 - mmseg - INFO - Iter [440/5000]\tlr: 9.214e-03, eta: 1:31:12, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1632, decode.acc_seg: 76.2482, aux.loss_seg: 0.0720, aux.acc_seg: 75.7042, loss: 0.2352\n",
      "2021-04-20 09:58:46,176 - mmseg - INFO - Iter [450/5000]\tlr: 9.196e-03, eta: 1:30:02, time: 0.627, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1795, decode.acc_seg: 77.0826, aux.loss_seg: 0.0782, aux.acc_seg: 76.3889, loss: 0.2577\n",
      "2021-04-20 09:58:52,460 - mmseg - INFO - Iter [460/5000]\tlr: 9.178e-03, eta: 1:28:55, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1498, decode.acc_seg: 77.3106, aux.loss_seg: 0.0648, aux.acc_seg: 76.9517, loss: 0.2146\n",
      "2021-04-20 09:58:58,746 - mmseg - INFO - Iter [470/5000]\tlr: 9.160e-03, eta: 1:27:50, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.2008, decode.acc_seg: 80.8104, aux.loss_seg: 0.0856, aux.acc_seg: 80.5111, loss: 0.2864\n",
      "2021-04-20 09:59:05,034 - mmseg - INFO - Iter [480/5000]\tlr: 9.142e-03, eta: 1:26:48, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1763, decode.acc_seg: 75.9619, aux.loss_seg: 0.0763, aux.acc_seg: 75.6198, loss: 0.2526\n",
      "2021-04-20 09:59:11,329 - mmseg - INFO - Iter [490/5000]\tlr: 9.124e-03, eta: 1:25:49, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1761, decode.acc_seg: 76.6358, aux.loss_seg: 0.0775, aux.acc_seg: 76.0459, loss: 0.2536\n",
      "2021-04-20 09:59:17,624 - mmseg - INFO - Iter [500/5000]\tlr: 9.106e-03, eta: 1:24:51, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1657, decode.acc_seg: 77.1732, aux.loss_seg: 0.0725, aux.acc_seg: 76.6608, loss: 0.2382\n",
      "2021-04-20 09:59:23,927 - mmseg - INFO - Iter [510/5000]\tlr: 9.088e-03, eta: 1:23:56, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1604, decode.acc_seg: 76.3341, aux.loss_seg: 0.0707, aux.acc_seg: 75.7421, loss: 0.2311\n",
      "2021-04-20 09:59:30,228 - mmseg - INFO - Iter [520/5000]\tlr: 9.070e-03, eta: 1:23:02, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1673, decode.acc_seg: 80.8060, aux.loss_seg: 0.0728, aux.acc_seg: 80.3153, loss: 0.2402\n",
      "2021-04-20 09:59:36,530 - mmseg - INFO - Iter [530/5000]\tlr: 9.052e-03, eta: 1:22:10, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1498, decode.acc_seg: 78.8582, aux.loss_seg: 0.0664, aux.acc_seg: 78.3651, loss: 0.2162\n",
      "2021-04-20 09:59:42,829 - mmseg - INFO - Iter [540/5000]\tlr: 9.034e-03, eta: 1:21:20, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1434, decode.acc_seg: 80.7037, aux.loss_seg: 0.0620, aux.acc_seg: 80.3591, loss: 0.2054\n",
      "2021-04-20 09:59:49,133 - mmseg - INFO - Iter [550/5000]\tlr: 9.016e-03, eta: 1:20:32, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1641, decode.acc_seg: 77.7218, aux.loss_seg: 0.0701, aux.acc_seg: 77.4151, loss: 0.2342\n",
      "2021-04-20 09:59:55,433 - mmseg - INFO - Iter [560/5000]\tlr: 8.998e-03, eta: 1:19:45, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1628, decode.acc_seg: 77.5574, aux.loss_seg: 0.0743, aux.acc_seg: 76.8095, loss: 0.2372\n",
      "2021-04-20 10:00:01,734 - mmseg - INFO - Iter [570/5000]\tlr: 8.980e-03, eta: 1:18:59, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1530, decode.acc_seg: 82.0741, aux.loss_seg: 0.0684, aux.acc_seg: 81.5019, loss: 0.2214\n",
      "2021-04-20 10:00:08,034 - mmseg - INFO - Iter [580/5000]\tlr: 8.962e-03, eta: 1:18:15, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1496, decode.acc_seg: 78.9304, aux.loss_seg: 0.0644, aux.acc_seg: 78.5604, loss: 0.2140\n",
      "2021-04-20 10:00:14,332 - mmseg - INFO - Iter [590/5000]\tlr: 8.944e-03, eta: 1:17:32, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1520, decode.acc_seg: 78.6763, aux.loss_seg: 0.0664, aux.acc_seg: 78.3050, loss: 0.2183\n",
      "2021-04-20 10:00:20,635 - mmseg - INFO - Saving checkpoint at 600 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.4 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:02:34,797 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:02:34,798 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 94.52 | 97.16 |\n",
      "| cell       | 89.78 | 95.18 |\n",
      "| cilia      | 60.88 | 73.89 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:02:34,798 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:02:34,799 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 81.73 | 88.74 | 94.71 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 10:02:34,806 - mmseg - INFO - Iter(val) [600]\tmIoU: 0.8173, mAcc: 0.8874, aAcc: 0.9471, IoU.background: 0.9452, IoU.cell: 0.8978, IoU.cilia: 0.6088, Acc.background: 0.9716, Acc.cell: 0.9518, Acc.cilia: 0.7389\n",
      "2021-04-20 10:02:41,050 - mmseg - INFO - Iter [610/5000]\tlr: 8.908e-03, eta: 1:31:25, time: 13.981, data_time: 13.361, memory: 9591, decode.loss_seg: 0.1592, decode.acc_seg: 80.5780, aux.loss_seg: 0.0695, aux.acc_seg: 80.1970, loss: 0.2287\n",
      "2021-04-20 10:02:47,275 - mmseg - INFO - Iter [620/5000]\tlr: 8.890e-03, eta: 1:30:29, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1254, decode.acc_seg: 77.8299, aux.loss_seg: 0.0552, aux.acc_seg: 77.4731, loss: 0.1806\n",
      "2021-04-20 10:02:53,506 - mmseg - INFO - Iter [630/5000]\tlr: 8.872e-03, eta: 1:29:33, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1381, decode.acc_seg: 76.6383, aux.loss_seg: 0.0581, aux.acc_seg: 76.5269, loss: 0.1963\n",
      "2021-04-20 10:02:59,750 - mmseg - INFO - Iter [640/5000]\tlr: 8.854e-03, eta: 1:28:40, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1339, decode.acc_seg: 81.5055, aux.loss_seg: 0.0589, aux.acc_seg: 81.1015, loss: 0.1927\n",
      "2021-04-20 10:03:06,032 - mmseg - INFO - Iter [650/5000]\tlr: 8.836e-03, eta: 1:27:48, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1288, decode.acc_seg: 78.2279, aux.loss_seg: 0.0560, aux.acc_seg: 77.8399, loss: 0.1849\n",
      "2021-04-20 10:03:12,323 - mmseg - INFO - Iter [660/5000]\tlr: 8.818e-03, eta: 1:26:58, time: 0.629, data_time: 0.005, memory: 9591, decode.loss_seg: 0.1415, decode.acc_seg: 80.2574, aux.loss_seg: 0.0620, aux.acc_seg: 79.9428, loss: 0.2035\n",
      "2021-04-20 10:03:18,611 - mmseg - INFO - Iter [670/5000]\tlr: 8.799e-03, eta: 1:26:09, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1367, decode.acc_seg: 79.7311, aux.loss_seg: 0.0589, aux.acc_seg: 79.4608, loss: 0.1956\n",
      "2021-04-20 10:03:24,911 - mmseg - INFO - Iter [680/5000]\tlr: 8.781e-03, eta: 1:25:21, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1415, decode.acc_seg: 79.6097, aux.loss_seg: 0.0613, aux.acc_seg: 79.3366, loss: 0.2028\n",
      "2021-04-20 10:03:31,212 - mmseg - INFO - Iter [690/5000]\tlr: 8.763e-03, eta: 1:24:34, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1326, decode.acc_seg: 78.6896, aux.loss_seg: 0.0587, aux.acc_seg: 78.2438, loss: 0.1913\n",
      "2021-04-20 10:03:37,517 - mmseg - INFO - Iter [700/5000]\tlr: 8.745e-03, eta: 1:23:49, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1454, decode.acc_seg: 79.4000, aux.loss_seg: 0.0652, aux.acc_seg: 78.8970, loss: 0.2106\n",
      "2021-04-20 10:03:43,829 - mmseg - INFO - Iter [710/5000]\tlr: 8.727e-03, eta: 1:23:05, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1363, decode.acc_seg: 79.1332, aux.loss_seg: 0.0581, aux.acc_seg: 78.9410, loss: 0.1944\n",
      "2021-04-20 10:03:50,138 - mmseg - INFO - Iter [720/5000]\tlr: 8.709e-03, eta: 1:22:21, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1195, decode.acc_seg: 80.9945, aux.loss_seg: 0.0527, aux.acc_seg: 80.6140, loss: 0.1721\n",
      "2021-04-20 10:03:56,447 - mmseg - INFO - Iter [730/5000]\tlr: 8.691e-03, eta: 1:21:39, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1330, decode.acc_seg: 79.6939, aux.loss_seg: 0.0587, aux.acc_seg: 79.2907, loss: 0.1917\n",
      "2021-04-20 10:04:02,760 - mmseg - INFO - Iter [740/5000]\tlr: 8.673e-03, eta: 1:20:58, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1320, decode.acc_seg: 79.6174, aux.loss_seg: 0.0562, aux.acc_seg: 79.4724, loss: 0.1882\n",
      "2021-04-20 10:04:09,070 - mmseg - INFO - Iter [750/5000]\tlr: 8.655e-03, eta: 1:20:18, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1475, decode.acc_seg: 78.8732, aux.loss_seg: 0.0656, aux.acc_seg: 78.4279, loss: 0.2131\n",
      "2021-04-20 10:04:15,381 - mmseg - INFO - Iter [760/5000]\tlr: 8.637e-03, eta: 1:19:38, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1250, decode.acc_seg: 80.5773, aux.loss_seg: 0.0577, aux.acc_seg: 79.9035, loss: 0.1827\n",
      "2021-04-20 10:04:21,689 - mmseg - INFO - Iter [770/5000]\tlr: 8.618e-03, eta: 1:19:00, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1369, decode.acc_seg: 79.4666, aux.loss_seg: 0.0599, aux.acc_seg: 79.0733, loss: 0.1968\n",
      "2021-04-20 10:04:27,998 - mmseg - INFO - Iter [780/5000]\tlr: 8.600e-03, eta: 1:18:22, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1335, decode.acc_seg: 78.0248, aux.loss_seg: 0.0572, aux.acc_seg: 77.7988, loss: 0.1907\n",
      "2021-04-20 10:04:34,308 - mmseg - INFO - Iter [790/5000]\tlr: 8.582e-03, eta: 1:17:45, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1392, decode.acc_seg: 78.2703, aux.loss_seg: 0.0608, aux.acc_seg: 77.9071, loss: 0.2000\n",
      "2021-04-20 10:04:40,612 - mmseg - INFO - Saving checkpoint at 800 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.1 task/s, elapsed: 117s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:06:55,682 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:06:55,682 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 95.0  | 97.61 |\n",
      "| cell       | 90.84 | 95.29 |\n",
      "| cilia      | 64.23 | 76.85 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:06:55,683 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:06:55,683 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 83.36 | 89.92 | 95.24 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 10:06:55,690 - mmseg - INFO - Iter(val) [800]\tmIoU: 0.8336, mAcc: 0.8992, aAcc: 0.9524, IoU.background: 0.9500, IoU.cell: 0.9084, IoU.cilia: 0.6423, Acc.background: 0.9761, Acc.cell: 0.9529, Acc.cilia: 0.7685\n",
      "2021-04-20 10:07:01,944 - mmseg - INFO - Iter [810/5000]\tlr: 8.546e-03, eta: 1:27:37, time: 14.077, data_time: 13.456, memory: 9591, decode.loss_seg: 0.1271, decode.acc_seg: 81.7976, aux.loss_seg: 0.0549, aux.acc_seg: 81.4463, loss: 0.1820\n",
      "2021-04-20 10:07:08,167 - mmseg - INFO - Iter [820/5000]\tlr: 8.528e-03, eta: 1:26:52, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1306, decode.acc_seg: 80.0920, aux.loss_seg: 0.0573, aux.acc_seg: 79.7140, loss: 0.1879\n",
      "2021-04-20 10:07:14,400 - mmseg - INFO - Iter [830/5000]\tlr: 8.510e-03, eta: 1:26:08, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1145, decode.acc_seg: 77.8946, aux.loss_seg: 0.0493, aux.acc_seg: 77.6488, loss: 0.1637\n",
      "2021-04-20 10:07:20,657 - mmseg - INFO - Iter [840/5000]\tlr: 8.492e-03, eta: 1:25:25, time: 0.626, data_time: 0.005, memory: 9591, decode.loss_seg: 0.1195, decode.acc_seg: 78.8517, aux.loss_seg: 0.0527, aux.acc_seg: 78.4858, loss: 0.1721\n",
      "2021-04-20 10:07:26,932 - mmseg - INFO - Iter [850/5000]\tlr: 8.473e-03, eta: 1:24:43, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1170, decode.acc_seg: 77.7436, aux.loss_seg: 0.0511, aux.acc_seg: 77.4977, loss: 0.1681\n",
      "2021-04-20 10:07:33,221 - mmseg - INFO - Iter [860/5000]\tlr: 8.455e-03, eta: 1:24:02, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1166, decode.acc_seg: 77.9687, aux.loss_seg: 0.0523, aux.acc_seg: 77.4668, loss: 0.1689\n",
      "2021-04-20 10:07:39,513 - mmseg - INFO - Iter [870/5000]\tlr: 8.437e-03, eta: 1:23:22, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1230, decode.acc_seg: 78.7581, aux.loss_seg: 0.0537, aux.acc_seg: 78.4259, loss: 0.1768\n",
      "2021-04-20 10:07:45,813 - mmseg - INFO - Iter [880/5000]\tlr: 8.419e-03, eta: 1:22:43, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1154, decode.acc_seg: 78.7955, aux.loss_seg: 0.0501, aux.acc_seg: 78.5011, loss: 0.1654\n",
      "2021-04-20 10:07:52,117 - mmseg - INFO - Iter [890/5000]\tlr: 8.401e-03, eta: 1:22:04, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1281, decode.acc_seg: 81.9670, aux.loss_seg: 0.0583, aux.acc_seg: 81.4194, loss: 0.1863\n",
      "2021-04-20 10:07:58,419 - mmseg - INFO - Iter [900/5000]\tlr: 8.383e-03, eta: 1:21:27, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1243, decode.acc_seg: 78.8023, aux.loss_seg: 0.0536, aux.acc_seg: 78.5076, loss: 0.1780\n",
      "2021-04-20 10:08:04,735 - mmseg - INFO - Iter [910/5000]\tlr: 8.364e-03, eta: 1:20:49, time: 0.632, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1161, decode.acc_seg: 81.2150, aux.loss_seg: 0.0507, aux.acc_seg: 80.9181, loss: 0.1667\n",
      "2021-04-20 10:08:11,043 - mmseg - INFO - Iter [920/5000]\tlr: 8.346e-03, eta: 1:20:13, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1169, decode.acc_seg: 83.8316, aux.loss_seg: 0.0506, aux.acc_seg: 83.5929, loss: 0.1675\n",
      "2021-04-20 10:08:17,352 - mmseg - INFO - Iter [930/5000]\tlr: 8.328e-03, eta: 1:19:37, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1034, decode.acc_seg: 76.7834, aux.loss_seg: 0.0456, aux.acc_seg: 76.4510, loss: 0.1490\n",
      "2021-04-20 10:08:23,663 - mmseg - INFO - Iter [940/5000]\tlr: 8.310e-03, eta: 1:19:02, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1172, decode.acc_seg: 78.2203, aux.loss_seg: 0.0509, aux.acc_seg: 77.8687, loss: 0.1681\n",
      "2021-04-20 10:08:29,971 - mmseg - INFO - Iter [950/5000]\tlr: 8.292e-03, eta: 1:18:27, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1242, decode.acc_seg: 78.9957, aux.loss_seg: 0.0542, aux.acc_seg: 78.7421, loss: 0.1784\n",
      "2021-04-20 10:08:36,278 - mmseg - INFO - Iter [960/5000]\tlr: 8.273e-03, eta: 1:17:53, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1046, decode.acc_seg: 80.8517, aux.loss_seg: 0.0466, aux.acc_seg: 80.5013, loss: 0.1512\n",
      "2021-04-20 10:08:42,588 - mmseg - INFO - Iter [970/5000]\tlr: 8.255e-03, eta: 1:17:20, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1123, decode.acc_seg: 81.5574, aux.loss_seg: 0.0502, aux.acc_seg: 81.1629, loss: 0.1625\n",
      "2021-04-20 10:08:48,906 - mmseg - INFO - Iter [980/5000]\tlr: 8.237e-03, eta: 1:16:47, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.1224, decode.acc_seg: 80.6526, aux.loss_seg: 0.0543, aux.acc_seg: 80.2679, loss: 0.1767\n",
      "2021-04-20 10:08:55,214 - mmseg - INFO - Iter [990/5000]\tlr: 8.219e-03, eta: 1:16:15, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1136, decode.acc_seg: 77.6450, aux.loss_seg: 0.0503, aux.acc_seg: 77.2388, loss: 0.1639\n",
      "2021-04-20 10:09:01,513 - mmseg - INFO - Saving checkpoint at 1000 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.2 task/s, elapsed: 117s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:11:16,459 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:11:16,460 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 95.8  | 97.78 |\n",
      "| cell       | 92.16 | 96.6  |\n",
      "| cilia      | 67.28 | 78.42 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:11:16,461 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:11:16,461 - mmseg - INFO - \n",
      "+--------+-------+-------+------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc |\n",
      "+--------+-------+-------+------+\n",
      "| global | 85.08 | 90.93 | 95.9 |\n",
      "+--------+-------+-------+------+\n",
      "2021-04-20 10:11:16,468 - mmseg - INFO - Iter(val) [1000]\tmIoU: 0.8508, mAcc: 0.9093, aAcc: 0.9590, IoU.background: 0.9580, IoU.cell: 0.9216, IoU.cilia: 0.6728, Acc.background: 0.9778, Acc.cell: 0.9660, Acc.cilia: 0.7842\n",
      "2021-04-20 10:11:22,721 - mmseg - INFO - Iter [1010/5000]\tlr: 8.182e-03, eta: 1:23:37, time: 14.060, data_time: 13.439, memory: 9591, decode.loss_seg: 0.1170, decode.acc_seg: 81.3333, aux.loss_seg: 0.0527, aux.acc_seg: 80.9242, loss: 0.1697\n",
      "2021-04-20 10:11:28,941 - mmseg - INFO - Iter [1020/5000]\tlr: 8.164e-03, eta: 1:23:00, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1206, decode.acc_seg: 78.4426, aux.loss_seg: 0.0530, aux.acc_seg: 78.0910, loss: 0.1736\n",
      "2021-04-20 10:11:35,167 - mmseg - INFO - Iter [1030/5000]\tlr: 8.146e-03, eta: 1:22:23, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1173, decode.acc_seg: 80.6368, aux.loss_seg: 0.0511, aux.acc_seg: 80.2721, loss: 0.1684\n",
      "2021-04-20 10:11:41,406 - mmseg - INFO - Iter [1040/5000]\tlr: 8.128e-03, eta: 1:21:47, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1000, decode.acc_seg: 78.2016, aux.loss_seg: 0.0441, aux.acc_seg: 77.9477, loss: 0.1441\n",
      "2021-04-20 10:11:47,672 - mmseg - INFO - Iter [1050/5000]\tlr: 8.109e-03, eta: 1:21:12, time: 0.627, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1204, decode.acc_seg: 84.8849, aux.loss_seg: 0.0543, aux.acc_seg: 84.3824, loss: 0.1747\n",
      "2021-04-20 10:11:56,050 - mmseg - INFO - Iter [1060/5000]\tlr: 8.091e-03, eta: 1:20:45, time: 0.838, data_time: 0.210, memory: 9591, decode.loss_seg: 0.1145, decode.acc_seg: 79.9378, aux.loss_seg: 0.0511, aux.acc_seg: 79.5163, loss: 0.1656\n",
      "2021-04-20 10:12:02,329 - mmseg - INFO - Iter [1070/5000]\tlr: 8.073e-03, eta: 1:20:10, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1153, decode.acc_seg: 77.4566, aux.loss_seg: 0.0500, aux.acc_seg: 77.1124, loss: 0.1653\n",
      "2021-04-20 10:12:08,621 - mmseg - INFO - Iter [1080/5000]\tlr: 8.055e-03, eta: 1:19:36, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1169, decode.acc_seg: 81.4794, aux.loss_seg: 0.0514, aux.acc_seg: 81.1132, loss: 0.1683\n",
      "2021-04-20 10:12:14,911 - mmseg - INFO - Iter [1090/5000]\tlr: 8.036e-03, eta: 1:19:03, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1196, decode.acc_seg: 78.6503, aux.loss_seg: 0.0519, aux.acc_seg: 78.2847, loss: 0.1715\n",
      "2021-04-20 10:12:21,213 - mmseg - INFO - Iter [1100/5000]\tlr: 8.018e-03, eta: 1:18:30, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1121, decode.acc_seg: 80.0079, aux.loss_seg: 0.0491, aux.acc_seg: 79.6412, loss: 0.1612\n",
      "2021-04-20 10:12:27,510 - mmseg - INFO - Iter [1110/5000]\tlr: 8.000e-03, eta: 1:17:58, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1050, decode.acc_seg: 83.3928, aux.loss_seg: 0.0469, aux.acc_seg: 83.0338, loss: 0.1520\n",
      "2021-04-20 10:12:33,808 - mmseg - INFO - Iter [1120/5000]\tlr: 7.982e-03, eta: 1:17:26, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1063, decode.acc_seg: 78.7445, aux.loss_seg: 0.0464, aux.acc_seg: 78.4653, loss: 0.1527\n",
      "2021-04-20 10:12:40,124 - mmseg - INFO - Iter [1130/5000]\tlr: 7.963e-03, eta: 1:16:55, time: 0.632, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1017, decode.acc_seg: 80.0095, aux.loss_seg: 0.0448, aux.acc_seg: 79.6748, loss: 0.1465\n",
      "2021-04-20 10:12:46,430 - mmseg - INFO - Iter [1140/5000]\tlr: 7.945e-03, eta: 1:16:24, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1047, decode.acc_seg: 83.8634, aux.loss_seg: 0.0465, aux.acc_seg: 83.5157, loss: 0.1512\n",
      "2021-04-20 10:12:52,735 - mmseg - INFO - Iter [1150/5000]\tlr: 7.927e-03, eta: 1:15:53, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1147, decode.acc_seg: 83.0510, aux.loss_seg: 0.0503, aux.acc_seg: 82.7007, loss: 0.1651\n",
      "2021-04-20 10:12:59,041 - mmseg - INFO - Iter [1160/5000]\tlr: 7.908e-03, eta: 1:15:23, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1278, decode.acc_seg: 83.7421, aux.loss_seg: 0.0563, aux.acc_seg: 83.3917, loss: 0.1841\n",
      "2021-04-20 10:13:05,345 - mmseg - INFO - Iter [1170/5000]\tlr: 7.890e-03, eta: 1:14:53, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1149, decode.acc_seg: 81.6756, aux.loss_seg: 0.0509, aux.acc_seg: 81.3595, loss: 0.1659\n",
      "2021-04-20 10:13:11,649 - mmseg - INFO - Iter [1180/5000]\tlr: 7.872e-03, eta: 1:14:24, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1211, decode.acc_seg: 78.9020, aux.loss_seg: 0.0550, aux.acc_seg: 78.2925, loss: 0.1761\n",
      "2021-04-20 10:13:17,952 - mmseg - INFO - Iter [1190/5000]\tlr: 7.853e-03, eta: 1:13:55, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1103, decode.acc_seg: 79.4330, aux.loss_seg: 0.0481, aux.acc_seg: 79.1286, loss: 0.1585\n",
      "2021-04-20 10:13:24,262 - mmseg - INFO - Saving checkpoint at 1200 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.3 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:15:39,014 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:15:39,015 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 96.0  | 97.95 |\n",
      "| cell       | 92.57 | 96.67 |\n",
      "| cilia      | 68.5  | 79.42 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:15:39,015 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:15:39,016 - mmseg - INFO - \n",
      "+--------+-------+-------+------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc |\n",
      "+--------+-------+-------+------+\n",
      "| global | 85.69 | 91.35 | 96.1 |\n",
      "+--------+-------+-------+------+\n",
      "2021-04-20 10:15:39,023 - mmseg - INFO - Iter(val) [1200]\tmIoU: 0.8569, mAcc: 0.9135, aAcc: 0.9610, IoU.background: 0.9600, IoU.cell: 0.9257, IoU.cilia: 0.6850, Acc.background: 0.9795, Acc.cell: 0.9667, Acc.cilia: 0.7942\n",
      "2021-04-20 10:15:45,263 - mmseg - INFO - Iter [1210/5000]\tlr: 7.817e-03, eta: 1:19:38, time: 14.024, data_time: 13.404, memory: 9591, decode.loss_seg: 0.1084, decode.acc_seg: 81.6272, aux.loss_seg: 0.0476, aux.acc_seg: 81.3207, loss: 0.1560\n",
      "2021-04-20 10:15:51,482 - mmseg - INFO - Iter [1220/5000]\tlr: 7.799e-03, eta: 1:19:06, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0996, decode.acc_seg: 80.9464, aux.loss_seg: 0.0446, aux.acc_seg: 80.6155, loss: 0.1442\n",
      "2021-04-20 10:15:57,713 - mmseg - INFO - Iter [1230/5000]\tlr: 7.780e-03, eta: 1:18:34, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1112, decode.acc_seg: 80.4247, aux.loss_seg: 0.0484, aux.acc_seg: 80.1005, loss: 0.1597\n",
      "2021-04-20 10:16:03,958 - mmseg - INFO - Iter [1240/5000]\tlr: 7.762e-03, eta: 1:18:02, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1033, decode.acc_seg: 82.0777, aux.loss_seg: 0.0450, aux.acc_seg: 81.7593, loss: 0.1483\n",
      "2021-04-20 10:16:10,233 - mmseg - INFO - Iter [1250/5000]\tlr: 7.744e-03, eta: 1:17:31, time: 0.627, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0991, decode.acc_seg: 80.1409, aux.loss_seg: 0.0426, aux.acc_seg: 79.9261, loss: 0.1416\n",
      "2021-04-20 10:16:16,520 - mmseg - INFO - Iter [1260/5000]\tlr: 7.725e-03, eta: 1:17:01, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1055, decode.acc_seg: 79.0229, aux.loss_seg: 0.0472, aux.acc_seg: 78.6717, loss: 0.1526\n",
      "2021-04-20 10:16:22,807 - mmseg - INFO - Iter [1270/5000]\tlr: 7.707e-03, eta: 1:16:31, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1036, decode.acc_seg: 80.6412, aux.loss_seg: 0.0451, aux.acc_seg: 80.4126, loss: 0.1487\n",
      "2021-04-20 10:16:29,112 - mmseg - INFO - Iter [1280/5000]\tlr: 7.688e-03, eta: 1:16:01, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1037, decode.acc_seg: 82.4626, aux.loss_seg: 0.0454, aux.acc_seg: 82.1516, loss: 0.1491\n",
      "2021-04-20 10:16:35,409 - mmseg - INFO - Iter [1290/5000]\tlr: 7.670e-03, eta: 1:15:31, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1025, decode.acc_seg: 79.0518, aux.loss_seg: 0.0436, aux.acc_seg: 78.8864, loss: 0.1461\n",
      "2021-04-20 10:16:41,706 - mmseg - INFO - Iter [1300/5000]\tlr: 7.652e-03, eta: 1:15:02, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1005, decode.acc_seg: 81.3221, aux.loss_seg: 0.0439, aux.acc_seg: 81.0328, loss: 0.1444\n",
      "2021-04-20 10:16:48,008 - mmseg - INFO - Iter [1310/5000]\tlr: 7.633e-03, eta: 1:14:34, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0916, decode.acc_seg: 81.4081, aux.loss_seg: 0.0406, aux.acc_seg: 81.1326, loss: 0.1322\n",
      "2021-04-20 10:16:54,312 - mmseg - INFO - Iter [1320/5000]\tlr: 7.615e-03, eta: 1:14:05, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0948, decode.acc_seg: 78.0033, aux.loss_seg: 0.0419, aux.acc_seg: 77.6950, loss: 0.1367\n",
      "2021-04-20 10:17:00,624 - mmseg - INFO - Iter [1330/5000]\tlr: 7.597e-03, eta: 1:13:37, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0952, decode.acc_seg: 80.7804, aux.loss_seg: 0.0416, aux.acc_seg: 80.5088, loss: 0.1368\n",
      "2021-04-20 10:17:06,929 - mmseg - INFO - Iter [1340/5000]\tlr: 7.578e-03, eta: 1:13:10, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1054, decode.acc_seg: 80.0939, aux.loss_seg: 0.0460, aux.acc_seg: 79.7452, loss: 0.1514\n",
      "2021-04-20 10:17:13,234 - mmseg - INFO - Iter [1350/5000]\tlr: 7.560e-03, eta: 1:12:42, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0923, decode.acc_seg: 80.3950, aux.loss_seg: 0.0403, aux.acc_seg: 80.1969, loss: 0.1327\n",
      "2021-04-20 10:17:19,541 - mmseg - INFO - Iter [1360/5000]\tlr: 7.542e-03, eta: 1:12:15, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.1017, decode.acc_seg: 85.6380, aux.loss_seg: 0.0440, aux.acc_seg: 85.3759, loss: 0.1456\n",
      "2021-04-20 10:17:25,844 - mmseg - INFO - Iter [1370/5000]\tlr: 7.523e-03, eta: 1:11:48, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0905, decode.acc_seg: 85.1127, aux.loss_seg: 0.0396, aux.acc_seg: 84.8716, loss: 0.1301\n",
      "2021-04-20 10:17:32,146 - mmseg - INFO - Iter [1380/5000]\tlr: 7.505e-03, eta: 1:11:22, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0915, decode.acc_seg: 83.3581, aux.loss_seg: 0.0399, aux.acc_seg: 83.1217, loss: 0.1314\n",
      "2021-04-20 10:17:38,451 - mmseg - INFO - Iter [1390/5000]\tlr: 7.486e-03, eta: 1:10:56, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0862, decode.acc_seg: 80.9987, aux.loss_seg: 0.0384, aux.acc_seg: 80.7333, loss: 0.1245\n",
      "2021-04-20 10:17:44,751 - mmseg - INFO - Saving checkpoint at 1400 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.3 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:19:59,434 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:19:59,435 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 96.3  | 97.88 |\n",
      "| cell       | 93.05 | 97.07 |\n",
      "| cilia      | 70.86 | 81.99 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:19:59,435 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:19:59,435 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 86.73 | 92.31 | 96.39 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 10:19:59,442 - mmseg - INFO - Iter(val) [1400]\tmIoU: 0.8673, mAcc: 0.9231, aAcc: 0.9639, IoU.background: 0.9630, IoU.cell: 0.9305, IoU.cilia: 0.7086, Acc.background: 0.9788, Acc.cell: 0.9707, Acc.cilia: 0.8199\n",
      "2021-04-20 10:20:05,675 - mmseg - INFO - Iter [1410/5000]\tlr: 7.449e-03, eta: 1:15:29, time: 14.034, data_time: 13.414, memory: 9591, decode.loss_seg: 0.0871, decode.acc_seg: 80.9816, aux.loss_seg: 0.0383, aux.acc_seg: 80.7373, loss: 0.1254\n",
      "2021-04-20 10:20:11,890 - mmseg - INFO - Iter [1420/5000]\tlr: 7.431e-03, eta: 1:15:01, time: 0.621, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0939, decode.acc_seg: 81.7978, aux.loss_seg: 0.0409, aux.acc_seg: 81.5250, loss: 0.1348\n",
      "2021-04-20 10:20:18,127 - mmseg - INFO - Iter [1430/5000]\tlr: 7.413e-03, eta: 1:14:32, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0891, decode.acc_seg: 80.1867, aux.loss_seg: 0.0391, aux.acc_seg: 79.8991, loss: 0.1281\n",
      "2021-04-20 10:20:24,377 - mmseg - INFO - Iter [1440/5000]\tlr: 7.394e-03, eta: 1:14:04, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0952, decode.acc_seg: 78.0988, aux.loss_seg: 0.0422, aux.acc_seg: 77.7805, loss: 0.1375\n",
      "2021-04-20 10:20:30,649 - mmseg - INFO - Iter [1450/5000]\tlr: 7.376e-03, eta: 1:13:37, time: 0.627, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0910, decode.acc_seg: 78.4097, aux.loss_seg: 0.0404, aux.acc_seg: 78.1849, loss: 0.1314\n",
      "2021-04-20 10:20:36,928 - mmseg - INFO - Iter [1460/5000]\tlr: 7.357e-03, eta: 1:13:09, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0960, decode.acc_seg: 81.7671, aux.loss_seg: 0.0424, aux.acc_seg: 81.5081, loss: 0.1384\n",
      "2021-04-20 10:20:43,222 - mmseg - INFO - Iter [1470/5000]\tlr: 7.339e-03, eta: 1:12:42, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0983, decode.acc_seg: 78.6599, aux.loss_seg: 0.0435, aux.acc_seg: 78.3680, loss: 0.1418\n",
      "2021-04-20 10:20:49,516 - mmseg - INFO - Iter [1480/5000]\tlr: 7.320e-03, eta: 1:12:15, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0897, decode.acc_seg: 82.5668, aux.loss_seg: 0.0399, aux.acc_seg: 82.2704, loss: 0.1296\n",
      "2021-04-20 10:20:55,819 - mmseg - INFO - Iter [1490/5000]\tlr: 7.302e-03, eta: 1:11:49, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0956, decode.acc_seg: 81.0449, aux.loss_seg: 0.0421, aux.acc_seg: 80.7630, loss: 0.1377\n",
      "2021-04-20 10:21:02,125 - mmseg - INFO - Iter [1500/5000]\tlr: 7.283e-03, eta: 1:11:23, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0938, decode.acc_seg: 81.0032, aux.loss_seg: 0.0403, aux.acc_seg: 80.8134, loss: 0.1341\n",
      "2021-04-20 10:21:08,434 - mmseg - INFO - Iter [1510/5000]\tlr: 7.265e-03, eta: 1:10:57, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0952, decode.acc_seg: 82.0002, aux.loss_seg: 0.0418, aux.acc_seg: 81.7914, loss: 0.1370\n",
      "2021-04-20 10:21:14,741 - mmseg - INFO - Iter [1520/5000]\tlr: 7.247e-03, eta: 1:10:31, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0887, decode.acc_seg: 80.9655, aux.loss_seg: 0.0397, aux.acc_seg: 80.6333, loss: 0.1284\n",
      "2021-04-20 10:21:21,045 - mmseg - INFO - Iter [1530/5000]\tlr: 7.228e-03, eta: 1:10:06, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0825, decode.acc_seg: 81.8001, aux.loss_seg: 0.0366, aux.acc_seg: 81.5354, loss: 0.1190\n",
      "2021-04-20 10:21:27,353 - mmseg - INFO - Iter [1540/5000]\tlr: 7.210e-03, eta: 1:09:40, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0861, decode.acc_seg: 83.1854, aux.loss_seg: 0.0384, aux.acc_seg: 82.8968, loss: 0.1245\n",
      "2021-04-20 10:21:33,661 - mmseg - INFO - Iter [1550/5000]\tlr: 7.191e-03, eta: 1:09:15, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0902, decode.acc_seg: 82.8847, aux.loss_seg: 0.0395, aux.acc_seg: 82.6543, loss: 0.1297\n",
      "2021-04-20 10:21:39,966 - mmseg - INFO - Iter [1560/5000]\tlr: 7.173e-03, eta: 1:08:51, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0835, decode.acc_seg: 81.2079, aux.loss_seg: 0.0360, aux.acc_seg: 81.0319, loss: 0.1195\n",
      "2021-04-20 10:21:46,274 - mmseg - INFO - Iter [1570/5000]\tlr: 7.154e-03, eta: 1:08:26, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0845, decode.acc_seg: 80.4046, aux.loss_seg: 0.0368, aux.acc_seg: 80.1965, loss: 0.1213\n",
      "2021-04-20 10:21:52,581 - mmseg - INFO - Iter [1580/5000]\tlr: 7.136e-03, eta: 1:08:02, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0855, decode.acc_seg: 82.0005, aux.loss_seg: 0.0379, aux.acc_seg: 81.7342, loss: 0.1234\n",
      "2021-04-20 10:21:58,888 - mmseg - INFO - Iter [1590/5000]\tlr: 7.117e-03, eta: 1:07:38, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0884, decode.acc_seg: 80.7062, aux.loss_seg: 0.0388, aux.acc_seg: 80.4155, loss: 0.1272\n",
      "2021-04-20 10:22:05,194 - mmseg - INFO - Saving checkpoint at 1600 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.4 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:24:19,419 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:24:19,420 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 96.44 | 98.12 |\n",
      "| cell       | 93.58 | 97.59 |\n",
      "| cilia      | 71.98 | 80.79 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:24:19,420 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:24:19,421 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 87.33 | 92.17 | 96.61 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 10:24:19,428 - mmseg - INFO - Iter(val) [1600]\tmIoU: 0.8733, mAcc: 0.9217, aAcc: 0.9661, IoU.background: 0.9644, IoU.cell: 0.9358, IoU.cilia: 0.7198, Acc.background: 0.9812, Acc.cell: 0.9759, Acc.cilia: 0.8079\n",
      "2021-04-20 10:24:25,662 - mmseg - INFO - Iter [1610/5000]\tlr: 7.080e-03, eta: 1:11:19, time: 13.989, data_time: 13.370, memory: 9591, decode.loss_seg: 0.0942, decode.acc_seg: 82.4956, aux.loss_seg: 0.0407, aux.acc_seg: 82.2752, loss: 0.1349\n",
      "2021-04-20 10:24:31,881 - mmseg - INFO - Iter [1620/5000]\tlr: 7.062e-03, eta: 1:10:53, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0810, decode.acc_seg: 81.9016, aux.loss_seg: 0.0364, aux.acc_seg: 81.5975, loss: 0.1174\n",
      "2021-04-20 10:24:38,118 - mmseg - INFO - Iter [1630/5000]\tlr: 7.043e-03, eta: 1:10:27, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0978, decode.acc_seg: 83.5213, aux.loss_seg: 0.0429, aux.acc_seg: 83.2402, loss: 0.1406\n",
      "2021-04-20 10:24:44,366 - mmseg - INFO - Iter [1640/5000]\tlr: 7.024e-03, eta: 1:10:01, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0744, decode.acc_seg: 80.3741, aux.loss_seg: 0.0332, aux.acc_seg: 80.1132, loss: 0.1076\n",
      "2021-04-20 10:24:50,642 - mmseg - INFO - Iter [1650/5000]\tlr: 7.006e-03, eta: 1:09:36, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0914, decode.acc_seg: 81.8556, aux.loss_seg: 0.0401, aux.acc_seg: 81.5785, loss: 0.1315\n",
      "2021-04-20 10:24:56,922 - mmseg - INFO - Iter [1660/5000]\tlr: 6.987e-03, eta: 1:09:11, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0812, decode.acc_seg: 79.6243, aux.loss_seg: 0.0357, aux.acc_seg: 79.3579, loss: 0.1169\n",
      "2021-04-20 10:25:03,216 - mmseg - INFO - Iter [1670/5000]\tlr: 6.969e-03, eta: 1:08:47, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0955, decode.acc_seg: 82.5910, aux.loss_seg: 0.0424, aux.acc_seg: 82.2094, loss: 0.1379\n",
      "2021-04-20 10:25:09,511 - mmseg - INFO - Iter [1680/5000]\tlr: 6.950e-03, eta: 1:08:22, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0796, decode.acc_seg: 83.7769, aux.loss_seg: 0.0359, aux.acc_seg: 83.4359, loss: 0.1155\n",
      "2021-04-20 10:25:15,820 - mmseg - INFO - Iter [1690/5000]\tlr: 6.932e-03, eta: 1:07:58, time: 0.631, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0968, decode.acc_seg: 83.1977, aux.loss_seg: 0.0428, aux.acc_seg: 82.8876, loss: 0.1396\n",
      "2021-04-20 10:25:22,122 - mmseg - INFO - Iter [1700/5000]\tlr: 6.913e-03, eta: 1:07:34, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0882, decode.acc_seg: 84.3336, aux.loss_seg: 0.0389, aux.acc_seg: 84.0834, loss: 0.1271\n",
      "2021-04-20 10:25:28,434 - mmseg - INFO - Iter [1710/5000]\tlr: 6.894e-03, eta: 1:07:10, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0791, decode.acc_seg: 81.3532, aux.loss_seg: 0.0352, aux.acc_seg: 81.1337, loss: 0.1143\n",
      "2021-04-20 10:25:34,744 - mmseg - INFO - Iter [1720/5000]\tlr: 6.876e-03, eta: 1:06:47, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0798, decode.acc_seg: 79.9644, aux.loss_seg: 0.0358, aux.acc_seg: 79.7579, loss: 0.1156\n",
      "2021-04-20 10:25:41,063 - mmseg - INFO - Iter [1730/5000]\tlr: 6.857e-03, eta: 1:06:23, time: 0.632, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0831, decode.acc_seg: 80.0718, aux.loss_seg: 0.0368, aux.acc_seg: 79.7882, loss: 0.1199\n",
      "2021-04-20 10:25:47,373 - mmseg - INFO - Iter [1740/5000]\tlr: 6.839e-03, eta: 1:06:00, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0881, decode.acc_seg: 83.9262, aux.loss_seg: 0.0391, aux.acc_seg: 83.6884, loss: 0.1272\n",
      "2021-04-20 10:25:53,686 - mmseg - INFO - Iter [1750/5000]\tlr: 6.820e-03, eta: 1:05:37, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0836, decode.acc_seg: 81.9702, aux.loss_seg: 0.0370, aux.acc_seg: 81.7197, loss: 0.1206\n",
      "2021-04-20 10:25:59,993 - mmseg - INFO - Iter [1760/5000]\tlr: 6.802e-03, eta: 1:05:14, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0785, decode.acc_seg: 81.3110, aux.loss_seg: 0.0349, aux.acc_seg: 81.0396, loss: 0.1134\n",
      "2021-04-20 10:26:06,309 - mmseg - INFO - Iter [1770/5000]\tlr: 6.783e-03, eta: 1:04:52, time: 0.632, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0868, decode.acc_seg: 83.0908, aux.loss_seg: 0.0384, aux.acc_seg: 82.8751, loss: 0.1252\n",
      "2021-04-20 10:26:12,618 - mmseg - INFO - Iter [1780/5000]\tlr: 6.764e-03, eta: 1:04:29, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0778, decode.acc_seg: 81.3496, aux.loss_seg: 0.0344, aux.acc_seg: 81.0933, loss: 0.1123\n",
      "2021-04-20 10:26:18,929 - mmseg - INFO - Iter [1790/5000]\tlr: 6.746e-03, eta: 1:04:07, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0836, decode.acc_seg: 79.9158, aux.loss_seg: 0.0357, aux.acc_seg: 79.7056, loss: 0.1194\n",
      "2021-04-20 10:26:25,231 - mmseg - INFO - Saving checkpoint at 1800 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.3 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:28:39,759 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:28:39,760 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 96.8  | 98.47 |\n",
      "| cell       | 94.18 | 96.91 |\n",
      "| cilia      | 74.6  | 85.15 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:28:39,760 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:28:39,761 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 88.53 | 93.51 | 96.93 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 10:28:39,768 - mmseg - INFO - Iter(val) [1800]\tmIoU: 0.8853, mAcc: 0.9351, aAcc: 0.9693, IoU.background: 0.9680, IoU.cell: 0.9418, IoU.cilia: 0.7460, Acc.background: 0.9847, Acc.cell: 0.9691, Acc.cilia: 0.8515\n",
      "2021-04-20 10:28:46,002 - mmseg - INFO - Iter [1810/5000]\tlr: 6.708e-03, eta: 1:07:08, time: 14.017, data_time: 13.398, memory: 9591, decode.loss_seg: 0.0817, decode.acc_seg: 83.7486, aux.loss_seg: 0.0355, aux.acc_seg: 83.5895, loss: 0.1171\n",
      "2021-04-20 10:28:52,228 - mmseg - INFO - Iter [1820/5000]\tlr: 6.690e-03, eta: 1:06:44, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0828, decode.acc_seg: 80.4604, aux.loss_seg: 0.0371, aux.acc_seg: 80.1117, loss: 0.1199\n",
      "2021-04-20 10:28:58,476 - mmseg - INFO - Iter [1830/5000]\tlr: 6.671e-03, eta: 1:06:20, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0862, decode.acc_seg: 81.4579, aux.loss_seg: 0.0379, aux.acc_seg: 81.2069, loss: 0.1241\n",
      "2021-04-20 10:29:04,734 - mmseg - INFO - Iter [1840/5000]\tlr: 6.652e-03, eta: 1:05:57, time: 0.626, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0751, decode.acc_seg: 80.0944, aux.loss_seg: 0.0328, aux.acc_seg: 79.9225, loss: 0.1080\n",
      "2021-04-20 10:29:11,016 - mmseg - INFO - Iter [1850/5000]\tlr: 6.634e-03, eta: 1:05:34, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0780, decode.acc_seg: 81.3121, aux.loss_seg: 0.0340, aux.acc_seg: 81.1239, loss: 0.1120\n",
      "2021-04-20 10:29:17,301 - mmseg - INFO - Iter [1860/5000]\tlr: 6.615e-03, eta: 1:05:11, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0845, decode.acc_seg: 81.7848, aux.loss_seg: 0.0374, aux.acc_seg: 81.4909, loss: 0.1218\n",
      "2021-04-20 10:29:23,594 - mmseg - INFO - Iter [1870/5000]\tlr: 6.596e-03, eta: 1:04:48, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0793, decode.acc_seg: 80.0685, aux.loss_seg: 0.0345, aux.acc_seg: 79.8423, loss: 0.1138\n",
      "2021-04-20 10:29:29,886 - mmseg - INFO - Iter [1880/5000]\tlr: 6.578e-03, eta: 1:04:26, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0756, decode.acc_seg: 82.9099, aux.loss_seg: 0.0332, aux.acc_seg: 82.7031, loss: 0.1088\n",
      "2021-04-20 10:29:36,189 - mmseg - INFO - Iter [1890/5000]\tlr: 6.559e-03, eta: 1:04:03, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0765, decode.acc_seg: 81.6047, aux.loss_seg: 0.0335, aux.acc_seg: 81.3759, loss: 0.1100\n",
      "2021-04-20 10:29:42,493 - mmseg - INFO - Iter [1900/5000]\tlr: 6.540e-03, eta: 1:03:41, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0739, decode.acc_seg: 79.8277, aux.loss_seg: 0.0327, aux.acc_seg: 79.5585, loss: 0.1066\n",
      "2021-04-20 10:29:48,802 - mmseg - INFO - Iter [1910/5000]\tlr: 6.522e-03, eta: 1:03:19, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0729, decode.acc_seg: 79.6545, aux.loss_seg: 0.0319, aux.acc_seg: 79.4223, loss: 0.1048\n",
      "2021-04-20 10:29:55,111 - mmseg - INFO - Iter [1920/5000]\tlr: 6.503e-03, eta: 1:02:57, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0828, decode.acc_seg: 81.7900, aux.loss_seg: 0.0365, aux.acc_seg: 81.5259, loss: 0.1193\n",
      "2021-04-20 10:30:01,421 - mmseg - INFO - Iter [1930/5000]\tlr: 6.484e-03, eta: 1:02:35, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0817, decode.acc_seg: 80.3430, aux.loss_seg: 0.0361, aux.acc_seg: 80.0914, loss: 0.1178\n",
      "2021-04-20 10:30:07,734 - mmseg - INFO - Iter [1940/5000]\tlr: 6.466e-03, eta: 1:02:14, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0765, decode.acc_seg: 83.1466, aux.loss_seg: 0.0338, aux.acc_seg: 82.8930, loss: 0.1103\n",
      "2021-04-20 10:30:14,048 - mmseg - INFO - Iter [1950/5000]\tlr: 6.447e-03, eta: 1:01:52, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0706, decode.acc_seg: 80.6688, aux.loss_seg: 0.0312, aux.acc_seg: 80.4276, loss: 0.1017\n",
      "2021-04-20 10:30:20,362 - mmseg - INFO - Iter [1960/5000]\tlr: 6.428e-03, eta: 1:01:31, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0797, decode.acc_seg: 82.6648, aux.loss_seg: 0.0350, aux.acc_seg: 82.4520, loss: 0.1147\n",
      "2021-04-20 10:30:26,671 - mmseg - INFO - Iter [1970/5000]\tlr: 6.409e-03, eta: 1:01:10, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0718, decode.acc_seg: 80.9195, aux.loss_seg: 0.0312, aux.acc_seg: 80.7435, loss: 0.1030\n",
      "2021-04-20 10:30:32,985 - mmseg - INFO - Iter [1980/5000]\tlr: 6.391e-03, eta: 1:00:49, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0802, decode.acc_seg: 81.2573, aux.loss_seg: 0.0351, aux.acc_seg: 81.0058, loss: 0.1152\n",
      "2021-04-20 10:30:39,294 - mmseg - INFO - Iter [1990/5000]\tlr: 6.372e-03, eta: 1:00:28, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0721, decode.acc_seg: 82.2543, aux.loss_seg: 0.0317, aux.acc_seg: 82.0362, loss: 0.1037\n",
      "2021-04-20 10:30:45,597 - mmseg - INFO - Saving checkpoint at 2000 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.7 task/s, elapsed: 115s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:32:58,931 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:32:58,931 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 96.81 | 98.0  |\n",
      "| cell       | 94.51 | 97.17 |\n",
      "| cilia      | 76.14 | 89.02 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:32:58,932 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:32:58,932 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 89.15 | 94.73 | 97.04 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 10:32:58,939 - mmseg - INFO - Iter(val) [2000]\tmIoU: 0.8915, mAcc: 0.9473, aAcc: 0.9704, IoU.background: 0.9681, IoU.cell: 0.9451, IoU.cilia: 0.7614, Acc.background: 0.9800, Acc.cell: 0.9717, Acc.cilia: 0.8902\n",
      "2021-04-20 10:33:05,171 - mmseg - INFO - Iter [2010/5000]\tlr: 6.334e-03, eta: 1:02:55, time: 13.900, data_time: 13.281, memory: 9591, decode.loss_seg: 0.0752, decode.acc_seg: 78.8664, aux.loss_seg: 0.0333, aux.acc_seg: 78.5875, loss: 0.1085\n",
      "2021-04-20 10:33:11,399 - mmseg - INFO - Iter [2020/5000]\tlr: 6.316e-03, eta: 1:02:33, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0854, decode.acc_seg: 81.8616, aux.loss_seg: 0.0367, aux.acc_seg: 81.6578, loss: 0.1221\n",
      "2021-04-20 10:33:17,634 - mmseg - INFO - Iter [2030/5000]\tlr: 6.297e-03, eta: 1:02:11, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0752, decode.acc_seg: 82.0343, aux.loss_seg: 0.0331, aux.acc_seg: 81.8091, loss: 0.1084\n",
      "2021-04-20 10:33:23,885 - mmseg - INFO - Iter [2040/5000]\tlr: 6.278e-03, eta: 1:01:49, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0771, decode.acc_seg: 80.2664, aux.loss_seg: 0.0335, aux.acc_seg: 80.0530, loss: 0.1106\n",
      "2021-04-20 10:33:30,162 - mmseg - INFO - Iter [2050/5000]\tlr: 6.259e-03, eta: 1:01:28, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0719, decode.acc_seg: 82.0609, aux.loss_seg: 0.0317, aux.acc_seg: 81.8675, loss: 0.1036\n",
      "2021-04-20 10:33:36,444 - mmseg - INFO - Iter [2060/5000]\tlr: 6.241e-03, eta: 1:01:06, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0723, decode.acc_seg: 80.3628, aux.loss_seg: 0.0319, aux.acc_seg: 80.1444, loss: 0.1042\n",
      "2021-04-20 10:33:42,738 - mmseg - INFO - Iter [2070/5000]\tlr: 6.222e-03, eta: 1:00:45, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0753, decode.acc_seg: 79.8583, aux.loss_seg: 0.0331, aux.acc_seg: 79.6361, loss: 0.1085\n",
      "2021-04-20 10:33:49,036 - mmseg - INFO - Iter [2080/5000]\tlr: 6.203e-03, eta: 1:00:24, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0798, decode.acc_seg: 83.6149, aux.loss_seg: 0.0344, aux.acc_seg: 83.4526, loss: 0.1142\n",
      "2021-04-20 10:33:55,327 - mmseg - INFO - Iter [2090/5000]\tlr: 6.184e-03, eta: 1:00:03, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0719, decode.acc_seg: 83.5307, aux.loss_seg: 0.0314, aux.acc_seg: 83.3311, loss: 0.1034\n",
      "2021-04-20 10:34:01,627 - mmseg - INFO - Iter [2100/5000]\tlr: 6.165e-03, eta: 0:59:42, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0743, decode.acc_seg: 82.4792, aux.loss_seg: 0.0326, aux.acc_seg: 82.2807, loss: 0.1068\n",
      "2021-04-20 10:34:07,928 - mmseg - INFO - Iter [2110/5000]\tlr: 6.147e-03, eta: 0:59:22, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0723, decode.acc_seg: 83.3112, aux.loss_seg: 0.0314, aux.acc_seg: 83.1423, loss: 0.1037\n",
      "2021-04-20 10:34:16,442 - mmseg - INFO - Iter [2120/5000]\tlr: 6.128e-03, eta: 0:59:04, time: 0.851, data_time: 0.221, memory: 9591, decode.loss_seg: 0.0796, decode.acc_seg: 83.7761, aux.loss_seg: 0.0352, aux.acc_seg: 83.5174, loss: 0.1148\n",
      "2021-04-20 10:34:22,746 - mmseg - INFO - Iter [2130/5000]\tlr: 6.109e-03, eta: 0:58:44, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0676, decode.acc_seg: 80.7195, aux.loss_seg: 0.0300, aux.acc_seg: 80.5140, loss: 0.0976\n",
      "2021-04-20 10:34:29,047 - mmseg - INFO - Iter [2140/5000]\tlr: 6.090e-03, eta: 0:58:23, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0716, decode.acc_seg: 84.9273, aux.loss_seg: 0.0313, aux.acc_seg: 84.7416, loss: 0.1029\n",
      "2021-04-20 10:34:35,356 - mmseg - INFO - Iter [2150/5000]\tlr: 6.071e-03, eta: 0:58:03, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0791, decode.acc_seg: 84.3816, aux.loss_seg: 0.0348, aux.acc_seg: 84.1514, loss: 0.1140\n",
      "2021-04-20 10:34:41,668 - mmseg - INFO - Iter [2160/5000]\tlr: 6.052e-03, eta: 0:57:43, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0690, decode.acc_seg: 80.0834, aux.loss_seg: 0.0309, aux.acc_seg: 79.8332, loss: 0.0999\n",
      "2021-04-20 10:34:47,982 - mmseg - INFO - Iter [2170/5000]\tlr: 6.033e-03, eta: 0:57:23, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0773, decode.acc_seg: 82.5610, aux.loss_seg: 0.0340, aux.acc_seg: 82.3707, loss: 0.1113\n",
      "2021-04-20 10:34:54,288 - mmseg - INFO - Iter [2180/5000]\tlr: 6.015e-03, eta: 0:57:04, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0669, decode.acc_seg: 81.0825, aux.loss_seg: 0.0297, aux.acc_seg: 80.9008, loss: 0.0966\n",
      "2021-04-20 10:35:00,597 - mmseg - INFO - Iter [2190/5000]\tlr: 5.996e-03, eta: 0:56:44, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0742, decode.acc_seg: 81.8259, aux.loss_seg: 0.0327, aux.acc_seg: 81.6038, loss: 0.1069\n",
      "2021-04-20 10:35:06,896 - mmseg - INFO - Saving checkpoint at 2200 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.1 task/s, elapsed: 117s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:37:22,205 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:37:22,206 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.07 | 98.46 |\n",
      "| cell       | 94.81 | 97.75 |\n",
      "| cilia      | 76.76 | 85.61 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:37:22,206 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:37:22,207 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 89.55 | 93.94 | 97.24 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 10:37:22,214 - mmseg - INFO - Iter(val) [2200]\tmIoU: 0.8955, mAcc: 0.9394, aAcc: 0.9724, IoU.background: 0.9707, IoU.cell: 0.9481, IoU.cilia: 0.7676, Acc.background: 0.9846, Acc.cell: 0.9775, Acc.cilia: 0.8561\n",
      "2021-04-20 10:37:28,454 - mmseg - INFO - Iter [2210/5000]\tlr: 5.958e-03, eta: 0:58:47, time: 14.081, data_time: 13.461, memory: 9591, decode.loss_seg: 0.0689, decode.acc_seg: 82.0695, aux.loss_seg: 0.0308, aux.acc_seg: 81.8519, loss: 0.0996\n",
      "2021-04-20 10:37:34,675 - mmseg - INFO - Iter [2220/5000]\tlr: 5.939e-03, eta: 0:58:26, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0730, decode.acc_seg: 81.2425, aux.loss_seg: 0.0328, aux.acc_seg: 80.9498, loss: 0.1058\n",
      "2021-04-20 10:37:40,909 - mmseg - INFO - Iter [2230/5000]\tlr: 5.920e-03, eta: 0:58:06, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0752, decode.acc_seg: 83.5850, aux.loss_seg: 0.0331, aux.acc_seg: 83.3663, loss: 0.1083\n",
      "2021-04-20 10:37:47,149 - mmseg - INFO - Iter [2240/5000]\tlr: 5.901e-03, eta: 0:57:45, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0688, decode.acc_seg: 79.5214, aux.loss_seg: 0.0302, aux.acc_seg: 79.3218, loss: 0.0991\n",
      "2021-04-20 10:37:53,419 - mmseg - INFO - Iter [2250/5000]\tlr: 5.882e-03, eta: 0:57:25, time: 0.627, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0769, decode.acc_seg: 82.4331, aux.loss_seg: 0.0339, aux.acc_seg: 82.2116, loss: 0.1107\n",
      "2021-04-20 10:37:59,706 - mmseg - INFO - Iter [2260/5000]\tlr: 5.863e-03, eta: 0:57:05, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0731, decode.acc_seg: 86.1268, aux.loss_seg: 0.0327, aux.acc_seg: 85.8585, loss: 0.1057\n",
      "2021-04-20 10:38:05,992 - mmseg - INFO - Iter [2270/5000]\tlr: 5.844e-03, eta: 0:56:45, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0685, decode.acc_seg: 80.1933, aux.loss_seg: 0.0300, aux.acc_seg: 79.9985, loss: 0.0985\n",
      "2021-04-20 10:38:12,284 - mmseg - INFO - Iter [2280/5000]\tlr: 5.826e-03, eta: 0:56:25, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0659, decode.acc_seg: 82.0823, aux.loss_seg: 0.0293, aux.acc_seg: 81.8632, loss: 0.0952\n",
      "2021-04-20 10:38:18,586 - mmseg - INFO - Iter [2290/5000]\tlr: 5.807e-03, eta: 0:56:06, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0748, decode.acc_seg: 85.1584, aux.loss_seg: 0.0331, aux.acc_seg: 84.9338, loss: 0.1079\n",
      "2021-04-20 10:38:24,887 - mmseg - INFO - Iter [2300/5000]\tlr: 5.788e-03, eta: 0:55:46, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0700, decode.acc_seg: 80.4021, aux.loss_seg: 0.0309, aux.acc_seg: 80.1511, loss: 0.1009\n",
      "2021-04-20 10:38:31,196 - mmseg - INFO - Iter [2310/5000]\tlr: 5.769e-03, eta: 0:55:26, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0744, decode.acc_seg: 80.4446, aux.loss_seg: 0.0318, aux.acc_seg: 80.2924, loss: 0.1062\n",
      "2021-04-20 10:38:37,508 - mmseg - INFO - Iter [2320/5000]\tlr: 5.750e-03, eta: 0:55:07, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0716, decode.acc_seg: 82.9690, aux.loss_seg: 0.0315, aux.acc_seg: 82.7508, loss: 0.1031\n",
      "2021-04-20 10:38:43,815 - mmseg - INFO - Iter [2330/5000]\tlr: 5.731e-03, eta: 0:54:48, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0652, decode.acc_seg: 85.3472, aux.loss_seg: 0.0287, aux.acc_seg: 85.1499, loss: 0.0939\n",
      "2021-04-20 10:38:50,118 - mmseg - INFO - Iter [2340/5000]\tlr: 5.712e-03, eta: 0:54:29, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0645, decode.acc_seg: 81.3057, aux.loss_seg: 0.0286, aux.acc_seg: 81.0903, loss: 0.0930\n",
      "2021-04-20 10:38:56,444 - mmseg - INFO - Iter [2350/5000]\tlr: 5.693e-03, eta: 0:54:10, time: 0.633, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0721, decode.acc_seg: 84.4012, aux.loss_seg: 0.0315, aux.acc_seg: 84.2247, loss: 0.1036\n",
      "2021-04-20 10:39:02,770 - mmseg - INFO - Iter [2360/5000]\tlr: 5.674e-03, eta: 0:53:51, time: 0.633, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0706, decode.acc_seg: 83.2597, aux.loss_seg: 0.0307, aux.acc_seg: 83.0961, loss: 0.1012\n",
      "2021-04-20 10:39:09,079 - mmseg - INFO - Iter [2370/5000]\tlr: 5.655e-03, eta: 0:53:32, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0692, decode.acc_seg: 85.5319, aux.loss_seg: 0.0307, aux.acc_seg: 85.3318, loss: 0.0999\n",
      "2021-04-20 10:39:15,391 - mmseg - INFO - Iter [2380/5000]\tlr: 5.636e-03, eta: 0:53:13, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0662, decode.acc_seg: 84.2496, aux.loss_seg: 0.0300, aux.acc_seg: 83.9407, loss: 0.0962\n",
      "2021-04-20 10:39:21,700 - mmseg - INFO - Iter [2390/5000]\tlr: 5.617e-03, eta: 0:52:55, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0697, decode.acc_seg: 80.7133, aux.loss_seg: 0.0307, aux.acc_seg: 80.4695, loss: 0.1004\n",
      "2021-04-20 10:39:28,001 - mmseg - INFO - Saving checkpoint at 2400 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.1 task/s, elapsed: 117s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:41:43,044 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:41:43,045 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.12 | 98.48 |\n",
      "| cell       | 94.82 | 96.98 |\n",
      "| cilia      | 77.53 | 89.16 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:41:43,045 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:41:43,046 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 89.82 | 94.87 | 97.26 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 10:41:43,053 - mmseg - INFO - Iter(val) [2400]\tmIoU: 0.8982, mAcc: 0.9487, aAcc: 0.9726, IoU.background: 0.9712, IoU.cell: 0.9482, IoU.cilia: 0.7753, Acc.background: 0.9848, Acc.cell: 0.9698, Acc.cilia: 0.8916\n",
      "2021-04-20 10:41:49,314 - mmseg - INFO - Iter [2410/5000]\tlr: 5.579e-03, eta: 0:54:35, time: 14.075, data_time: 13.453, memory: 9591, decode.loss_seg: 0.0682, decode.acc_seg: 82.2094, aux.loss_seg: 0.0304, aux.acc_seg: 82.0070, loss: 0.0987\n",
      "2021-04-20 10:41:55,535 - mmseg - INFO - Iter [2420/5000]\tlr: 5.560e-03, eta: 0:54:16, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0698, decode.acc_seg: 83.7942, aux.loss_seg: 0.0309, aux.acc_seg: 83.5519, loss: 0.1007\n",
      "2021-04-20 10:42:01,771 - mmseg - INFO - Iter [2430/5000]\tlr: 5.541e-03, eta: 0:53:57, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0611, decode.acc_seg: 81.5765, aux.loss_seg: 0.0275, aux.acc_seg: 81.3562, loss: 0.0886\n",
      "2021-04-20 10:42:08,019 - mmseg - INFO - Iter [2440/5000]\tlr: 5.522e-03, eta: 0:53:37, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0697, decode.acc_seg: 81.7732, aux.loss_seg: 0.0308, aux.acc_seg: 81.5521, loss: 0.1005\n",
      "2021-04-20 10:42:14,284 - mmseg - INFO - Iter [2450/5000]\tlr: 5.503e-03, eta: 0:53:18, time: 0.626, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0716, decode.acc_seg: 83.3824, aux.loss_seg: 0.0318, aux.acc_seg: 83.1456, loss: 0.1033\n",
      "2021-04-20 10:42:20,566 - mmseg - INFO - Iter [2460/5000]\tlr: 5.484e-03, eta: 0:52:59, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0678, decode.acc_seg: 82.9024, aux.loss_seg: 0.0301, aux.acc_seg: 82.6893, loss: 0.0979\n",
      "2021-04-20 10:42:26,856 - mmseg - INFO - Iter [2470/5000]\tlr: 5.464e-03, eta: 0:52:40, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0761, decode.acc_seg: 83.2109, aux.loss_seg: 0.0338, aux.acc_seg: 82.9494, loss: 0.1100\n",
      "2021-04-20 10:42:33,155 - mmseg - INFO - Iter [2480/5000]\tlr: 5.445e-03, eta: 0:52:21, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0714, decode.acc_seg: 81.6329, aux.loss_seg: 0.0312, aux.acc_seg: 81.4264, loss: 0.1026\n",
      "2021-04-20 10:42:39,452 - mmseg - INFO - Iter [2490/5000]\tlr: 5.426e-03, eta: 0:52:03, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0637, decode.acc_seg: 81.6766, aux.loss_seg: 0.0281, aux.acc_seg: 81.4909, loss: 0.0918\n",
      "2021-04-20 10:42:45,763 - mmseg - INFO - Iter [2500/5000]\tlr: 5.407e-03, eta: 0:51:44, time: 0.631, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0671, decode.acc_seg: 82.6598, aux.loss_seg: 0.0293, aux.acc_seg: 82.4844, loss: 0.0965\n",
      "2021-04-20 10:42:52,070 - mmseg - INFO - Iter [2510/5000]\tlr: 5.388e-03, eta: 0:51:26, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0676, decode.acc_seg: 82.3687, aux.loss_seg: 0.0299, aux.acc_seg: 82.1503, loss: 0.0975\n",
      "2021-04-20 10:42:58,377 - mmseg - INFO - Iter [2520/5000]\tlr: 5.369e-03, eta: 0:51:07, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0658, decode.acc_seg: 79.4311, aux.loss_seg: 0.0292, aux.acc_seg: 79.1905, loss: 0.0950\n",
      "2021-04-20 10:43:04,686 - mmseg - INFO - Iter [2530/5000]\tlr: 5.350e-03, eta: 0:50:49, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0681, decode.acc_seg: 81.0396, aux.loss_seg: 0.0300, aux.acc_seg: 80.7994, loss: 0.0981\n",
      "2021-04-20 10:43:10,994 - mmseg - INFO - Iter [2540/5000]\tlr: 5.331e-03, eta: 0:50:31, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0618, decode.acc_seg: 80.9791, aux.loss_seg: 0.0271, aux.acc_seg: 80.7946, loss: 0.0889\n",
      "2021-04-20 10:43:17,305 - mmseg - INFO - Iter [2550/5000]\tlr: 5.312e-03, eta: 0:50:13, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0646, decode.acc_seg: 81.9278, aux.loss_seg: 0.0285, aux.acc_seg: 81.7506, loss: 0.0931\n",
      "2021-04-20 10:43:23,612 - mmseg - INFO - Iter [2560/5000]\tlr: 5.292e-03, eta: 0:49:55, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0638, decode.acc_seg: 80.1679, aux.loss_seg: 0.0279, aux.acc_seg: 79.9986, loss: 0.0917\n",
      "2021-04-20 10:43:29,917 - mmseg - INFO - Iter [2570/5000]\tlr: 5.273e-03, eta: 0:49:37, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0680, decode.acc_seg: 82.4322, aux.loss_seg: 0.0301, aux.acc_seg: 82.2201, loss: 0.0981\n",
      "2021-04-20 10:43:36,224 - mmseg - INFO - Iter [2580/5000]\tlr: 5.254e-03, eta: 0:49:19, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0688, decode.acc_seg: 82.6168, aux.loss_seg: 0.0301, aux.acc_seg: 82.4306, loss: 0.0990\n",
      "2021-04-20 10:43:42,543 - mmseg - INFO - Iter [2590/5000]\tlr: 5.235e-03, eta: 0:49:01, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0632, decode.acc_seg: 80.5642, aux.loss_seg: 0.0280, aux.acc_seg: 80.3546, loss: 0.0912\n",
      "2021-04-20 10:43:48,845 - mmseg - INFO - Saving checkpoint at 2600 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.5 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:46:02,749 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:46:02,750 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.24 | 98.45 |\n",
      "| cell       | 95.17 | 97.81 |\n",
      "| cilia      | 78.45 | 87.84 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:46:02,751 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:46:02,751 - mmseg - INFO - \n",
      "+--------+-------+------+-------+\n",
      "| Scope  | mIoU  | mAcc | aAcc  |\n",
      "+--------+-------+------+-------+\n",
      "| global | 90.29 | 94.7 | 97.42 |\n",
      "+--------+-------+------+-------+\n",
      "2021-04-20 10:46:02,758 - mmseg - INFO - Iter(val) [2600]\tmIoU: 0.9029, mAcc: 0.9470, aAcc: 0.9742, IoU.background: 0.9724, IoU.cell: 0.9517, IoU.cilia: 0.7845, Acc.background: 0.9845, Acc.cell: 0.9781, Acc.cilia: 0.8784\n",
      "2021-04-20 10:46:08,987 - mmseg - INFO - Iter [2610/5000]\tlr: 5.197e-03, eta: 0:50:22, time: 13.963, data_time: 13.344, memory: 9591, decode.loss_seg: 0.0625, decode.acc_seg: 82.5392, aux.loss_seg: 0.0278, aux.acc_seg: 82.3288, loss: 0.0903\n",
      "2021-04-20 10:46:15,210 - mmseg - INFO - Iter [2620/5000]\tlr: 5.177e-03, eta: 0:50:04, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0613, decode.acc_seg: 79.3529, aux.loss_seg: 0.0270, aux.acc_seg: 79.1983, loss: 0.0883\n",
      "2021-04-20 10:46:21,450 - mmseg - INFO - Iter [2630/5000]\tlr: 5.158e-03, eta: 0:49:45, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0677, decode.acc_seg: 81.6066, aux.loss_seg: 0.0295, aux.acc_seg: 81.4489, loss: 0.0972\n",
      "2021-04-20 10:46:27,697 - mmseg - INFO - Iter [2640/5000]\tlr: 5.139e-03, eta: 0:49:27, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0632, decode.acc_seg: 85.0750, aux.loss_seg: 0.0281, aux.acc_seg: 84.8560, loss: 0.0913\n",
      "2021-04-20 10:46:33,963 - mmseg - INFO - Iter [2650/5000]\tlr: 5.120e-03, eta: 0:49:09, time: 0.627, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0721, decode.acc_seg: 80.3777, aux.loss_seg: 0.0319, aux.acc_seg: 80.1371, loss: 0.1040\n",
      "2021-04-20 10:46:40,246 - mmseg - INFO - Iter [2660/5000]\tlr: 5.101e-03, eta: 0:48:51, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0600, decode.acc_seg: 82.4991, aux.loss_seg: 0.0263, aux.acc_seg: 82.3171, loss: 0.0864\n",
      "2021-04-20 10:46:46,545 - mmseg - INFO - Iter [2670/5000]\tlr: 5.081e-03, eta: 0:48:33, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0622, decode.acc_seg: 83.6636, aux.loss_seg: 0.0278, aux.acc_seg: 83.4455, loss: 0.0900\n",
      "2021-04-20 10:46:52,841 - mmseg - INFO - Iter [2680/5000]\tlr: 5.062e-03, eta: 0:48:15, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0710, decode.acc_seg: 85.4625, aux.loss_seg: 0.0311, aux.acc_seg: 85.2571, loss: 0.1021\n",
      "2021-04-20 10:46:59,142 - mmseg - INFO - Iter [2690/5000]\tlr: 5.043e-03, eta: 0:47:57, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0638, decode.acc_seg: 83.0492, aux.loss_seg: 0.0280, aux.acc_seg: 82.8393, loss: 0.0918\n",
      "2021-04-20 10:47:05,445 - mmseg - INFO - Iter [2700/5000]\tlr: 5.024e-03, eta: 0:47:40, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0604, decode.acc_seg: 82.5213, aux.loss_seg: 0.0267, aux.acc_seg: 82.3524, loss: 0.0871\n",
      "2021-04-20 10:47:11,753 - mmseg - INFO - Iter [2710/5000]\tlr: 5.004e-03, eta: 0:47:22, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0690, decode.acc_seg: 82.1544, aux.loss_seg: 0.0299, aux.acc_seg: 82.0087, loss: 0.0989\n",
      "2021-04-20 10:47:18,063 - mmseg - INFO - Iter [2720/5000]\tlr: 4.985e-03, eta: 0:47:04, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0626, decode.acc_seg: 82.9657, aux.loss_seg: 0.0278, aux.acc_seg: 82.7653, loss: 0.0904\n",
      "2021-04-20 10:47:24,384 - mmseg - INFO - Iter [2730/5000]\tlr: 4.966e-03, eta: 0:46:47, time: 0.632, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0663, decode.acc_seg: 79.4105, aux.loss_seg: 0.0294, aux.acc_seg: 79.1871, loss: 0.0957\n",
      "2021-04-20 10:47:30,696 - mmseg - INFO - Iter [2740/5000]\tlr: 4.947e-03, eta: 0:46:30, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0654, decode.acc_seg: 83.5412, aux.loss_seg: 0.0288, aux.acc_seg: 83.3720, loss: 0.0942\n",
      "2021-04-20 10:47:37,015 - mmseg - INFO - Iter [2750/5000]\tlr: 4.927e-03, eta: 0:46:12, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0560, decode.acc_seg: 79.1198, aux.loss_seg: 0.0248, aux.acc_seg: 78.9306, loss: 0.0808\n",
      "2021-04-20 10:47:43,323 - mmseg - INFO - Iter [2760/5000]\tlr: 4.908e-03, eta: 0:45:55, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0637, decode.acc_seg: 83.3213, aux.loss_seg: 0.0281, aux.acc_seg: 83.1492, loss: 0.0918\n",
      "2021-04-20 10:47:49,630 - mmseg - INFO - Iter [2770/5000]\tlr: 4.889e-03, eta: 0:45:38, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0616, decode.acc_seg: 81.5661, aux.loss_seg: 0.0271, aux.acc_seg: 81.3766, loss: 0.0887\n",
      "2021-04-20 10:47:55,936 - mmseg - INFO - Iter [2780/5000]\tlr: 4.869e-03, eta: 0:45:21, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0625, decode.acc_seg: 81.6914, aux.loss_seg: 0.0276, aux.acc_seg: 81.5127, loss: 0.0901\n",
      "2021-04-20 10:48:02,252 - mmseg - INFO - Iter [2790/5000]\tlr: 4.850e-03, eta: 0:45:04, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0598, decode.acc_seg: 81.4185, aux.loss_seg: 0.0262, aux.acc_seg: 81.2549, loss: 0.0860\n",
      "2021-04-20 10:48:08,553 - mmseg - INFO - Saving checkpoint at 2800 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.4 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:50:22,727 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:50:22,728 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.2  | 98.47 |\n",
      "| cell       | 95.29 | 97.55 |\n",
      "| cilia      | 78.97 | 89.19 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:50:22,728 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:50:22,729 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 90.48 | 95.07 | 97.45 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 10:50:22,735 - mmseg - INFO - Iter(val) [2800]\tmIoU: 0.9048, mAcc: 0.9507, aAcc: 0.9745, IoU.background: 0.9720, IoU.cell: 0.9529, IoU.cilia: 0.7897, Acc.background: 0.9847, Acc.cell: 0.9755, Acc.cilia: 0.8919\n",
      "2021-04-20 10:50:28,995 - mmseg - INFO - Iter [2810/5000]\tlr: 4.811e-03, eta: 0:46:09, time: 13.985, data_time: 13.363, memory: 9591, decode.loss_seg: 0.0663, decode.acc_seg: 81.2417, aux.loss_seg: 0.0294, aux.acc_seg: 81.0670, loss: 0.0957\n",
      "2021-04-20 10:50:35,210 - mmseg - INFO - Iter [2820/5000]\tlr: 4.792e-03, eta: 0:45:52, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0629, decode.acc_seg: 79.3805, aux.loss_seg: 0.0277, aux.acc_seg: 79.1726, loss: 0.0906\n",
      "2021-04-20 10:50:41,443 - mmseg - INFO - Iter [2830/5000]\tlr: 4.773e-03, eta: 0:45:34, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0617, decode.acc_seg: 80.8380, aux.loss_seg: 0.0274, aux.acc_seg: 80.6328, loss: 0.0892\n",
      "2021-04-20 10:50:47,691 - mmseg - INFO - Iter [2840/5000]\tlr: 4.753e-03, eta: 0:45:17, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0594, decode.acc_seg: 81.2166, aux.loss_seg: 0.0261, aux.acc_seg: 81.0484, loss: 0.0855\n",
      "2021-04-20 10:50:53,972 - mmseg - INFO - Iter [2850/5000]\tlr: 4.734e-03, eta: 0:45:00, time: 0.628, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0667, decode.acc_seg: 78.3673, aux.loss_seg: 0.0293, aux.acc_seg: 78.1842, loss: 0.0960\n",
      "2021-04-20 10:51:00,261 - mmseg - INFO - Iter [2860/5000]\tlr: 4.714e-03, eta: 0:44:42, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0652, decode.acc_seg: 81.9162, aux.loss_seg: 0.0290, aux.acc_seg: 81.7058, loss: 0.0942\n",
      "2021-04-20 10:51:06,555 - mmseg - INFO - Iter [2870/5000]\tlr: 4.695e-03, eta: 0:44:25, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0756, decode.acc_seg: 83.0566, aux.loss_seg: 0.0318, aux.acc_seg: 82.8675, loss: 0.1074\n",
      "2021-04-20 10:51:12,850 - mmseg - INFO - Iter [2880/5000]\tlr: 4.676e-03, eta: 0:44:08, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0587, decode.acc_seg: 78.6259, aux.loss_seg: 0.0258, aux.acc_seg: 78.4678, loss: 0.0845\n",
      "2021-04-20 10:51:19,152 - mmseg - INFO - Iter [2890/5000]\tlr: 4.656e-03, eta: 0:43:51, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0647, decode.acc_seg: 84.8712, aux.loss_seg: 0.0282, aux.acc_seg: 84.6957, loss: 0.0930\n",
      "2021-04-20 10:51:25,461 - mmseg - INFO - Iter [2900/5000]\tlr: 4.637e-03, eta: 0:43:34, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0600, decode.acc_seg: 84.1587, aux.loss_seg: 0.0262, aux.acc_seg: 83.9856, loss: 0.0862\n",
      "2021-04-20 10:51:31,773 - mmseg - INFO - Iter [2910/5000]\tlr: 4.617e-03, eta: 0:43:17, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0584, decode.acc_seg: 79.6789, aux.loss_seg: 0.0256, aux.acc_seg: 79.5327, loss: 0.0841\n",
      "2021-04-20 10:51:38,082 - mmseg - INFO - Iter [2920/5000]\tlr: 4.598e-03, eta: 0:43:00, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0647, decode.acc_seg: 81.4042, aux.loss_seg: 0.0285, aux.acc_seg: 81.1967, loss: 0.0932\n",
      "2021-04-20 10:51:44,391 - mmseg - INFO - Iter [2930/5000]\tlr: 4.578e-03, eta: 0:42:44, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0612, decode.acc_seg: 78.1161, aux.loss_seg: 0.0270, aux.acc_seg: 77.9164, loss: 0.0881\n",
      "2021-04-20 10:51:50,705 - mmseg - INFO - Iter [2940/5000]\tlr: 4.559e-03, eta: 0:42:27, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0683, decode.acc_seg: 81.7655, aux.loss_seg: 0.0298, aux.acc_seg: 81.5813, loss: 0.0982\n",
      "2021-04-20 10:51:57,016 - mmseg - INFO - Iter [2950/5000]\tlr: 4.539e-03, eta: 0:42:11, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0609, decode.acc_seg: 80.6782, aux.loss_seg: 0.0264, aux.acc_seg: 80.5367, loss: 0.0873\n",
      "2021-04-20 10:52:03,337 - mmseg - INFO - Iter [2960/5000]\tlr: 4.520e-03, eta: 0:41:54, time: 0.632, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0576, decode.acc_seg: 80.8749, aux.loss_seg: 0.0253, aux.acc_seg: 80.7133, loss: 0.0828\n",
      "2021-04-20 10:52:09,657 - mmseg - INFO - Iter [2970/5000]\tlr: 4.500e-03, eta: 0:41:38, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0585, decode.acc_seg: 85.8520, aux.loss_seg: 0.0261, aux.acc_seg: 85.6367, loss: 0.0846\n",
      "2021-04-20 10:52:15,966 - mmseg - INFO - Iter [2980/5000]\tlr: 4.481e-03, eta: 0:41:21, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0624, decode.acc_seg: 82.3941, aux.loss_seg: 0.0273, aux.acc_seg: 82.2459, loss: 0.0896\n",
      "2021-04-20 10:52:22,275 - mmseg - INFO - Iter [2990/5000]\tlr: 4.461e-03, eta: 0:41:05, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0755, decode.acc_seg: 82.7036, aux.loss_seg: 0.0333, aux.acc_seg: 82.4886, loss: 0.1088\n",
      "2021-04-20 10:52:28,578 - mmseg - INFO - Saving checkpoint at 3000 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.3 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:54:43,082 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:54:43,083 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.33 | 98.55 |\n",
      "| cell       | 95.35 | 97.8  |\n",
      "| cilia      | 79.29 | 88.44 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:54:43,083 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:54:43,084 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 90.66 | 94.93 | 97.52 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 10:54:43,091 - mmseg - INFO - Iter(val) [3000]\tmIoU: 0.9066, mAcc: 0.9493, aAcc: 0.9752, IoU.background: 0.9733, IoU.cell: 0.9535, IoU.cilia: 0.7929, Acc.background: 0.9855, Acc.cell: 0.9780, Acc.cilia: 0.8844\n",
      "2021-04-20 10:54:49,343 - mmseg - INFO - Iter [3010/5000]\tlr: 4.422e-03, eta: 0:41:57, time: 14.019, data_time: 13.398, memory: 9591, decode.loss_seg: 0.0523, decode.acc_seg: 80.2109, aux.loss_seg: 0.0233, aux.acc_seg: 80.0268, loss: 0.0755\n",
      "2021-04-20 10:54:55,567 - mmseg - INFO - Iter [3020/5000]\tlr: 4.403e-03, eta: 0:41:40, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0641, decode.acc_seg: 84.0824, aux.loss_seg: 0.0287, aux.acc_seg: 83.8285, loss: 0.0928\n",
      "2021-04-20 10:55:01,797 - mmseg - INFO - Iter [3030/5000]\tlr: 4.383e-03, eta: 0:41:23, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0565, decode.acc_seg: 80.8510, aux.loss_seg: 0.0248, aux.acc_seg: 80.7118, loss: 0.0813\n",
      "2021-04-20 10:55:08,042 - mmseg - INFO - Iter [3040/5000]\tlr: 4.364e-03, eta: 0:41:06, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0623, decode.acc_seg: 81.9750, aux.loss_seg: 0.0276, aux.acc_seg: 81.7634, loss: 0.0899\n",
      "2021-04-20 10:55:14,301 - mmseg - INFO - Iter [3050/5000]\tlr: 4.344e-03, eta: 0:40:50, time: 0.626, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0582, decode.acc_seg: 81.1952, aux.loss_seg: 0.0255, aux.acc_seg: 81.0434, loss: 0.0837\n",
      "2021-04-20 10:55:20,582 - mmseg - INFO - Iter [3060/5000]\tlr: 4.325e-03, eta: 0:40:33, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0610, decode.acc_seg: 83.9506, aux.loss_seg: 0.0266, aux.acc_seg: 83.8027, loss: 0.0875\n",
      "2021-04-20 10:55:26,871 - mmseg - INFO - Iter [3070/5000]\tlr: 4.305e-03, eta: 0:40:17, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0654, decode.acc_seg: 83.1609, aux.loss_seg: 0.0287, aux.acc_seg: 82.9848, loss: 0.0941\n",
      "2021-04-20 10:55:33,165 - mmseg - INFO - Iter [3080/5000]\tlr: 4.285e-03, eta: 0:40:00, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0611, decode.acc_seg: 79.2062, aux.loss_seg: 0.0277, aux.acc_seg: 78.9655, loss: 0.0888\n",
      "2021-04-20 10:55:39,464 - mmseg - INFO - Iter [3090/5000]\tlr: 4.266e-03, eta: 0:39:44, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0576, decode.acc_seg: 83.6104, aux.loss_seg: 0.0251, aux.acc_seg: 83.4792, loss: 0.0826\n",
      "2021-04-20 10:55:45,778 - mmseg - INFO - Iter [3100/5000]\tlr: 4.246e-03, eta: 0:39:28, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0622, decode.acc_seg: 82.1474, aux.loss_seg: 0.0277, aux.acc_seg: 81.9433, loss: 0.0898\n",
      "2021-04-20 10:55:52,085 - mmseg - INFO - Iter [3110/5000]\tlr: 4.227e-03, eta: 0:39:12, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0605, decode.acc_seg: 81.4999, aux.loss_seg: 0.0265, aux.acc_seg: 81.3503, loss: 0.0870\n",
      "2021-04-20 10:55:58,397 - mmseg - INFO - Iter [3120/5000]\tlr: 4.207e-03, eta: 0:38:55, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0662, decode.acc_seg: 80.2093, aux.loss_seg: 0.0289, aux.acc_seg: 80.0047, loss: 0.0951\n",
      "2021-04-20 10:56:04,705 - mmseg - INFO - Iter [3130/5000]\tlr: 4.187e-03, eta: 0:38:39, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0682, decode.acc_seg: 79.9950, aux.loss_seg: 0.0303, aux.acc_seg: 79.7570, loss: 0.0986\n",
      "2021-04-20 10:56:11,017 - mmseg - INFO - Iter [3140/5000]\tlr: 4.168e-03, eta: 0:38:23, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0670, decode.acc_seg: 79.5623, aux.loss_seg: 0.0307, aux.acc_seg: 79.4132, loss: 0.0977\n",
      "2021-04-20 10:56:17,330 - mmseg - INFO - Iter [3150/5000]\tlr: 4.148e-03, eta: 0:38:07, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0618, decode.acc_seg: 83.6299, aux.loss_seg: 0.0274, aux.acc_seg: 83.4264, loss: 0.0892\n",
      "2021-04-20 10:56:23,633 - mmseg - INFO - Iter [3160/5000]\tlr: 4.128e-03, eta: 0:37:52, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0649, decode.acc_seg: 79.5649, aux.loss_seg: 0.0286, aux.acc_seg: 79.4223, loss: 0.0935\n",
      "2021-04-20 10:56:32,150 - mmseg - INFO - Iter [3170/5000]\tlr: 4.108e-03, eta: 0:37:37, time: 0.852, data_time: 0.223, memory: 9591, decode.loss_seg: 0.0597, decode.acc_seg: 82.2031, aux.loss_seg: 0.0266, aux.acc_seg: 81.9738, loss: 0.0863\n",
      "2021-04-20 10:56:38,453 - mmseg - INFO - Iter [3180/5000]\tlr: 4.089e-03, eta: 0:37:21, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0623, decode.acc_seg: 81.0982, aux.loss_seg: 0.0273, aux.acc_seg: 80.9097, loss: 0.0896\n",
      "2021-04-20 10:56:44,753 - mmseg - INFO - Iter [3190/5000]\tlr: 4.069e-03, eta: 0:37:05, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0603, decode.acc_seg: 82.2810, aux.loss_seg: 0.0267, aux.acc_seg: 82.0987, loss: 0.0870\n",
      "2021-04-20 10:56:51,049 - mmseg - INFO - Saving checkpoint at 3200 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 35.8 task/s, elapsed: 118s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 10:59:07,490 - mmseg - INFO - per class results:\n",
      "2021-04-20 10:59:07,491 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.45 | 98.56 |\n",
      "| cell       | 95.55 | 97.59 |\n",
      "| cilia      | 80.34 | 90.66 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 10:59:07,491 - mmseg - INFO - Summary:\n",
      "2021-04-20 10:59:07,492 - mmseg - INFO - \n",
      "+--------+-------+------+-------+\n",
      "| Scope  | mIoU  | mAcc | aAcc  |\n",
      "+--------+-------+------+-------+\n",
      "| global | 91.11 | 95.6 | 97.63 |\n",
      "+--------+-------+------+-------+\n",
      "2021-04-20 10:59:07,499 - mmseg - INFO - Iter(val) [3200]\tmIoU: 0.9111, mAcc: 0.9560, aAcc: 0.9763, IoU.background: 0.9745, IoU.cell: 0.9555, IoU.cilia: 0.8034, Acc.background: 0.9856, Acc.cell: 0.9759, Acc.cilia: 0.9066\n",
      "2021-04-20 10:59:13,744 - mmseg - INFO - Iter [3210/5000]\tlr: 4.030e-03, eta: 0:37:46, time: 14.212, data_time: 13.591, memory: 9591, decode.loss_seg: 0.0616, decode.acc_seg: 83.3149, aux.loss_seg: 0.0276, aux.acc_seg: 83.0735, loss: 0.0892\n",
      "2021-04-20 10:59:19,957 - mmseg - INFO - Iter [3220/5000]\tlr: 4.010e-03, eta: 0:37:30, time: 0.621, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0619, decode.acc_seg: 81.8301, aux.loss_seg: 0.0274, aux.acc_seg: 81.6186, loss: 0.0893\n",
      "2021-04-20 10:59:26,189 - mmseg - INFO - Iter [3230/5000]\tlr: 3.990e-03, eta: 0:37:14, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0529, decode.acc_seg: 84.5228, aux.loss_seg: 0.0231, aux.acc_seg: 84.3941, loss: 0.0760\n",
      "2021-04-20 10:59:32,436 - mmseg - INFO - Iter [3240/5000]\tlr: 3.970e-03, eta: 0:36:58, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0545, decode.acc_seg: 80.4324, aux.loss_seg: 0.0238, aux.acc_seg: 80.2649, loss: 0.0784\n",
      "2021-04-20 10:59:38,688 - mmseg - INFO - Iter [3250/5000]\tlr: 3.951e-03, eta: 0:36:42, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0582, decode.acc_seg: 85.6341, aux.loss_seg: 0.0256, aux.acc_seg: 85.4465, loss: 0.0837\n",
      "2021-04-20 10:59:44,972 - mmseg - INFO - Iter [3260/5000]\tlr: 3.931e-03, eta: 0:36:26, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0643, decode.acc_seg: 85.5210, aux.loss_seg: 0.0287, aux.acc_seg: 85.2614, loss: 0.0930\n",
      "2021-04-20 10:59:51,273 - mmseg - INFO - Iter [3270/5000]\tlr: 3.911e-03, eta: 0:36:10, time: 0.630, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0586, decode.acc_seg: 80.1416, aux.loss_seg: 0.0260, aux.acc_seg: 79.9606, loss: 0.0846\n",
      "2021-04-20 10:59:57,564 - mmseg - INFO - Iter [3280/5000]\tlr: 3.891e-03, eta: 0:35:54, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0573, decode.acc_seg: 82.2629, aux.loss_seg: 0.0257, aux.acc_seg: 82.0752, loss: 0.0831\n",
      "2021-04-20 11:00:03,866 - mmseg - INFO - Iter [3290/5000]\tlr: 3.871e-03, eta: 0:35:38, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0643, decode.acc_seg: 79.2018, aux.loss_seg: 0.0282, aux.acc_seg: 79.0078, loss: 0.0926\n",
      "2021-04-20 11:00:10,165 - mmseg - INFO - Iter [3300/5000]\tlr: 3.851e-03, eta: 0:35:23, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0586, decode.acc_seg: 82.7862, aux.loss_seg: 0.0259, aux.acc_seg: 82.6057, loss: 0.0845\n",
      "2021-04-20 11:00:16,463 - mmseg - INFO - Iter [3310/5000]\tlr: 3.832e-03, eta: 0:35:07, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0597, decode.acc_seg: 82.4766, aux.loss_seg: 0.0262, aux.acc_seg: 82.3136, loss: 0.0859\n",
      "2021-04-20 11:00:22,767 - mmseg - INFO - Iter [3320/5000]\tlr: 3.812e-03, eta: 0:34:52, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0541, decode.acc_seg: 84.2688, aux.loss_seg: 0.0237, aux.acc_seg: 84.1414, loss: 0.0778\n",
      "2021-04-20 11:00:29,072 - mmseg - INFO - Iter [3330/5000]\tlr: 3.792e-03, eta: 0:34:36, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0631, decode.acc_seg: 84.0072, aux.loss_seg: 0.0274, aux.acc_seg: 83.8412, loss: 0.0905\n",
      "2021-04-20 11:00:35,388 - mmseg - INFO - Iter [3340/5000]\tlr: 3.772e-03, eta: 0:34:20, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0654, decode.acc_seg: 83.7198, aux.loss_seg: 0.0291, aux.acc_seg: 83.4891, loss: 0.0946\n",
      "2021-04-20 11:00:41,705 - mmseg - INFO - Iter [3350/5000]\tlr: 3.752e-03, eta: 0:34:05, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0577, decode.acc_seg: 81.3448, aux.loss_seg: 0.0259, aux.acc_seg: 81.1150, loss: 0.0836\n",
      "2021-04-20 11:00:48,007 - mmseg - INFO - Iter [3360/5000]\tlr: 3.732e-03, eta: 0:33:50, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0551, decode.acc_seg: 83.6345, aux.loss_seg: 0.0241, aux.acc_seg: 83.5091, loss: 0.0792\n",
      "2021-04-20 11:00:54,315 - mmseg - INFO - Iter [3370/5000]\tlr: 3.712e-03, eta: 0:33:34, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0587, decode.acc_seg: 82.2352, aux.loss_seg: 0.0256, aux.acc_seg: 82.0778, loss: 0.0843\n",
      "2021-04-20 11:01:00,623 - mmseg - INFO - Iter [3380/5000]\tlr: 3.692e-03, eta: 0:33:19, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0568, decode.acc_seg: 78.5357, aux.loss_seg: 0.0252, aux.acc_seg: 78.3484, loss: 0.0820\n",
      "2021-04-20 11:01:06,930 - mmseg - INFO - Iter [3390/5000]\tlr: 3.672e-03, eta: 0:33:04, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0521, decode.acc_seg: 80.8865, aux.loss_seg: 0.0228, aux.acc_seg: 80.7424, loss: 0.0749\n",
      "2021-04-20 11:01:13,233 - mmseg - INFO - Saving checkpoint at 3400 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.3 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 11:03:27,599 - mmseg - INFO - per class results:\n",
      "2021-04-20 11:03:27,600 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.5  | 98.65 |\n",
      "| cell       | 95.68 | 97.89 |\n",
      "| cilia      | 80.79 | 89.56 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 11:03:27,600 - mmseg - INFO - Summary:\n",
      "2021-04-20 11:03:27,600 - mmseg - INFO - \n",
      "+--------+-------+-------+------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc |\n",
      "+--------+-------+-------+------+\n",
      "| global | 91.32 | 95.37 | 97.7 |\n",
      "+--------+-------+-------+------+\n",
      "2021-04-20 11:03:27,607 - mmseg - INFO - Iter(val) [3400]\tmIoU: 0.9132, mAcc: 0.9537, aAcc: 0.9770, IoU.background: 0.9750, IoU.cell: 0.9568, IoU.cilia: 0.8079, Acc.background: 0.9865, Acc.cell: 0.9789, Acc.cilia: 0.8956\n",
      "2021-04-20 11:03:33,863 - mmseg - INFO - Iter [3410/5000]\tlr: 3.632e-03, eta: 0:33:33, time: 14.030, data_time: 13.408, memory: 9591, decode.loss_seg: 0.0595, decode.acc_seg: 84.2388, aux.loss_seg: 0.0265, aux.acc_seg: 84.0387, loss: 0.0859\n",
      "2021-04-20 11:03:40,082 - mmseg - INFO - Iter [3420/5000]\tlr: 3.612e-03, eta: 0:33:18, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0562, decode.acc_seg: 82.0704, aux.loss_seg: 0.0246, aux.acc_seg: 81.9422, loss: 0.0808\n",
      "2021-04-20 11:03:46,319 - mmseg - INFO - Iter [3430/5000]\tlr: 3.592e-03, eta: 0:33:02, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0618, decode.acc_seg: 80.9406, aux.loss_seg: 0.0274, aux.acc_seg: 80.7419, loss: 0.0893\n",
      "2021-04-20 11:03:52,565 - mmseg - INFO - Iter [3440/5000]\tlr: 3.572e-03, eta: 0:32:46, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0568, decode.acc_seg: 83.5360, aux.loss_seg: 0.0253, aux.acc_seg: 83.3844, loss: 0.0822\n",
      "2021-04-20 11:03:58,825 - mmseg - INFO - Iter [3450/5000]\tlr: 3.552e-03, eta: 0:32:31, time: 0.626, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0529, decode.acc_seg: 84.6508, aux.loss_seg: 0.0231, aux.acc_seg: 84.5245, loss: 0.0761\n",
      "2021-04-20 11:04:05,106 - mmseg - INFO - Iter [3460/5000]\tlr: 3.532e-03, eta: 0:32:16, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0575, decode.acc_seg: 85.2269, aux.loss_seg: 0.0254, aux.acc_seg: 85.0407, loss: 0.0830\n",
      "2021-04-20 11:04:11,391 - mmseg - INFO - Iter [3470/5000]\tlr: 3.512e-03, eta: 0:32:00, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0573, decode.acc_seg: 83.1421, aux.loss_seg: 0.0253, aux.acc_seg: 82.9630, loss: 0.0826\n",
      "2021-04-20 11:04:17,691 - mmseg - INFO - Iter [3480/5000]\tlr: 3.492e-03, eta: 0:31:45, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0525, decode.acc_seg: 83.0349, aux.loss_seg: 0.0231, aux.acc_seg: 82.8848, loss: 0.0756\n",
      "2021-04-20 11:04:23,986 - mmseg - INFO - Iter [3490/5000]\tlr: 3.472e-03, eta: 0:31:30, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0500, decode.acc_seg: 79.7138, aux.loss_seg: 0.0221, aux.acc_seg: 79.5520, loss: 0.0721\n",
      "2021-04-20 11:04:30,288 - mmseg - INFO - Iter [3500/5000]\tlr: 3.452e-03, eta: 0:31:15, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0631, decode.acc_seg: 82.8493, aux.loss_seg: 0.0276, aux.acc_seg: 82.7091, loss: 0.0907\n",
      "2021-04-20 11:04:36,598 - mmseg - INFO - Iter [3510/5000]\tlr: 3.432e-03, eta: 0:30:59, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0567, decode.acc_seg: 83.1354, aux.loss_seg: 0.0253, aux.acc_seg: 82.9319, loss: 0.0820\n",
      "2021-04-20 11:04:42,904 - mmseg - INFO - Iter [3520/5000]\tlr: 3.412e-03, eta: 0:30:44, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0551, decode.acc_seg: 84.2951, aux.loss_seg: 0.0242, aux.acc_seg: 84.1196, loss: 0.0794\n",
      "2021-04-20 11:04:49,221 - mmseg - INFO - Iter [3530/5000]\tlr: 3.392e-03, eta: 0:30:29, time: 0.632, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0586, decode.acc_seg: 85.1735, aux.loss_seg: 0.0257, aux.acc_seg: 85.0413, loss: 0.0843\n",
      "2021-04-20 11:04:55,529 - mmseg - INFO - Iter [3540/5000]\tlr: 3.372e-03, eta: 0:30:14, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0503, decode.acc_seg: 82.6061, aux.loss_seg: 0.0224, aux.acc_seg: 82.4377, loss: 0.0727\n",
      "2021-04-20 11:05:01,837 - mmseg - INFO - Iter [3550/5000]\tlr: 3.351e-03, eta: 0:29:59, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0560, decode.acc_seg: 81.3021, aux.loss_seg: 0.0246, aux.acc_seg: 81.1384, loss: 0.0806\n",
      "2021-04-20 11:05:08,144 - mmseg - INFO - Iter [3560/5000]\tlr: 3.331e-03, eta: 0:29:45, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0552, decode.acc_seg: 84.4713, aux.loss_seg: 0.0245, aux.acc_seg: 84.2759, loss: 0.0797\n",
      "2021-04-20 11:05:14,463 - mmseg - INFO - Iter [3570/5000]\tlr: 3.311e-03, eta: 0:29:30, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0570, decode.acc_seg: 83.4986, aux.loss_seg: 0.0250, aux.acc_seg: 83.3416, loss: 0.0820\n",
      "2021-04-20 11:05:20,772 - mmseg - INFO - Iter [3580/5000]\tlr: 3.291e-03, eta: 0:29:15, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0602, decode.acc_seg: 81.8769, aux.loss_seg: 0.0267, aux.acc_seg: 81.6842, loss: 0.0869\n",
      "2021-04-20 11:05:27,080 - mmseg - INFO - Iter [3590/5000]\tlr: 3.271e-03, eta: 0:29:00, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0571, decode.acc_seg: 80.3892, aux.loss_seg: 0.0255, aux.acc_seg: 80.1869, loss: 0.0827\n",
      "2021-04-20 11:05:33,384 - mmseg - INFO - Saving checkpoint at 3600 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.4 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 11:07:47,425 - mmseg - INFO - per class results:\n",
      "2021-04-20 11:07:47,426 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.5  | 98.48 |\n",
      "| cell       | 95.8  | 97.8  |\n",
      "| cilia      | 81.16 | 91.62 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 11:07:47,426 - mmseg - INFO - Summary:\n",
      "2021-04-20 11:07:47,426 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 91.49 | 95.97 | 97.72 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 11:07:47,434 - mmseg - INFO - Iter(val) [3600]\tmIoU: 0.9149, mAcc: 0.9597, aAcc: 0.9772, IoU.background: 0.9750, IoU.cell: 0.9580, IoU.cilia: 0.8116, Acc.background: 0.9848, Acc.cell: 0.9780, Acc.cilia: 0.9162\n",
      "2021-04-20 11:07:53,677 - mmseg - INFO - Iter [3610/5000]\tlr: 3.230e-03, eta: 0:29:20, time: 13.994, data_time: 13.374, memory: 9591, decode.loss_seg: 0.0545, decode.acc_seg: 84.5924, aux.loss_seg: 0.0240, aux.acc_seg: 84.4368, loss: 0.0784\n",
      "2021-04-20 11:07:59,891 - mmseg - INFO - Iter [3620/5000]\tlr: 3.210e-03, eta: 0:29:05, time: 0.621, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0567, decode.acc_seg: 80.5582, aux.loss_seg: 0.0247, aux.acc_seg: 80.4143, loss: 0.0813\n",
      "2021-04-20 11:08:06,125 - mmseg - INFO - Iter [3630/5000]\tlr: 3.190e-03, eta: 0:28:50, time: 0.623, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0551, decode.acc_seg: 80.4022, aux.loss_seg: 0.0246, aux.acc_seg: 80.1969, loss: 0.0797\n",
      "2021-04-20 11:08:12,372 - mmseg - INFO - Iter [3640/5000]\tlr: 3.169e-03, eta: 0:28:35, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0587, decode.acc_seg: 79.6414, aux.loss_seg: 0.0259, aux.acc_seg: 79.4544, loss: 0.0846\n",
      "2021-04-20 11:08:18,640 - mmseg - INFO - Iter [3650/5000]\tlr: 3.149e-03, eta: 0:28:20, time: 0.627, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0599, decode.acc_seg: 84.5400, aux.loss_seg: 0.0267, aux.acc_seg: 84.3603, loss: 0.0866\n",
      "2021-04-20 11:08:24,921 - mmseg - INFO - Iter [3660/5000]\tlr: 3.129e-03, eta: 0:28:05, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0633, decode.acc_seg: 81.7593, aux.loss_seg: 0.0279, aux.acc_seg: 81.5480, loss: 0.0913\n",
      "2021-04-20 11:08:31,210 - mmseg - INFO - Iter [3670/5000]\tlr: 3.108e-03, eta: 0:27:50, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0548, decode.acc_seg: 81.8458, aux.loss_seg: 0.0240, aux.acc_seg: 81.6888, loss: 0.0788\n",
      "2021-04-20 11:08:37,507 - mmseg - INFO - Iter [3680/5000]\tlr: 3.088e-03, eta: 0:27:35, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0510, decode.acc_seg: 83.5356, aux.loss_seg: 0.0225, aux.acc_seg: 83.3783, loss: 0.0734\n",
      "2021-04-20 11:08:43,806 - mmseg - INFO - Iter [3690/5000]\tlr: 3.068e-03, eta: 0:27:20, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0617, decode.acc_seg: 82.5493, aux.loss_seg: 0.0266, aux.acc_seg: 82.4142, loss: 0.0884\n",
      "2021-04-20 11:08:50,107 - mmseg - INFO - Iter [3700/5000]\tlr: 3.047e-03, eta: 0:27:06, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0548, decode.acc_seg: 81.7175, aux.loss_seg: 0.0241, aux.acc_seg: 81.5542, loss: 0.0789\n",
      "2021-04-20 11:08:56,412 - mmseg - INFO - Iter [3710/5000]\tlr: 3.027e-03, eta: 0:26:51, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0528, decode.acc_seg: 81.4966, aux.loss_seg: 0.0228, aux.acc_seg: 81.3575, loss: 0.0756\n",
      "2021-04-20 11:09:02,720 - mmseg - INFO - Iter [3720/5000]\tlr: 3.006e-03, eta: 0:26:36, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0564, decode.acc_seg: 82.4896, aux.loss_seg: 0.0249, aux.acc_seg: 82.3109, loss: 0.0814\n",
      "2021-04-20 11:09:09,038 - mmseg - INFO - Iter [3730/5000]\tlr: 2.986e-03, eta: 0:26:22, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0512, decode.acc_seg: 79.3274, aux.loss_seg: 0.0226, aux.acc_seg: 79.1593, loss: 0.0738\n",
      "2021-04-20 11:09:15,350 - mmseg - INFO - Iter [3740/5000]\tlr: 2.966e-03, eta: 0:26:07, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0542, decode.acc_seg: 84.9423, aux.loss_seg: 0.0237, aux.acc_seg: 84.7838, loss: 0.0779\n",
      "2021-04-20 11:09:21,659 - mmseg - INFO - Iter [3750/5000]\tlr: 2.945e-03, eta: 0:25:53, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0540, decode.acc_seg: 83.3881, aux.loss_seg: 0.0239, aux.acc_seg: 83.2136, loss: 0.0780\n",
      "2021-04-20 11:09:27,968 - mmseg - INFO - Iter [3760/5000]\tlr: 2.925e-03, eta: 0:25:38, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0554, decode.acc_seg: 82.2339, aux.loss_seg: 0.0243, aux.acc_seg: 82.0837, loss: 0.0797\n",
      "2021-04-20 11:09:34,277 - mmseg - INFO - Iter [3770/5000]\tlr: 2.904e-03, eta: 0:25:24, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0491, decode.acc_seg: 83.9748, aux.loss_seg: 0.0218, aux.acc_seg: 83.8251, loss: 0.0709\n",
      "2021-04-20 11:09:40,590 - mmseg - INFO - Iter [3780/5000]\tlr: 2.884e-03, eta: 0:25:10, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0543, decode.acc_seg: 84.6984, aux.loss_seg: 0.0240, aux.acc_seg: 84.5165, loss: 0.0783\n",
      "2021-04-20 11:09:46,899 - mmseg - INFO - Iter [3790/5000]\tlr: 2.863e-03, eta: 0:24:55, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0572, decode.acc_seg: 82.0689, aux.loss_seg: 0.0250, aux.acc_seg: 81.8913, loss: 0.0822\n",
      "2021-04-20 11:09:53,204 - mmseg - INFO - Saving checkpoint at 3800 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.3 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 11:12:07,701 - mmseg - INFO - per class results:\n",
      "2021-04-20 11:12:07,702 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.52 | 98.58 |\n",
      "| cell       | 95.9  | 97.91 |\n",
      "| cilia      | 81.28 | 90.81 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 11:12:07,702 - mmseg - INFO - Summary:\n",
      "2021-04-20 11:12:07,703 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 91.57 | 95.77 | 97.76 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 11:12:07,709 - mmseg - INFO - Iter(val) [3800]\tmIoU: 0.9157, mAcc: 0.9577, aAcc: 0.9776, IoU.background: 0.9752, IoU.cell: 0.9590, IoU.cilia: 0.8128, Acc.background: 0.9858, Acc.cell: 0.9791, Acc.cilia: 0.9081\n",
      "2021-04-20 11:12:13,966 - mmseg - INFO - Iter [3810/5000]\tlr: 2.822e-03, eta: 0:25:07, time: 14.039, data_time: 13.417, memory: 9591, decode.loss_seg: 0.0552, decode.acc_seg: 82.5868, aux.loss_seg: 0.0239, aux.acc_seg: 82.4454, loss: 0.0791\n",
      "2021-04-20 11:12:20,184 - mmseg - INFO - Iter [3820/5000]\tlr: 2.801e-03, eta: 0:24:52, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0544, decode.acc_seg: 80.6992, aux.loss_seg: 0.0239, aux.acc_seg: 80.5322, loss: 0.0782\n",
      "2021-04-20 11:12:26,421 - mmseg - INFO - Iter [3830/5000]\tlr: 2.781e-03, eta: 0:24:37, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0531, decode.acc_seg: 83.7449, aux.loss_seg: 0.0237, aux.acc_seg: 83.5676, loss: 0.0768\n",
      "2021-04-20 11:12:32,668 - mmseg - INFO - Iter [3840/5000]\tlr: 2.760e-03, eta: 0:24:23, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0482, decode.acc_seg: 83.1229, aux.loss_seg: 0.0216, aux.acc_seg: 82.9258, loss: 0.0698\n",
      "2021-04-20 11:12:38,940 - mmseg - INFO - Iter [3850/5000]\tlr: 2.740e-03, eta: 0:24:08, time: 0.627, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0594, decode.acc_seg: 82.6335, aux.loss_seg: 0.0265, aux.acc_seg: 82.4217, loss: 0.0860\n",
      "2021-04-20 11:12:45,232 - mmseg - INFO - Iter [3860/5000]\tlr: 2.719e-03, eta: 0:23:54, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0558, decode.acc_seg: 81.6889, aux.loss_seg: 0.0243, aux.acc_seg: 81.5464, loss: 0.0801\n",
      "2021-04-20 11:12:51,536 - mmseg - INFO - Iter [3870/5000]\tlr: 2.698e-03, eta: 0:23:39, time: 0.630, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0539, decode.acc_seg: 81.8225, aux.loss_seg: 0.0238, aux.acc_seg: 81.6424, loss: 0.0777\n",
      "2021-04-20 11:12:57,846 - mmseg - INFO - Iter [3880/5000]\tlr: 2.678e-03, eta: 0:23:25, time: 0.631, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0604, decode.acc_seg: 79.7764, aux.loss_seg: 0.0264, aux.acc_seg: 79.5928, loss: 0.0868\n",
      "2021-04-20 11:13:04,151 - mmseg - INFO - Iter [3890/5000]\tlr: 2.657e-03, eta: 0:23:11, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0582, decode.acc_seg: 81.0453, aux.loss_seg: 0.0252, aux.acc_seg: 80.9060, loss: 0.0834\n",
      "2021-04-20 11:13:10,455 - mmseg - INFO - Iter [3900/5000]\tlr: 2.636e-03, eta: 0:22:56, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0572, decode.acc_seg: 84.4426, aux.loss_seg: 0.0252, aux.acc_seg: 84.2715, loss: 0.0824\n",
      "2021-04-20 11:13:16,763 - mmseg - INFO - Iter [3910/5000]\tlr: 2.615e-03, eta: 0:22:42, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0526, decode.acc_seg: 82.9449, aux.loss_seg: 0.0230, aux.acc_seg: 82.8010, loss: 0.0757\n",
      "2021-04-20 11:13:23,072 - mmseg - INFO - Iter [3920/5000]\tlr: 2.595e-03, eta: 0:22:28, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0496, decode.acc_seg: 84.0723, aux.loss_seg: 0.0221, aux.acc_seg: 83.9181, loss: 0.0717\n",
      "2021-04-20 11:13:29,388 - mmseg - INFO - Iter [3930/5000]\tlr: 2.574e-03, eta: 0:22:14, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0529, decode.acc_seg: 80.5856, aux.loss_seg: 0.0231, aux.acc_seg: 80.4234, loss: 0.0760\n",
      "2021-04-20 11:13:35,699 - mmseg - INFO - Iter [3940/5000]\tlr: 2.553e-03, eta: 0:22:00, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0506, decode.acc_seg: 82.8634, aux.loss_seg: 0.0224, aux.acc_seg: 82.7153, loss: 0.0730\n",
      "2021-04-20 11:13:42,007 - mmseg - INFO - Iter [3950/5000]\tlr: 2.532e-03, eta: 0:21:46, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0494, decode.acc_seg: 83.1905, aux.loss_seg: 0.0224, aux.acc_seg: 83.0012, loss: 0.0718\n",
      "2021-04-20 11:13:48,317 - mmseg - INFO - Iter [3960/5000]\tlr: 2.511e-03, eta: 0:21:32, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0552, decode.acc_seg: 83.4441, aux.loss_seg: 0.0241, aux.acc_seg: 83.3210, loss: 0.0793\n",
      "2021-04-20 11:13:54,630 - mmseg - INFO - Iter [3970/5000]\tlr: 2.491e-03, eta: 0:21:18, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0501, decode.acc_seg: 83.9272, aux.loss_seg: 0.0221, aux.acc_seg: 83.7773, loss: 0.0722\n",
      "2021-04-20 11:14:00,943 - mmseg - INFO - Iter [3980/5000]\tlr: 2.470e-03, eta: 0:21:04, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0549, decode.acc_seg: 84.5485, aux.loss_seg: 0.0243, aux.acc_seg: 84.3580, loss: 0.0792\n",
      "2021-04-20 11:14:07,257 - mmseg - INFO - Iter [3990/5000]\tlr: 2.449e-03, eta: 0:20:50, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0515, decode.acc_seg: 80.4189, aux.loss_seg: 0.0229, aux.acc_seg: 80.2412, loss: 0.0744\n",
      "2021-04-20 11:14:13,563 - mmseg - INFO - Saving checkpoint at 4000 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.3 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 11:16:28,011 - mmseg - INFO - per class results:\n",
      "2021-04-20 11:16:28,012 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.62 | 98.65 |\n",
      "| cell       | 95.97 | 98.03 |\n",
      "| cilia      | 81.97 | 90.76 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 11:16:28,013 - mmseg - INFO - Summary:\n",
      "2021-04-20 11:16:28,013 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 91.85 | 95.82 | 97.84 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 11:16:28,020 - mmseg - INFO - Iter(val) [4000]\tmIoU: 0.9185, mAcc: 0.9582, aAcc: 0.9784, IoU.background: 0.9762, IoU.cell: 0.9597, IoU.cilia: 0.8197, Acc.background: 0.9865, Acc.cell: 0.9803, Acc.cilia: 0.9076\n",
      "2021-04-20 11:16:34,279 - mmseg - INFO - Iter [4010/5000]\tlr: 2.407e-03, eta: 0:20:53, time: 14.037, data_time: 13.415, memory: 9591, decode.loss_seg: 0.0506, decode.acc_seg: 81.1993, aux.loss_seg: 0.0224, aux.acc_seg: 81.0309, loss: 0.0731\n",
      "2021-04-20 11:16:40,498 - mmseg - INFO - Iter [4020/5000]\tlr: 2.386e-03, eta: 0:20:39, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0515, decode.acc_seg: 82.9260, aux.loss_seg: 0.0228, aux.acc_seg: 82.7476, loss: 0.0744\n",
      "2021-04-20 11:16:46,740 - mmseg - INFO - Iter [4030/5000]\tlr: 2.365e-03, eta: 0:20:25, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0533, decode.acc_seg: 81.2382, aux.loss_seg: 0.0237, aux.acc_seg: 81.0587, loss: 0.0769\n",
      "2021-04-20 11:16:52,994 - mmseg - INFO - Iter [4040/5000]\tlr: 2.344e-03, eta: 0:20:11, time: 0.625, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0551, decode.acc_seg: 80.1467, aux.loss_seg: 0.0241, aux.acc_seg: 79.9769, loss: 0.0793\n",
      "2021-04-20 11:16:59,251 - mmseg - INFO - Iter [4050/5000]\tlr: 2.323e-03, eta: 0:19:57, time: 0.626, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0509, decode.acc_seg: 85.0109, aux.loss_seg: 0.0223, aux.acc_seg: 84.8640, loss: 0.0732\n",
      "2021-04-20 11:17:05,532 - mmseg - INFO - Iter [4060/5000]\tlr: 2.302e-03, eta: 0:19:43, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0533, decode.acc_seg: 86.2777, aux.loss_seg: 0.0235, aux.acc_seg: 86.1268, loss: 0.0768\n",
      "2021-04-20 11:17:11,838 - mmseg - INFO - Iter [4070/5000]\tlr: 2.281e-03, eta: 0:19:29, time: 0.631, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0473, decode.acc_seg: 82.1592, aux.loss_seg: 0.0208, aux.acc_seg: 82.0218, loss: 0.0681\n",
      "2021-04-20 11:17:18,135 - mmseg - INFO - Iter [4080/5000]\tlr: 2.260e-03, eta: 0:19:15, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0544, decode.acc_seg: 84.9698, aux.loss_seg: 0.0241, aux.acc_seg: 84.8070, loss: 0.0785\n",
      "2021-04-20 11:17:24,437 - mmseg - INFO - Iter [4090/5000]\tlr: 2.239e-03, eta: 0:19:01, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0518, decode.acc_seg: 82.7320, aux.loss_seg: 0.0226, aux.acc_seg: 82.6058, loss: 0.0744\n",
      "2021-04-20 11:17:30,741 - mmseg - INFO - Iter [4100/5000]\tlr: 2.217e-03, eta: 0:18:47, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0582, decode.acc_seg: 84.3737, aux.loss_seg: 0.0255, aux.acc_seg: 84.2074, loss: 0.0837\n",
      "2021-04-20 11:17:37,048 - mmseg - INFO - Iter [4110/5000]\tlr: 2.196e-03, eta: 0:18:33, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0509, decode.acc_seg: 81.9689, aux.loss_seg: 0.0225, aux.acc_seg: 81.8207, loss: 0.0734\n",
      "2021-04-20 11:17:43,365 - mmseg - INFO - Iter [4120/5000]\tlr: 2.175e-03, eta: 0:18:19, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0571, decode.acc_seg: 80.8010, aux.loss_seg: 0.0253, aux.acc_seg: 80.6151, loss: 0.0824\n",
      "2021-04-20 11:17:49,675 - mmseg - INFO - Iter [4130/5000]\tlr: 2.154e-03, eta: 0:18:05, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0608, decode.acc_seg: 81.8310, aux.loss_seg: 0.0267, aux.acc_seg: 81.6726, loss: 0.0876\n",
      "2021-04-20 11:17:55,986 - mmseg - INFO - Iter [4140/5000]\tlr: 2.133e-03, eta: 0:17:52, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0530, decode.acc_seg: 82.8628, aux.loss_seg: 0.0239, aux.acc_seg: 82.6209, loss: 0.0769\n",
      "2021-04-20 11:18:02,295 - mmseg - INFO - Iter [4150/5000]\tlr: 2.111e-03, eta: 0:17:38, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0555, decode.acc_seg: 83.4357, aux.loss_seg: 0.0243, aux.acc_seg: 83.2910, loss: 0.0798\n",
      "2021-04-20 11:18:08,607 - mmseg - INFO - Iter [4160/5000]\tlr: 2.090e-03, eta: 0:17:24, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0560, decode.acc_seg: 80.3406, aux.loss_seg: 0.0250, aux.acc_seg: 80.1337, loss: 0.0810\n",
      "2021-04-20 11:18:14,920 - mmseg - INFO - Iter [4170/5000]\tlr: 2.069e-03, eta: 0:17:10, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0552, decode.acc_seg: 80.4351, aux.loss_seg: 0.0243, aux.acc_seg: 80.2719, loss: 0.0795\n",
      "2021-04-20 11:18:21,232 - mmseg - INFO - Iter [4180/5000]\tlr: 2.047e-03, eta: 0:16:57, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0553, decode.acc_seg: 82.0385, aux.loss_seg: 0.0242, aux.acc_seg: 81.8778, loss: 0.0795\n",
      "2021-04-20 11:18:27,559 - mmseg - INFO - Iter [4190/5000]\tlr: 2.026e-03, eta: 0:16:43, time: 0.633, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0721, decode.acc_seg: 85.2706, aux.loss_seg: 0.0297, aux.acc_seg: 85.0761, loss: 0.1019\n",
      "2021-04-20 11:18:33,857 - mmseg - INFO - Saving checkpoint at 4200 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.3 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 11:20:48,235 - mmseg - INFO - per class results:\n",
      "2021-04-20 11:20:48,236 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.62 | 98.76 |\n",
      "| cell       | 95.93 | 97.64 |\n",
      "| cilia      | 81.99 | 91.42 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 11:20:48,237 - mmseg - INFO - Summary:\n",
      "2021-04-20 11:20:48,237 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 91.84 | 95.94 | 97.83 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 11:20:48,244 - mmseg - INFO - Iter(val) [4200]\tmIoU: 0.9184, mAcc: 0.9594, aAcc: 0.9783, IoU.background: 0.9762, IoU.cell: 0.9593, IoU.cilia: 0.8199, Acc.background: 0.9876, Acc.cell: 0.9764, Acc.cilia: 0.9142\n",
      "2021-04-20 11:20:54,473 - mmseg - INFO - Iter [4210/5000]\tlr: 1.983e-03, eta: 0:16:40, time: 14.026, data_time: 13.407, memory: 9591, decode.loss_seg: 0.0555, decode.acc_seg: 84.8768, aux.loss_seg: 0.0242, aux.acc_seg: 84.7450, loss: 0.0796\n",
      "2021-04-20 11:21:00,686 - mmseg - INFO - Iter [4220/5000]\tlr: 1.962e-03, eta: 0:16:26, time: 0.621, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0554, decode.acc_seg: 85.2377, aux.loss_seg: 0.0243, aux.acc_seg: 85.0899, loss: 0.0797\n",
      "2021-04-20 11:21:09,067 - mmseg - INFO - Iter [4230/5000]\tlr: 1.940e-03, eta: 0:16:13, time: 0.838, data_time: 0.216, memory: 9591, decode.loss_seg: 0.0572, decode.acc_seg: 83.3084, aux.loss_seg: 0.0250, aux.acc_seg: 83.1619, loss: 0.0821\n",
      "2021-04-20 11:21:15,312 - mmseg - INFO - Iter [4240/5000]\tlr: 1.919e-03, eta: 0:15:59, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0534, decode.acc_seg: 77.7820, aux.loss_seg: 0.0236, aux.acc_seg: 77.6278, loss: 0.0770\n",
      "2021-04-20 11:21:21,561 - mmseg - INFO - Iter [4250/5000]\tlr: 1.897e-03, eta: 0:15:45, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0535, decode.acc_seg: 84.7355, aux.loss_seg: 0.0236, aux.acc_seg: 84.5813, loss: 0.0772\n",
      "2021-04-20 11:21:27,839 - mmseg - INFO - Iter [4260/5000]\tlr: 1.876e-03, eta: 0:15:32, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0508, decode.acc_seg: 81.4799, aux.loss_seg: 0.0224, aux.acc_seg: 81.3391, loss: 0.0731\n",
      "2021-04-20 11:21:34,123 - mmseg - INFO - Iter [4270/5000]\tlr: 1.854e-03, eta: 0:15:18, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0535, decode.acc_seg: 83.0756, aux.loss_seg: 0.0237, aux.acc_seg: 82.9061, loss: 0.0772\n",
      "2021-04-20 11:21:40,413 - mmseg - INFO - Iter [4280/5000]\tlr: 1.833e-03, eta: 0:15:04, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0519, decode.acc_seg: 84.7262, aux.loss_seg: 0.0230, aux.acc_seg: 84.5676, loss: 0.0748\n",
      "2021-04-20 11:21:46,710 - mmseg - INFO - Iter [4290/5000]\tlr: 1.811e-03, eta: 0:14:51, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0531, decode.acc_seg: 80.1155, aux.loss_seg: 0.0230, aux.acc_seg: 79.9836, loss: 0.0761\n",
      "2021-04-20 11:21:53,012 - mmseg - INFO - Iter [4300/5000]\tlr: 1.789e-03, eta: 0:14:37, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0541, decode.acc_seg: 81.2946, aux.loss_seg: 0.0240, aux.acc_seg: 81.1081, loss: 0.0781\n",
      "2021-04-20 11:21:59,325 - mmseg - INFO - Iter [4310/5000]\tlr: 1.768e-03, eta: 0:14:24, time: 0.631, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0474, decode.acc_seg: 80.1152, aux.loss_seg: 0.0207, aux.acc_seg: 79.9893, loss: 0.0680\n",
      "2021-04-20 11:22:05,632 - mmseg - INFO - Iter [4320/5000]\tlr: 1.746e-03, eta: 0:14:10, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0573, decode.acc_seg: 83.8530, aux.loss_seg: 0.0252, aux.acc_seg: 83.6643, loss: 0.0825\n",
      "2021-04-20 11:22:11,950 - mmseg - INFO - Iter [4330/5000]\tlr: 1.724e-03, eta: 0:13:57, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0556, decode.acc_seg: 81.3361, aux.loss_seg: 0.0246, aux.acc_seg: 81.1627, loss: 0.0802\n",
      "2021-04-20 11:22:18,261 - mmseg - INFO - Iter [4340/5000]\tlr: 1.702e-03, eta: 0:13:43, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0537, decode.acc_seg: 80.7167, aux.loss_seg: 0.0236, aux.acc_seg: 80.5525, loss: 0.0774\n",
      "2021-04-20 11:22:24,569 - mmseg - INFO - Iter [4350/5000]\tlr: 1.680e-03, eta: 0:13:30, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0569, decode.acc_seg: 81.3992, aux.loss_seg: 0.0249, aux.acc_seg: 81.2355, loss: 0.0818\n",
      "2021-04-20 11:22:30,877 - mmseg - INFO - Iter [4360/5000]\tlr: 1.659e-03, eta: 0:13:16, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0565, decode.acc_seg: 81.0766, aux.loss_seg: 0.0248, aux.acc_seg: 80.9098, loss: 0.0813\n",
      "2021-04-20 11:22:37,188 - mmseg - INFO - Iter [4370/5000]\tlr: 1.637e-03, eta: 0:13:03, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0547, decode.acc_seg: 81.7155, aux.loss_seg: 0.0241, aux.acc_seg: 81.5579, loss: 0.0788\n",
      "2021-04-20 11:22:43,498 - mmseg - INFO - Iter [4380/5000]\tlr: 1.615e-03, eta: 0:12:50, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0583, decode.acc_seg: 81.9043, aux.loss_seg: 0.0258, aux.acc_seg: 81.7014, loss: 0.0841\n",
      "2021-04-20 11:22:49,807 - mmseg - INFO - Iter [4390/5000]\tlr: 1.593e-03, eta: 0:12:37, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0487, decode.acc_seg: 83.6652, aux.loss_seg: 0.0217, aux.acc_seg: 83.5196, loss: 0.0704\n",
      "2021-04-20 11:22:56,111 - mmseg - INFO - Saving checkpoint at 4400 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.4 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 11:25:10,382 - mmseg - INFO - per class results:\n",
      "2021-04-20 11:25:10,383 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.62 | 98.61 |\n",
      "| cell       | 96.03 | 97.92 |\n",
      "| cilia      | 82.21 | 91.74 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 11:25:10,383 - mmseg - INFO - Summary:\n",
      "2021-04-20 11:25:10,384 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 91.96 | 96.09 | 97.86 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 11:25:10,391 - mmseg - INFO - Iter(val) [4400]\tmIoU: 0.9196, mAcc: 0.9609, aAcc: 0.9786, IoU.background: 0.9762, IoU.cell: 0.9603, IoU.cilia: 0.8221, Acc.background: 0.9861, Acc.cell: 0.9792, Acc.cilia: 0.9174\n",
      "2021-04-20 11:25:16,674 - mmseg - INFO - Iter [4410/5000]\tlr: 1.549e-03, eta: 0:12:27, time: 14.006, data_time: 13.382, memory: 9591, decode.loss_seg: 0.0536, decode.acc_seg: 80.6412, aux.loss_seg: 0.0238, aux.acc_seg: 80.4628, loss: 0.0774\n",
      "2021-04-20 11:25:22,892 - mmseg - INFO - Iter [4420/5000]\tlr: 1.527e-03, eta: 0:12:14, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0554, decode.acc_seg: 84.3666, aux.loss_seg: 0.0247, aux.acc_seg: 84.1857, loss: 0.0801\n",
      "2021-04-20 11:25:29,138 - mmseg - INFO - Iter [4430/5000]\tlr: 1.505e-03, eta: 0:12:00, time: 0.625, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0484, decode.acc_seg: 80.3238, aux.loss_seg: 0.0215, aux.acc_seg: 80.1761, loss: 0.0699\n",
      "2021-04-20 11:25:35,384 - mmseg - INFO - Iter [4440/5000]\tlr: 1.482e-03, eta: 0:11:47, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0592, decode.acc_seg: 81.0354, aux.loss_seg: 0.0265, aux.acc_seg: 80.8075, loss: 0.0856\n",
      "2021-04-20 11:25:41,656 - mmseg - INFO - Iter [4450/5000]\tlr: 1.460e-03, eta: 0:11:33, time: 0.627, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0509, decode.acc_seg: 82.3344, aux.loss_seg: 0.0225, aux.acc_seg: 82.1872, loss: 0.0734\n",
      "2021-04-20 11:25:47,948 - mmseg - INFO - Iter [4460/5000]\tlr: 1.438e-03, eta: 0:11:20, time: 0.629, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0512, decode.acc_seg: 84.3749, aux.loss_seg: 0.0225, aux.acc_seg: 84.2375, loss: 0.0736\n",
      "2021-04-20 11:25:54,236 - mmseg - INFO - Iter [4470/5000]\tlr: 1.416e-03, eta: 0:11:07, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0529, decode.acc_seg: 82.2785, aux.loss_seg: 0.0235, aux.acc_seg: 82.0981, loss: 0.0764\n",
      "2021-04-20 11:26:00,532 - mmseg - INFO - Iter [4480/5000]\tlr: 1.393e-03, eta: 0:10:53, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0559, decode.acc_seg: 82.8129, aux.loss_seg: 0.0243, aux.acc_seg: 82.6727, loss: 0.0802\n",
      "2021-04-20 11:26:06,829 - mmseg - INFO - Iter [4490/5000]\tlr: 1.371e-03, eta: 0:10:40, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0511, decode.acc_seg: 82.1291, aux.loss_seg: 0.0224, aux.acc_seg: 81.9903, loss: 0.0734\n",
      "2021-04-20 11:26:13,131 - mmseg - INFO - Iter [4500/5000]\tlr: 1.349e-03, eta: 0:10:27, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0515, decode.acc_seg: 83.7645, aux.loss_seg: 0.0227, aux.acc_seg: 83.6081, loss: 0.0742\n",
      "2021-04-20 11:26:19,437 - mmseg - INFO - Iter [4510/5000]\tlr: 1.326e-03, eta: 0:10:13, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0488, decode.acc_seg: 82.9762, aux.loss_seg: 0.0214, aux.acc_seg: 82.8519, loss: 0.0703\n",
      "2021-04-20 11:26:25,741 - mmseg - INFO - Iter [4520/5000]\tlr: 1.304e-03, eta: 0:10:00, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0527, decode.acc_seg: 83.7975, aux.loss_seg: 0.0234, aux.acc_seg: 83.6207, loss: 0.0761\n",
      "2021-04-20 11:26:32,048 - mmseg - INFO - Iter [4530/5000]\tlr: 1.281e-03, eta: 0:09:47, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0538, decode.acc_seg: 87.7026, aux.loss_seg: 0.0238, aux.acc_seg: 87.5317, loss: 0.0776\n",
      "2021-04-20 11:26:38,359 - mmseg - INFO - Iter [4540/5000]\tlr: 1.258e-03, eta: 0:09:34, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0516, decode.acc_seg: 86.3619, aux.loss_seg: 0.0232, aux.acc_seg: 86.1900, loss: 0.0747\n",
      "2021-04-20 11:26:44,669 - mmseg - INFO - Iter [4550/5000]\tlr: 1.236e-03, eta: 0:09:21, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0505, decode.acc_seg: 83.8468, aux.loss_seg: 0.0224, aux.acc_seg: 83.6919, loss: 0.0729\n",
      "2021-04-20 11:26:50,979 - mmseg - INFO - Iter [4560/5000]\tlr: 1.213e-03, eta: 0:09:08, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0484, decode.acc_seg: 83.6755, aux.loss_seg: 0.0215, aux.acc_seg: 83.5050, loss: 0.0699\n",
      "2021-04-20 11:26:57,289 - mmseg - INFO - Iter [4570/5000]\tlr: 1.190e-03, eta: 0:08:55, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0528, decode.acc_seg: 85.1494, aux.loss_seg: 0.0232, aux.acc_seg: 84.9878, loss: 0.0760\n",
      "2021-04-20 11:27:03,598 - mmseg - INFO - Iter [4580/5000]\tlr: 1.168e-03, eta: 0:08:42, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0521, decode.acc_seg: 84.4411, aux.loss_seg: 0.0230, aux.acc_seg: 84.2935, loss: 0.0751\n",
      "2021-04-20 11:27:09,906 - mmseg - INFO - Iter [4590/5000]\tlr: 1.145e-03, eta: 0:08:29, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0465, decode.acc_seg: 81.9176, aux.loss_seg: 0.0205, aux.acc_seg: 81.7938, loss: 0.0670\n",
      "2021-04-20 11:27:16,210 - mmseg - INFO - Saving checkpoint at 4600 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.1 task/s, elapsed: 117s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 11:29:31,247 - mmseg - INFO - per class results:\n",
      "2021-04-20 11:29:31,248 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.66 | 98.71 |\n",
      "| cell       | 96.08 | 98.08 |\n",
      "| cilia      | 82.36 | 90.77 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 11:29:31,249 - mmseg - INFO - Summary:\n",
      "2021-04-20 11:29:31,249 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 92.03 | 95.85 | 97.89 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 11:29:31,256 - mmseg - INFO - Iter(val) [4600]\tmIoU: 0.9203, mAcc: 0.9585, aAcc: 0.9789, IoU.background: 0.9766, IoU.cell: 0.9608, IoU.cilia: 0.8236, Acc.background: 0.9871, Acc.cell: 0.9808, Acc.cilia: 0.9077\n",
      "2021-04-20 11:29:37,511 - mmseg - INFO - Iter [4610/5000]\tlr: 1.099e-03, eta: 0:08:14, time: 14.097, data_time: 13.476, memory: 9591, decode.loss_seg: 0.0541, decode.acc_seg: 82.8898, aux.loss_seg: 0.0239, aux.acc_seg: 82.7469, loss: 0.0780\n",
      "2021-04-20 11:29:43,730 - mmseg - INFO - Iter [4620/5000]\tlr: 1.076e-03, eta: 0:08:01, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0523, decode.acc_seg: 83.4580, aux.loss_seg: 0.0231, aux.acc_seg: 83.2843, loss: 0.0754\n",
      "2021-04-20 11:29:49,968 - mmseg - INFO - Iter [4630/5000]\tlr: 1.053e-03, eta: 0:07:47, time: 0.624, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0523, decode.acc_seg: 84.0125, aux.loss_seg: 0.0231, aux.acc_seg: 83.8478, loss: 0.0754\n",
      "2021-04-20 11:29:56,222 - mmseg - INFO - Iter [4640/5000]\tlr: 1.030e-03, eta: 0:07:34, time: 0.625, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0517, decode.acc_seg: 82.3018, aux.loss_seg: 0.0226, aux.acc_seg: 82.1545, loss: 0.0743\n",
      "2021-04-20 11:30:02,480 - mmseg - INFO - Iter [4650/5000]\tlr: 1.006e-03, eta: 0:07:21, time: 0.626, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0532, decode.acc_seg: 82.2586, aux.loss_seg: 0.0231, aux.acc_seg: 82.1324, loss: 0.0763\n",
      "2021-04-20 11:30:08,774 - mmseg - INFO - Iter [4660/5000]\tlr: 9.832e-04, eta: 0:07:08, time: 0.629, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0492, decode.acc_seg: 84.6556, aux.loss_seg: 0.0216, aux.acc_seg: 84.5193, loss: 0.0708\n",
      "2021-04-20 11:30:15,063 - mmseg - INFO - Iter [4670/5000]\tlr: 9.598e-04, eta: 0:06:55, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0518, decode.acc_seg: 81.2945, aux.loss_seg: 0.0228, aux.acc_seg: 81.1485, loss: 0.0745\n",
      "2021-04-20 11:30:21,363 - mmseg - INFO - Iter [4680/5000]\tlr: 9.364e-04, eta: 0:06:42, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0521, decode.acc_seg: 80.7830, aux.loss_seg: 0.0228, aux.acc_seg: 80.6148, loss: 0.0749\n",
      "2021-04-20 11:30:27,670 - mmseg - INFO - Iter [4690/5000]\tlr: 9.129e-04, eta: 0:06:29, time: 0.631, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0507, decode.acc_seg: 80.1838, aux.loss_seg: 0.0222, aux.acc_seg: 80.0406, loss: 0.0729\n",
      "2021-04-20 11:30:33,973 - mmseg - INFO - Iter [4700/5000]\tlr: 8.894e-04, eta: 0:06:16, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0569, decode.acc_seg: 82.6729, aux.loss_seg: 0.0249, aux.acc_seg: 82.5347, loss: 0.0818\n",
      "2021-04-20 11:30:40,282 - mmseg - INFO - Iter [4710/5000]\tlr: 8.657e-04, eta: 0:06:03, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0492, decode.acc_seg: 83.3871, aux.loss_seg: 0.0217, aux.acc_seg: 83.2340, loss: 0.0708\n",
      "2021-04-20 11:30:46,589 - mmseg - INFO - Iter [4720/5000]\tlr: 8.420e-04, eta: 0:05:50, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0498, decode.acc_seg: 83.8114, aux.loss_seg: 0.0223, aux.acc_seg: 83.6288, loss: 0.0721\n",
      "2021-04-20 11:30:52,898 - mmseg - INFO - Iter [4730/5000]\tlr: 8.182e-04, eta: 0:05:37, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0489, decode.acc_seg: 82.7019, aux.loss_seg: 0.0217, aux.acc_seg: 82.5391, loss: 0.0706\n",
      "2021-04-20 11:30:59,207 - mmseg - INFO - Iter [4740/5000]\tlr: 7.943e-04, eta: 0:05:24, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0476, decode.acc_seg: 84.3566, aux.loss_seg: 0.0210, aux.acc_seg: 84.2246, loss: 0.0686\n",
      "2021-04-20 11:31:05,516 - mmseg - INFO - Iter [4750/5000]\tlr: 7.703e-04, eta: 0:05:12, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0493, decode.acc_seg: 85.5197, aux.loss_seg: 0.0219, aux.acc_seg: 85.3657, loss: 0.0712\n",
      "2021-04-20 11:31:11,840 - mmseg - INFO - Iter [4760/5000]\tlr: 7.462e-04, eta: 0:04:59, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0517, decode.acc_seg: 79.1484, aux.loss_seg: 0.0231, aux.acc_seg: 78.9669, loss: 0.0747\n",
      "2021-04-20 11:31:18,163 - mmseg - INFO - Iter [4770/5000]\tlr: 7.220e-04, eta: 0:04:46, time: 0.632, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0517, decode.acc_seg: 80.9891, aux.loss_seg: 0.0223, aux.acc_seg: 80.8798, loss: 0.0740\n",
      "2021-04-20 11:31:24,477 - mmseg - INFO - Iter [4780/5000]\tlr: 6.977e-04, eta: 0:04:33, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0482, decode.acc_seg: 80.8424, aux.loss_seg: 0.0210, aux.acc_seg: 80.7181, loss: 0.0692\n",
      "2021-04-20 11:31:30,787 - mmseg - INFO - Iter [4790/5000]\tlr: 6.733e-04, eta: 0:04:21, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0480, decode.acc_seg: 82.7084, aux.loss_seg: 0.0211, aux.acc_seg: 82.5845, loss: 0.0691\n",
      "2021-04-20 11:31:37,092 - mmseg - INFO - Saving checkpoint at 4800 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.5 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 11:33:50,688 - mmseg - INFO - per class results:\n",
      "2021-04-20 11:33:50,689 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.7  | 98.75 |\n",
      "| cell       | 96.11 | 98.06 |\n",
      "| cilia      | 82.57 | 90.89 |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 11:33:50,689 - mmseg - INFO - Summary:\n",
      "2021-04-20 11:33:50,689 - mmseg - INFO - \n",
      "+--------+-------+------+-------+\n",
      "| Scope  | mIoU  | mAcc | aAcc  |\n",
      "+--------+-------+------+-------+\n",
      "| global | 92.13 | 95.9 | 97.92 |\n",
      "+--------+-------+------+-------+\n",
      "2021-04-20 11:33:50,696 - mmseg - INFO - Iter(val) [4800]\tmIoU: 0.9213, mAcc: 0.9590, aAcc: 0.9792, IoU.background: 0.9770, IoU.cell: 0.9611, IoU.cilia: 0.8257, Acc.background: 0.9875, Acc.cell: 0.9806, Acc.cilia: 0.9089\n",
      "2021-04-20 11:33:56,940 - mmseg - INFO - Iter [4810/5000]\tlr: 6.242e-04, eta: 0:04:00, time: 13.951, data_time: 13.330, memory: 9591, decode.loss_seg: 0.0526, decode.acc_seg: 84.2828, aux.loss_seg: 0.0232, aux.acc_seg: 84.1184, loss: 0.0757\n",
      "2021-04-20 11:34:03,160 - mmseg - INFO - Iter [4820/5000]\tlr: 5.994e-04, eta: 0:03:47, time: 0.622, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0560, decode.acc_seg: 84.9056, aux.loss_seg: 0.0248, aux.acc_seg: 84.7204, loss: 0.0808\n",
      "2021-04-20 11:34:09,407 - mmseg - INFO - Iter [4830/5000]\tlr: 5.745e-04, eta: 0:03:34, time: 0.625, data_time: 0.005, memory: 9591, decode.loss_seg: 0.0508, decode.acc_seg: 83.8741, aux.loss_seg: 0.0227, aux.acc_seg: 83.6872, loss: 0.0735\n",
      "2021-04-20 11:34:15,657 - mmseg - INFO - Iter [4840/5000]\tlr: 5.495e-04, eta: 0:03:22, time: 0.625, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0578, decode.acc_seg: 85.9441, aux.loss_seg: 0.0256, aux.acc_seg: 85.7574, loss: 0.0835\n",
      "2021-04-20 11:34:21,914 - mmseg - INFO - Iter [4850/5000]\tlr: 5.243e-04, eta: 0:03:09, time: 0.626, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0460, decode.acc_seg: 83.1365, aux.loss_seg: 0.0202, aux.acc_seg: 82.9874, loss: 0.0662\n",
      "2021-04-20 11:34:28,195 - mmseg - INFO - Iter [4860/5000]\tlr: 4.989e-04, eta: 0:02:56, time: 0.628, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0494, decode.acc_seg: 81.3504, aux.loss_seg: 0.0218, aux.acc_seg: 81.1912, loss: 0.0713\n",
      "2021-04-20 11:34:34,489 - mmseg - INFO - Iter [4870/5000]\tlr: 4.733e-04, eta: 0:02:43, time: 0.629, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0473, decode.acc_seg: 81.7381, aux.loss_seg: 0.0208, aux.acc_seg: 81.6049, loss: 0.0680\n",
      "2021-04-20 11:34:40,788 - mmseg - INFO - Iter [4880/5000]\tlr: 4.476e-04, eta: 0:02:30, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0477, decode.acc_seg: 85.3305, aux.loss_seg: 0.0210, aux.acc_seg: 85.1912, loss: 0.0686\n",
      "2021-04-20 11:34:47,088 - mmseg - INFO - Iter [4890/5000]\tlr: 4.216e-04, eta: 0:02:18, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0501, decode.acc_seg: 83.1524, aux.loss_seg: 0.0219, aux.acc_seg: 83.0277, loss: 0.0720\n",
      "2021-04-20 11:34:53,389 - mmseg - INFO - Iter [4900/5000]\tlr: 3.954e-04, eta: 0:02:05, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0468, decode.acc_seg: 81.6812, aux.loss_seg: 0.0205, aux.acc_seg: 81.5281, loss: 0.0673\n",
      "2021-04-20 11:34:59,691 - mmseg - INFO - Iter [4910/5000]\tlr: 3.690e-04, eta: 0:01:52, time: 0.630, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0525, decode.acc_seg: 86.0128, aux.loss_seg: 0.0232, aux.acc_seg: 85.8434, loss: 0.0757\n",
      "2021-04-20 11:35:05,997 - mmseg - INFO - Iter [4920/5000]\tlr: 3.422e-04, eta: 0:01:40, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0466, decode.acc_seg: 80.6341, aux.loss_seg: 0.0205, aux.acc_seg: 80.4992, loss: 0.0670\n",
      "2021-04-20 11:35:12,308 - mmseg - INFO - Iter [4930/5000]\tlr: 3.151e-04, eta: 0:01:27, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0522, decode.acc_seg: 82.3833, aux.loss_seg: 0.0229, aux.acc_seg: 82.2413, loss: 0.0751\n",
      "2021-04-20 11:35:18,617 - mmseg - INFO - Iter [4940/5000]\tlr: 2.877e-04, eta: 0:01:15, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0493, decode.acc_seg: 78.6000, aux.loss_seg: 0.0218, aux.acc_seg: 78.4495, loss: 0.0711\n",
      "2021-04-20 11:35:24,927 - mmseg - INFO - Iter [4950/5000]\tlr: 2.597e-04, eta: 0:01:02, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0535, decode.acc_seg: 83.7501, aux.loss_seg: 0.0234, aux.acc_seg: 83.5978, loss: 0.0769\n",
      "2021-04-20 11:35:31,236 - mmseg - INFO - Iter [4960/5000]\tlr: 2.312e-04, eta: 0:00:49, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0511, decode.acc_seg: 82.3694, aux.loss_seg: 0.0224, aux.acc_seg: 82.2149, loss: 0.0734\n",
      "2021-04-20 11:35:37,545 - mmseg - INFO - Iter [4970/5000]\tlr: 2.020e-04, eta: 0:00:37, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0540, decode.acc_seg: 82.8928, aux.loss_seg: 0.0239, aux.acc_seg: 82.7513, loss: 0.0779\n",
      "2021-04-20 11:35:43,855 - mmseg - INFO - Iter [4980/5000]\tlr: 1.719e-04, eta: 0:00:24, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0464, decode.acc_seg: 80.0232, aux.loss_seg: 0.0207, aux.acc_seg: 79.8695, loss: 0.0671\n",
      "2021-04-20 11:35:50,164 - mmseg - INFO - Iter [4990/5000]\tlr: 1.402e-04, eta: 0:00:12, time: 0.631, data_time: 0.004, memory: 9591, decode.loss_seg: 0.0506, decode.acc_seg: 84.1421, aux.loss_seg: 0.0225, aux.acc_seg: 83.9773, loss: 0.0731\n",
      "2021-04-20 11:35:56,475 - mmseg - INFO - Saving checkpoint at 5000 iterations\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[>>>>>>>>>>>>>>>>>>>>>>>>>>] 4220/4220, 36.4 task/s, elapsed: 116s, ETA:     0s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-20 11:38:10,382 - mmseg - INFO - per class results:\n",
      "2021-04-20 11:38:10,383 - mmseg - INFO - \n",
      "+------------+-------+-------+\n",
      "| Class      | IoU   | Acc   |\n",
      "+------------+-------+-------+\n",
      "| background | 97.72 | 98.72 |\n",
      "| cell       | 96.15 | 98.07 |\n",
      "| cilia      | 82.87 | 91.4  |\n",
      "+------------+-------+-------+\n",
      "2021-04-20 11:38:10,383 - mmseg - INFO - Summary:\n",
      "2021-04-20 11:38:10,383 - mmseg - INFO - \n",
      "+--------+-------+-------+-------+\n",
      "| Scope  | mIoU  | mAcc  | aAcc  |\n",
      "+--------+-------+-------+-------+\n",
      "| global | 92.25 | 96.06 | 97.94 |\n",
      "+--------+-------+-------+-------+\n",
      "2021-04-20 11:38:10,390 - mmseg - INFO - Iter(val) [5000]\tmIoU: 0.9225, mAcc: 0.9606, aAcc: 0.9794, IoU.background: 0.9772, IoU.cell: 0.9615, IoU.cilia: 0.8287, Acc.background: 0.9872, Acc.cell: 0.9807, Acc.cilia: 0.9140\n"
     ]
    }
   ],
   "source": [
    "from mmseg.datasets import build_dataset\n",
    "from mmseg.models import build_segmentor\n",
    "from mmseg.apis import train_segmentor\n",
    "\n",
    "\n",
    "# Build the dataset\n",
    "datasets = [build_dataset(cfg.data.train)]\n",
    "\n",
    "# Build the detector\n",
    "model = build_segmentor(\n",
    "    cfg.model, train_cfg=cfg.get('train_cfg'), test_cfg=cfg.get('test_cfg'))\n",
    "# Add an attribute for visualization convenience\n",
    "model.CLASSES = datasets[0].CLASSES\n",
    "\n",
    "# Create work_dir\n",
    "mmcv.mkdir_or_exist(osp.abspath(cfg.work_dir))\n",
    "train_segmentor(model, datasets, cfg, distributed=False, validate=True, \n",
    "                meta=dict())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/esbg-lab/Desktop/P3/mmsegmentation/mmseg/models/segmentors/base.py:271: UserWarning: show==False and out_file is not specified, only result image will be returned\n",
      "  warnings.warn('show==False and out_file is not specified, only '\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 576x432 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "img = mmcv.imread('../data/my_dataset/img_dir/train/007f736aedbc4ca67989f8ca62f1bbeb447ad76698351fe387923963ee50e5ae_frame0000.png')\n",
    "\n",
    "model.cfg = cfg\n",
    "result = inference_segmentor(model, img)\n",
    "plt.figure(figsize=(8, 6))\n",
    "show_result_pyplot(model, img, result, palette)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "img = Image.open('../data/my_dataset/ann_dir/train/007f736aedbc4ca67989f8ca62f1bbeb447ad76698351fe387923963ee50e5ae_frame0000.png')\n",
    "plt.figure(figsize=(8, 6))\n",
    "im = plt.imshow(np.array(img))\n",
    "\n",
    "# create a patch (proxy artist) for every color \n",
    "patches = [mpatches.Patch(color=np.array(palette[i])/255., \n",
    "                          label=classes[i]) for i in range(3)]\n",
    "# put those patched as legend-handles into the legend\n",
    "plt.legend(handles=patches, bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0., \n",
    "           fontsize='large')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "rootdir = '/home/esbg-lab/Desktop/P3/data/my_dataset/img_dir/test'\n",
    "\n",
    "for subdir, dirs, files in os.walk(rootdir):\n",
    "    for file in files:\n",
    "        if str(file).split(\"_frame\")[1]==\"0000.png\":\n",
    "#             print(str(file).split(\"_frame\")[0])\n",
    "            img = mmcv.imread(os.path.join(subdir, file))\n",
    "            result = inference_segmentor(model, img)\n",
    "            im = Image.fromarray(result[0].astype(np.uint8))\n",
    "            im.save(\"/home/esbg-lab/Desktop/P3/predictions/\"+str(str(file).split(\"_frame\")[0])+\".png\")\n",
    "#             print(result[0].shape)\n",
    "            \n",
    "#         print(str(file).split(\"_frame\")[1])\n",
    "#         print(str(subdir).split(\"/\")[-1])\n",
    "#         print(img)\n",
    "#         new_name = str(subdir).split(\"/\")[-1] +\"_\"+ file\n",
    "#         shutil.copy(os.path.join(subdir, file), \"./data/my_dataset/img_dir/test/\"+new_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}