PaddleX---dEEPlABv3+语义分割
2024-05-16 10:49:01
PaddleX简介：
如前几篇关于PaddleX的介绍一样，它是飞桨全流程开发工具，方便用户根据实际生产需求进行直接调用或二次开发。如有需要可以点击如下链接进行查询。
PaddleX代码GitHub链接：https://github.com/PaddlePaddle/PaddleX/tree/develop
PaddleX文档链接：https://paddlex.readthedocs.io/zh_CN/latest/index.html
PaddleX官网链接：https://www.paddlepaddle.org.cn/paddle/paddlex
DeepLabV3+简介
DeepLabv3在DeepLab模型的基础上，通过encoder进行多尺度信息的融合，提高了语义分割的运行效率。本示例在一个小数据集上展示了如何通过PaddleX进行训练。
首先依旧是安装PaddleX
1. 安装PaddleX
! pip install paddlex -i https://mirror.baidu.com/pypi/simple此处安装后的跟其他的paddlex的一样，就省去了。
2.准备视盘语义分割数据集
#获取paddlex下数据集中的optic_disc_seg.tar.gz
! wget https://bj.bcebos.com/paddlex/datasets/optic_disc_seg.tar.gz
#解压数据集optic_disc_seg.tar.gz
! tar xzf optic_disc_seg.tar.gz--2020-05-09 09:29:27-- https://bj.bcebos.com/paddlex/datasets/optic_disc_seg.tar.gz Resolving bj.bcebos.com (bj.bcebos.com)... 182.61.200.229, 182.61.200.195 Connecting to bj.bcebos.com (bj.bcebos.com)|182.61.200.229|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 19077558 (18M) [application/octet-stream] Saving to: ‘optic_disc_seg.tar.gz’optic_disc_seg.tar. 100%[===================>] 18.19M 47.9MB/s in 0.4s 2020-05-09 09:29:28 (47.9 MB/s) - ‘optic_disc_seg.tar.gz’ saved [19077558/19077558]
下载的数据集为
解压此数据集为
3. 模型训练
3.1 配置GPU
#设置使用0号GPU卡（如无GPU,执行此代码后仍会使用GPU训练模型）
import matplotlib
matplotlib.use('Agg')
import os
os.environ[CUDA_VISIBLE_DEVICES'] = '0'
import paddlex as pdx2020-05-09 09:32:51,743-INFO: font search path ['/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/mpl-data/fonts/ttf', '/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/mpl-data/fonts/afm', '/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/mpl-data/fonts/pdfcorefonts'] 2020-05-09 09:32:52,200-INFO: generated new fontManager
3.2 定义图像预处理流程transforms
定义数据处理流程，其中训练和测试需分别定义，训练过程包括了部分测试过程中不需要的数据增强操作，如在本示例中，训练过程使用了RandomHorizontalFlip和RandomPaddingCrop两种数据增强方式。
from paddlex.seg import transforms
train_transforms = transforms.Compose([transforms.RandomHorizontalFlip(),transforms.Resize(target_size=512),transformsRandomPaddingCrop(crop_size=500),transforms.Normalize()
])
eval_transforms = transforms.Compose([transforms.Resize(512),transforms.Normalize()
])3.3 定义数据集Dataset
实例分割使用SegDataset格式的数据集，因此采用pdx.datasets.SegDataset来加载数据集
train_dataset = pdx.datasets.SegDateset(data_dir='optic_disc_seg',file_list='optic_sisc_seg/train_list.txt',label_list='optic_disc_seg/labels.txt',transforms=transforms,shuffle=True)
eval_dataset = pdx.datasets.SegDataset(data_dir='optic_disc_seg',file_list='optic_disc_seg/val_list.txt',label_list='optic_disc_seg/labels.txt',transforms=eval_transforms)2020-05-09 09:45:52 [INFO] 267 samples in file optic_disc_seg/train_list.txt 2020-05-09 09:45:52 [INFO] 76 samples in file optic_disc_seg/val_list.txt
3.4 模型开始训练
使用本数据集在P 40上训练，如有GPU,模型的训练过程预估为13分钟；如无GPU，则预估5个小时左右。模型训练过程每间隔save_interval_epochs轮会保存一次模型在save_dir目录下，同时在保存的过程中也会在验证数据集上计算相关指标。
num_classes = len(train_dataset.labels)
model = pdx.seg.DeepLabv3p(num_classes=num_classes)
model.train(num_epochs=40,train_dataset=train_dataset,train_batch_size=4,eval_dataset=eval_dataset,learning_rate=0.01,save_interval_epochs=1,save_dir='output/deeplab')
有算力就是好仅仅训练了9分多就完成了。下附上训练结果，因为训练的结果有点多，所以这里只付上一部分训练结果。
Downloading MobileNetV2_pretrained.tar [==================================================] 100.00% Uncompress /home/aistudio/.paddlehub/tmp/tmp5n809v9n/MobileNetV2_pretrained.tar [==================================================] 100.00% 2020-05-09 09:46:32 [INFO] Load pretrain weights from output/deeplab/pretrain/MobileNetV2_x1.0. 2020-05-09 09:46:32 [INFO] There are 260 varaibles in output/deeplab/pretrain/MobileNetV2_x1.0 are loaded. 2020-05-09 09:46:37 [INFO] [TRAIN] Epoch=1/40, Step=2/66, loss=0.565835, lr=0.009997, time_each_step=2.38s, eta=2:14:37 2020-05-09 09:46:37 [INFO] [TRAIN] Epoch=1/40, Step=4/66, loss=0.2271, lr=0.00999, time_each_step=1.23s, eta=1:9:32 2020-05-09 09:46:37 [INFO] [TRAIN] Epoch=1/40, Step=6/66, loss=0.119317, lr=0.009983, time_each_step=0.84s, eta=0:47:41 2020-05-09 09:46:38 [INFO] [TRAIN] Epoch=1/40, Step=8/66, loss=0.093658, lr=0.009976, time_each_step=0.66s, eta=0:37:4 2020-05-09 09:46:38 [INFO] [TRAIN] Epoch=1/40, Step=10/66, loss=0.112628, lr=0.009969, time_each_step=0.54s, eta=0:30:32 2020-05-09 09:46:38 [INFO] [TRAIN] Epoch=1/40, Step=12/66, loss=0.131241, lr=0.009962, time_each_step=0.46s, eta=0:26:10 2020-05-09 09:46:38 [INFO] [TRAIN] Epoch=1/40, Step=14/66, loss=0.126554, lr=0.009956, time_each_step=0.41s, eta=0:23:4 2020-05-09 09:46:38 [INFO] [TRAIN] Epoch=1/40, Step=16/66, loss=0.132383, lr=0.009949, time_each_step=0.37s, eta=0:20:42 2020-05-09 09:46:38 [INFO] [TRAIN] Epoch=1/40, Step=18/66, loss=0.13974, lr=0.009942, time_each_step=0.33s, eta=0:18:51 2020-05-09 09:46:38 [INFO] [TRAIN] Epoch=1/40, Step=20/66, loss=0.151852, lr=0.009935, time_each_step=0.31s, eta=0:17:23 2020-05-09 09:46:39 [INFO] [TRAIN] Epoch=1/40, Step=22/66, loss=0.177936, lr=0.009928, time_each_step=0.08s, eta=0:4:24 2020-05-09 09:46:39 [INFO] [TRAIN] Epoch=1/40, Step=24/66, loss=0.13828, lr=0.009922, time_each_step=0.08s, eta=0:4:19 2020-05-09 09:46:39 [INFO] [TRAIN] Epoch=1/40, Step=26/66, loss=0.13588, lr=0.009915, time_each_step=0.08s, eta=0:4:17 2020-05-09 09:46:39 [INFO] [TRAIN] Epoch=1/40, Step=28/66, loss=0.084726, lr=0.009908, time_each_step=0.07s, eta=0:4:7 2020-05-09 09:46:39 [INFO] [TRAIN] Epoch=1/40, Step=30/66, loss=0.139684, lr=0.009901, time_each_step=0.07s, eta=0:4:1 2020-05-09 09:46:39 [INFO] [TRAIN] Epoch=1/40, Step=32/66, loss=0.166004, lr=0.009894, time_each_step=0.07s, eta=0:3:57 2020-05-09 09:46:39 [INFO] [TRAIN] Epoch=1/40, Step=34/66, loss=0.091609, lr=0.009887, time_each_step=0.07s, eta=0:3:52 2020-05-09 09:46:40 [INFO] [TRAIN] Epoch=1/40, Step=36/66, loss=0.133962, lr=0.009881, time_each_step=0.07s, eta=0:3:49 2020-05-09 09:46:40 [INFO] [TRAIN] Epoch=1/40, Step=38/66, loss=0.09712, lr=0.009874, time_each_step=0.07s, eta=0:3:45 2020-05-09 09:46:40 [INFO] [TRAIN] Epoch=1/40, Step=40/66, loss=0.097227, lr=0.009867, time_each_step=0.07s, eta=0:3:42 2020-05-09 09:46:40 [INFO] [TRAIN] Epoch=1/40, Step=42/66, loss=0.125341, lr=0.00986, time_each_step=0.07s, eta=0:3:39 2020-05-09 09:46:40 [INFO] [TRAIN] Epoch=1/40, Step=44/66, loss=0.056299, lr=0.009853, time_each_step=0.07s, eta=0:3:38 2020-05-09 09:46:40 [INFO] [TRAIN] Epoch=1/40, Step=46/66, loss=0.068326, lr=0.009846, time_each_step=0.07s, eta=0:3:38 2020-05-09 09:46:40 [INFO] [TRAIN] Epoch=1/40, Step=48/66, loss=0.059824, lr=0.00984, time_each_step=0.07s, eta=0:3:38 2020-05-09 09:46:40 [INFO] [TRAIN] Epoch=1/40, Step=50/66, loss=0.070903, lr=0.009833, time_each_step=0.07s, eta=0:3:38 2020-05-09 09:46:41 [INFO] [TRAIN] Epoch=1/40, Step=52/66, loss=0.070057, lr=0.009826, time_each_step=0.07s, eta=0:3:38 2020-05-09 09:46:41 [INFO] [TRAIN] Epoch=1/40, Step=54/66, loss=0.044742, lr=0.009819, time_each_step=0.07s, eta=0:3:39 2020-05-09 09:46:41 [INFO] [TRAIN] Epoch=1/40, Step=56/66, loss=0.046969, lr=0.009812, time_each_step=0.07s, eta=0:3:38 2020-05-09 09:46:41 [INFO] [TRAIN] Epoch=1/40, Step=58/66, loss=0.037851, lr=0.009805, time_each_step=0.07s, eta=0:3:39 2020-05-09 09:46:41 [INFO] [TRAIN] Epoch=1/40, Step=60/66, loss=0.046815, lr=0.009799, time_each_step=0.07s, eta=0:3:39 2020-05-09 09:46:41 [INFO] [TRAIN] Epoch=1/40, Step=62/66, loss=0.047356, lr=0.009792, time_each_step=0.07s, eta=0:3:39 2020-05-09 09:46:41 [INFO] [TRAIN] Epoch=1/40, Step=64/66, loss=0.050003, lr=0.009785, time_each_step=0.07s, eta=0:3:40 2020-05-09 09:46:42 [INFO] [TRAIN] Epoch=1/40, Step=66/66, loss=0.040461, lr=0.009778, time_each_step=0.07s, eta=0:3:40 2020-05-09 09:46:42 [INFO] [TRAIN] Epoch 1 finished, loss=0.123195, lr=0.009889 . 2020-05-09 09:46:42 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
0%| | 0/19 [00:00<?, ?it/s]share_vars_from is set, scope is ignored. 100%|██████████| 19/19 [00:03<00:00, 5.46it/s]
2020-05-09 09:46:45 [INFO] [EVAL] Finished, Epoch=1, miou=0.490853, category_iou=[0.98170577 0. ], macc=0.981706, category_acc=[0.98170577 0. ], kappa=0.0 . 2020-05-09 09:46:45 [INFO] Model saved in output/deeplab/best_model. 2020-05-09 09:46:46 [INFO] Model saved in output/deeplab/epoch_1. 2020-05-09 09:46:46 [INFO] Current evaluated best model in eval_dataset is epoch_1, miou=0.4908528829875745 2020-05-09 09:46:48 [INFO] [TRAIN] Epoch=2/40, Step=2/66, loss=0.048643, lr=0.009771, time_each_step=0.17s, eta=0:8:43 2020-05-09 09:46:48 [INFO] [TRAIN] Epoch=2/40, Step=4/66, loss=0.065821, lr=0.009764, time_each_step=0.18s, eta=0:8:43 2020-05-09 09:46:48 [INFO] [TRAIN] Epoch=2/40, Step=6/66, loss=0.033492, lr=0.009758, time_each_step=0.18s, eta=0:8:43 2020-05-09 09:46:49 [INFO] [TRAIN] Epoch=2/40, Step=8/66, loss=0.054411, lr=0.009751, time_each_step=0.18s, eta=0:8:42 2020-05-09 09:46:49 [INFO] [TRAIN] Epoch=2/40, Step=10/66, loss=0.055043, lr=0.009744, time_each_step=0.18s, eta=0:8:42 2020-05-09 09:46:49 [INFO] [TRAIN] Epoch=2/40, Step=12/66, loss=0.044917, lr=0.009737, time_each_step=0.19s, eta=0:8:42 2020-05-09 09:46:49 [INFO] [TRAIN] Epoch=2/40, Step=14/66, loss=0.051386, lr=0.00973, time_each_step=0.19s, eta=0:8:42 2020-05-09 09:46:49 [INFO] [TRAIN] Epoch=2/40, Step=16/66, loss=0.037487, lr=0.009723, time_each_step=0.19s, eta=0:8:41 2020-05-09 09:46:49 [INFO] [TRAIN] Epoch=2/40, Step=18/66, loss=0.047655, lr=0.009717, time_each_step=0.2s, eta=0:8:41 2020-05-09 09:46:50 [INFO] [TRAIN] Epoch=2/40, Step=20/66, loss=0.044596, lr=0.00971, time_each_step=0.2s, eta=0:8:41 2020-05-09 09:46:50 [INFO] [TRAIN] Epoch=2/40, Step=22/66, loss=0.05101, lr=0.009703, time_each_step=0.09s, eta=0:8:36 2020-05-09 09:46:50 [INFO] [TRAIN] Epoch=2/40, Step=24/66, loss=0.027706, lr=0.009696, time_each_step=0.09s, eta=0:8:36 2020-05-09 09:46:50 [INFO] [TRAIN] Epoch=2/40, Step=26/66, loss=0.036218, lr=0.009689, time_each_step=0.09s, eta=0:8:35 2020-05-09 09:46:50 [INFO] [TRAIN] Epoch=2/40, Step=28/66, loss=0.038468, lr=0.009682, time_each_step=0.08s, eta=0:8:35 2020-05-09 09:46:50 [INFO] [TRAIN] Epoch=2/40, Step=30/66, loss=0.032189, lr=0.009676, time_each_step=0.08s, eta=0:8:35 2020-05-09 09:46:50 [INFO] [TRAIN] Epoch=2/40, Step=32/66, loss=0.037459, lr=0.009669, time_each_step=0.08s, eta=0:8:34 2020-05-09 09:46:51 [INFO] [TRAIN] Epoch=2/40, Step=34/66, loss=0.05304, lr=0.009662, time_each_step=0.07s, eta=0:8:34 2020-05-09 09:46:51 [INFO] [TRAIN] Epoch=2/40, Step=36/66, loss=0.05954, lr=0.009655, time_each_step=0.07s, eta=0:8:34 2020-05-09 09:46:51 [INFO] [TRAIN] Epoch=2/40, Step=38/66, loss=0.052924, lr=0.009648, time_each_step=0.07s, eta=0:8:34 2020-05-09 09:46:51 [INFO] [TRAIN] Epoch=2/40, Step=40/66, loss=0.03714, lr=0.009641, time_each_step=0.07s, eta=0:8:33 2020-05-09 09:46:51 [INFO] [TRAIN] Epoch=2/40, Step=42/66, loss=0.045424, lr=0.009634, time_each_step=0.07s, eta=0:8:33 2020-05-09 09:46:51 [INFO] [TRAIN] Epoch=2/40, Step=44/66, loss=0.022456, lr=0.009628, time_each_step=0.07s, eta=0:8:33 2020-05-09 09:46:51 [INFO] [TRAIN] Epoch=2/40, Step=46/66, loss=0.02337, lr=0.009621, time_each_step=0.07s, eta=0:8:33 2020-05-09 09:46:52 [INFO] [TRAIN] Epoch=2/40, Step=48/66, loss=0.020118, lr=0.009614, time_each_step=0.07s, eta=0:8:33 2020-05-09 09:46:52 [INFO] [TRAIN] Epoch=2/40, Step=50/66, loss=0.041169, lr=0.009607, time_each_step=0.07s, eta=0:8:33 2020-05-09 09:46:52 [INFO] [TRAIN] Epoch=2/40, Step=52/66, loss=0.032431, lr=0.0096, time_each_step=0.07s, eta=0:8:33 2020-05-09 09:46:52 [INFO] [TRAIN] Epoch=2/40, Step=54/66, loss=0.030603, lr=0.009593, time_each_step=0.07s, eta=0:8:33 2020-05-09 09:46:52 [INFO] [TRAIN] Epoch=2/40, Step=56/66, loss=0.037192, lr=0.009587, time_each_step=0.07s, eta=0:8:32 2020-05-09 09:46:52 [INFO] [TRAIN] Epoch=2/40, Step=58/66, loss=0.065011, lr=0.00958, time_each_step=0.07s, eta=0:8:32 2020-05-09 09:46:52 [INFO] [TRAIN] Epoch=2/40, Step=60/66, loss=0.044167, lr=0.009573, time_each_step=0.07s, eta=0:8:32 2020-05-09 09:46:52 [INFO] [TRAIN] Epoch=2/40, Step=62/66, loss=0.043026, lr=0.009566, time_each_step=0.07s, eta=0:8:32 2020-05-09 09:46:53 [INFO] [TRAIN] Epoch=2/40, Step=64/66, loss=0.026524, lr=0.009559, time_each_step=0.07s, eta=0:8:32 2020-05-09 09:46:53 [INFO] [TRAIN] Epoch=2/40, Step=66/66, loss=0.027805, lr=0.009552, time_each_step=0.07s, eta=0:8:32 2020-05-09 09:46:53 [INFO] [TRAIN] Epoch 2 finished, loss=0.042302, lr=0.009664 . 2020-05-09 09:46:53 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:03<00:00, 5.01it/s]
2020-05-09 09:53:06 [INFO] [EVAL] Finished, Epoch=29, miou=0.883191, category_iou=[0.99496381 0.77141844], macc=0.995048, category_acc=[0.99838548 0.83219193], kappa=0.868442 . 2020-05-09 09:53:06 [INFO] Model saved in output/deeplab/epoch_29. 2020-05-09 09:53:06 [INFO] Current evaluated best model in eval_dataset is epoch_26, miou=0.8850852681815866 2020-05-09 09:53:11 [INFO] [TRAIN] Epoch=30/40, Step=2/66, loss=0.006863, lr=0.003125, time_each_step=0.27s, eta=0:2:17 2020-05-09 09:53:11 [INFO] [TRAIN] Epoch=30/40, Step=4/66, loss=0.009762, lr=0.003117, time_each_step=0.27s, eta=0:2:17 2020-05-09 09:53:11 [INFO] [TRAIN] Epoch=30/40, Step=6/66, loss=0.009979, lr=0.00311, time_each_step=0.27s, eta=0:2:16 2020-05-09 09:53:11 [INFO] [TRAIN] Epoch=30/40, Step=8/66, loss=0.00579, lr=0.003102, time_each_step=0.28s, eta=0:2:16 2020-05-09 09:53:11 [INFO] [TRAIN] Epoch=30/40, Step=10/66, loss=0.009733, lr=0.003094, time_each_step=0.28s, eta=0:2:15 2020-05-09 09:53:11 [INFO] [TRAIN] Epoch=30/40, Step=12/66, loss=0.005823, lr=0.003086, time_each_step=0.28s, eta=0:2:15 2020-05-09 09:53:12 [INFO] [TRAIN] Epoch=30/40, Step=14/66, loss=0.011094, lr=0.003078, time_each_step=0.28s, eta=0:2:14 2020-05-09 09:53:12 [INFO] [TRAIN] Epoch=30/40, Step=16/66, loss=0.009109, lr=0.003071, time_each_step=0.29s, eta=0:2:14 2020-05-09 09:53:12 [INFO] [TRAIN] Epoch=30/40, Step=18/66, loss=0.008076, lr=0.003063, time_each_step=0.29s, eta=0:2:14 2020-05-09 09:53:12 [INFO] [TRAIN] Epoch=30/40, Step=20/66, loss=0.012225, lr=0.003055, time_each_step=0.29s, eta=0:2:13 2020-05-09 09:53:12 [INFO] [TRAIN] Epoch=30/40, Step=22/66, loss=0.008668, lr=0.003047, time_each_step=0.09s, eta=0:2:4 2020-05-09 09:53:12 [INFO] [TRAIN] Epoch=30/40, Step=24/66, loss=0.007403, lr=0.00304, time_each_step=0.09s, eta=0:2:3 2020-05-09 09:53:13 [INFO] [TRAIN] Epoch=30/40, Step=26/66, loss=0.006372, lr=0.003032, time_each_step=0.08s, eta=0:2:3 2020-05-09 09:53:13 [INFO] [TRAIN] Epoch=30/40, Step=28/66, loss=0.016677, lr=0.003024, time_each_step=0.08s, eta=0:2:3 2020-05-09 09:53:13 [INFO] [TRAIN] Epoch=30/40, Step=30/66, loss=0.010489, lr=0.003016, time_each_step=0.08s, eta=0:2:3 2020-05-09 09:53:13 [INFO] [TRAIN] Epoch=30/40, Step=32/66, loss=0.006755, lr=0.003008, time_each_step=0.08s, eta=0:2:2 2020-05-09 09:53:13 [INFO] [TRAIN] Epoch=30/40, Step=34/66, loss=0.006045, lr=0.003001, time_each_step=0.07s, eta=0:2:2 2020-05-09 09:53:13 [INFO] [TRAIN] Epoch=30/40, Step=36/66, loss=0.007014, lr=0.002993, time_each_step=0.07s, eta=0:2:2 2020-05-09 09:53:13 [INFO] [TRAIN] Epoch=30/40, Step=38/66, loss=0.005759, lr=0.002985, time_each_step=0.07s, eta=0:2:2 2020-05-09 09:53:14 [INFO] [TRAIN] Epoch=30/40, Step=40/66, loss=0.004762, lr=0.002977, time_each_step=0.07s, eta=0:2:1 2020-05-09 09:53:14 [INFO] [TRAIN] Epoch=30/40, Step=42/66, loss=0.005977, lr=0.002969, time_each_step=0.07s, eta=0:2:1 2020-05-09 09:53:14 [INFO] [TRAIN] Epoch=30/40, Step=44/66, loss=0.007629, lr=0.002962, time_each_step=0.07s, eta=0:2:1 2020-05-09 09:53:14 [INFO] [TRAIN] Epoch=30/40, Step=46/66, loss=0.007897, lr=0.002954, time_each_step=0.07s, eta=0:2:1 2020-05-09 09:53:14 [INFO] [TRAIN] Epoch=30/40, Step=48/66, loss=0.009535, lr=0.002946, time_each_step=0.07s, eta=0:2:1 2020-05-09 09:53:14 [INFO] [TRAIN] Epoch=30/40, Step=50/66, loss=0.008363, lr=0.002938, time_each_step=0.07s, eta=0:2:1 2020-05-09 09:53:14 [INFO] [TRAIN] Epoch=30/40, Step=52/66, loss=0.008497, lr=0.00293, time_each_step=0.07s, eta=0:2:1 2020-05-09 09:53:14 [INFO] [TRAIN] Epoch=30/40, Step=54/66, loss=0.007527, lr=0.002923, time_each_step=0.07s, eta=0:2:0 2020-05-09 09:53:15 [INFO] [TRAIN] Epoch=30/40, Step=56/66, loss=0.008486, lr=0.002915, time_each_step=0.07s, eta=0:2:0 2020-05-09 09:53:15 [INFO] [TRAIN] Epoch=30/40, Step=58/66, loss=0.007435, lr=0.002907, time_each_step=0.07s, eta=0:2:0 2020-05-09 09:53:15 [INFO] [TRAIN] Epoch=30/40, Step=60/66, loss=0.005084, lr=0.002899, time_each_step=0.07s, eta=0:2:0 2020-05-09 09:53:15 [INFO] [TRAIN] Epoch=30/40, Step=62/66, loss=0.008406, lr=0.002891, time_each_step=0.07s, eta=0:2:0 2020-05-09 09:53:15 [INFO] [TRAIN] Epoch=30/40, Step=64/66, loss=0.007653, lr=0.002883, time_each_step=0.06s, eta=0:2:0 2020-05-09 09:53:15 [INFO] [TRAIN] Epoch=30/40, Step=66/66, loss=0.005077, lr=0.002876, time_each_step=0.06s, eta=0:2:0 2020-05-09 09:53:15 [INFO] [TRAIN] Epoch 30 finished, loss=0.008108, lr=0.003003 . 2020-05-09 09:53:15 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:05<00:00, 3.20it/s]
2020-05-09 09:53:21 [INFO] [EVAL] Finished, Epoch=30, miou=0.886507, category_iou=[0.99522117 0.77779254], macc=0.9953, category_acc=[0.99812163 0.8519899 ], kappa=0.872616 . 2020-05-09 09:53:22 [INFO] Model saved in output/deeplab/best_model. 2020-05-09 09:53:22 [INFO] Model saved in output/deeplab/epoch_30. 2020-05-09 09:53:22 [INFO] Current evaluated best model in eval_dataset is epoch_30, miou=0.8865068559163839 2020-05-09 09:53:25 [INFO] [TRAIN] Epoch=31/40, Step=2/66, loss=0.00641, lr=0.002868, time_each_step=0.18s, eta=0:2:39 2020-05-09 09:53:25 [INFO] [TRAIN] Epoch=31/40, Step=4/66, loss=0.009021, lr=0.00286, time_each_step=0.19s, eta=0:2:38 2020-05-09 09:53:25 [INFO] [TRAIN] Epoch=31/40, Step=6/66, loss=0.005458, lr=0.002852, time_each_step=0.19s, eta=0:2:38 2020-05-09 09:53:25 [INFO] [TRAIN] Epoch=31/40, Step=8/66, loss=0.008455, lr=0.002844, time_each_step=0.19s, eta=0:2:38 2020-05-09 09:53:25 [INFO] [TRAIN] Epoch=31/40, Step=10/66, loss=0.007805, lr=0.002836, time_each_step=0.19s, eta=0:2:38 2020-05-09 09:53:25 [INFO] [TRAIN] Epoch=31/40, Step=12/66, loss=0.008099, lr=0.002829, time_each_step=0.2s, eta=0:2:38 2020-05-09 09:53:26 [INFO] [TRAIN] Epoch=31/40, Step=14/66, loss=0.007417, lr=0.002821, time_each_step=0.2s, eta=0:2:37 2020-05-09 09:53:26 [INFO] [TRAIN] Epoch=31/40, Step=16/66, loss=0.00497, lr=0.002813, time_each_step=0.2s, eta=0:2:37 2020-05-09 09:53:26 [INFO] [TRAIN] Epoch=31/40, Step=18/66, loss=0.008049, lr=0.002805, time_each_step=0.2s, eta=0:2:37 2020-05-09 09:53:26 [INFO] [TRAIN] Epoch=31/40, Step=20/66, loss=0.00691, lr=0.002797, time_each_step=0.21s, eta=0:2:36 2020-05-09 09:53:26 [INFO] [TRAIN] Epoch=31/40, Step=22/66, loss=0.008314, lr=0.002789, time_each_step=0.09s, eta=0:2:31 2020-05-09 09:53:26 [INFO] [TRAIN] Epoch=31/40, Step=24/66, loss=0.007918, lr=0.002782, time_each_step=0.09s, eta=0:2:31 2020-05-09 09:53:27 [INFO] [TRAIN] Epoch=31/40, Step=26/66, loss=0.008154, lr=0.002774, time_each_step=0.09s, eta=0:2:30 2020-05-09 09:53:27 [INFO] [TRAIN] Epoch=31/40, Step=28/66, loss=0.005817, lr=0.002766, time_each_step=0.08s, eta=0:2:30 2020-05-09 09:53:27 [INFO] [TRAIN] Epoch=31/40, Step=30/66, loss=0.01673, lr=0.002758, time_each_step=0.08s, eta=0:2:30 2020-05-09 09:53:27 [INFO] [TRAIN] Epoch=31/40, Step=32/66, loss=0.009255, lr=0.00275, time_each_step=0.08s, eta=0:2:30 2020-05-09 09:53:27 [INFO] [TRAIN] Epoch=31/40, Step=34/66, loss=0.011975, lr=0.002742, time_each_step=0.08s, eta=0:2:29 2020-05-09 09:53:27 [INFO] [TRAIN] Epoch=31/40, Step=36/66, loss=0.006252, lr=0.002734, time_each_step=0.08s, eta=0:2:29 2020-05-09 09:53:27 [INFO] [TRAIN] Epoch=31/40, Step=38/66, loss=0.010236, lr=0.002726, time_each_step=0.07s, eta=0:2:29 2020-05-09 09:53:28 [INFO] [TRAIN] Epoch=31/40, Step=40/66, loss=0.006026, lr=0.002719, time_each_step=0.07s, eta=0:2:29 2020-05-09 09:53:28 [INFO] [TRAIN] Epoch=31/40, Step=42/66, loss=0.009318, lr=0.002711, time_each_step=0.07s, eta=0:2:29 2020-05-09 09:53:28 [INFO] [TRAIN] Epoch=31/40, Step=44/66, loss=0.005679, lr=0.002703, time_each_step=0.07s, eta=0:2:28 2020-05-09 09:53:28 [INFO] [TRAIN] Epoch=31/40, Step=46/66, loss=0.005522, lr=0.002695, time_each_step=0.07s, eta=0:2:28 2020-05-09 09:53:28 [INFO] [TRAIN] Epoch=31/40, Step=48/66, loss=0.009917, lr=0.002687, time_each_step=0.07s, eta=0:2:28 2020-05-09 09:53:28 [INFO] [TRAIN] Epoch=31/40, Step=50/66, loss=0.009573, lr=0.002679, time_each_step=0.07s, eta=0:2:28 2020-05-09 09:53:28 [INFO] [TRAIN] Epoch=31/40, Step=52/66, loss=0.011337, lr=0.002671, time_each_step=0.07s, eta=0:2:28 2020-05-09 09:53:29 [INFO] [TRAIN] Epoch=31/40, Step=54/66, loss=0.007056, lr=0.002663, time_each_step=0.07s, eta=0:2:28 2020-05-09 09:53:29 [INFO] [TRAIN] Epoch=31/40, Step=56/66, loss=0.00622, lr=0.002655, time_each_step=0.07s, eta=0:2:28 2020-05-09 09:53:29 [INFO] [TRAIN] Epoch=31/40, Step=58/66, loss=0.007693, lr=0.002648, time_each_step=0.07s, eta=0:2:28 2020-05-09 09:53:29 [INFO] [TRAIN] Epoch=31/40, Step=60/66, loss=0.006469, lr=0.00264, time_each_step=0.07s, eta=0:2:27 2020-05-09 09:53:29 [INFO] [TRAIN] Epoch=31/40, Step=62/66, loss=0.006688, lr=0.002632, time_each_step=0.07s, eta=0:2:27 2020-05-09 09:53:29 [INFO] [TRAIN] Epoch=31/40, Step=64/66, loss=0.007176, lr=0.002624, time_each_step=0.07s, eta=0:2:27 2020-05-09 09:53:29 [INFO] [TRAIN] Epoch=31/40, Step=66/66, loss=0.008351, lr=0.002616, time_each_step=0.07s, eta=0:2:27 2020-05-09 09:53:29 [INFO] [TRAIN] Epoch 31 finished, loss=0.007701, lr=0.002744 . 2020-05-09 09:53:29 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:04<00:00, 4.49it/s]
2020-05-09 09:53:34 [INFO] [EVAL] Finished, Epoch=31, miou=0.879857, category_iou=[0.99485834 0.7648558 ], macc=0.994943, category_acc=[0.99811799 0.83662956], kappa=0.864189 . 2020-05-09 09:53:34 [INFO] Model saved in output/deeplab/epoch_31. 2020-05-09 09:53:34 [INFO] Current evaluated best model in eval_dataset is epoch_30, miou=0.8865068559163839 2020-05-09 09:53:36 [INFO] [TRAIN] Epoch=32/40, Step=2/66, loss=0.006874, lr=0.002608, time_each_step=0.18s, eta=0:1:51 2020-05-09 09:53:37 [INFO] [TRAIN] Epoch=32/40, Step=4/66, loss=0.010941, lr=0.0026, time_each_step=0.18s, eta=0:1:50 2020-05-09 09:53:37 [INFO] [TRAIN] Epoch=32/40, Step=6/66, loss=0.007639, lr=0.002592, time_each_step=0.19s, eta=0:1:50 2020-05-09 09:53:37 [INFO] [TRAIN] Epoch=32/40, Step=8/66, loss=0.006818, lr=0.002584, time_each_step=0.19s, eta=0:1:50 2020-05-09 09:53:37 [INFO] [TRAIN] Epoch=32/40, Step=10/66, loss=0.005039, lr=0.002576, time_each_step=0.19s, eta=0:1:50 2020-05-09 09:53:37 [INFO] [TRAIN] Epoch=32/40, Step=12/66, loss=0.010708, lr=0.002568, time_each_step=0.2s, eta=0:1:50 2020-05-09 09:53:37 [INFO] [TRAIN] Epoch=32/40, Step=14/66, loss=0.00596, lr=0.00256, time_each_step=0.2s, eta=0:1:49 2020-05-09 09:53:38 [INFO] [TRAIN] Epoch=32/40, Step=16/66, loss=0.006019, lr=0.002553, time_each_step=0.2s, eta=0:1:49 2020-05-09 09:53:38 [INFO] [TRAIN] Epoch=32/40, Step=18/66, loss=0.007748, lr=0.002545, time_each_step=0.2s, eta=0:1:49 2020-05-09 09:53:38 [INFO] [TRAIN] Epoch=32/40, Step=20/66, loss=0.008574, lr=0.002537, time_each_step=0.21s, eta=0:1:49 2020-05-09 09:53:38 [INFO] [TRAIN] Epoch=32/40, Step=22/66, loss=0.011029, lr=0.002529, time_each_step=0.09s, eta=0:1:43 2020-05-09 09:53:38 [INFO] [TRAIN] Epoch=32/40, Step=24/66, loss=0.007234, lr=0.002521, time_each_step=0.09s, eta=0:1:43 2020-05-09 09:53:38 [INFO] [TRAIN] Epoch=32/40, Step=26/66, loss=0.014135, lr=0.002513, time_each_step=0.09s, eta=0:1:42 2020-05-09 09:53:39 [INFO] [TRAIN] Epoch=32/40, Step=28/66, loss=0.005983, lr=0.002505, time_each_step=0.08s, eta=0:1:42 2020-05-09 09:53:39 [INFO] [TRAIN] Epoch=32/40, Step=30/66, loss=0.005911, lr=0.002497, time_each_step=0.08s, eta=0:1:42 2020-05-09 09:53:39 [INFO] [TRAIN] Epoch=32/40, Step=32/66, loss=0.006717, lr=0.002489, time_each_step=0.08s, eta=0:1:42 2020-05-09 09:53:39 [INFO] [TRAIN] Epoch=32/40, Step=34/66, loss=0.014635, lr=0.002481, time_each_step=0.08s, eta=0:1:41 2020-05-09 09:53:39 [INFO] [TRAIN] Epoch=32/40, Step=36/66, loss=0.007887, lr=0.002473, time_each_step=0.07s, eta=0:1:41 2020-05-09 09:53:39 [INFO] [TRAIN] Epoch=32/40, Step=38/66, loss=0.006465, lr=0.002465, time_each_step=0.07s, eta=0:1:41 2020-05-09 09:53:39 [INFO] [TRAIN] Epoch=32/40, Step=40/66, loss=0.008387, lr=0.002457, time_each_step=0.07s, eta=0:1:41 2020-05-09 09:53:40 [INFO] [TRAIN] Epoch=32/40, Step=42/66, loss=0.005994, lr=0.002449, time_each_step=0.07s, eta=0:1:41 2020-05-09 09:53:40 [INFO] [TRAIN] Epoch=32/40, Step=44/66, loss=0.006445, lr=0.002441, time_each_step=0.07s, eta=0:1:40 2020-05-09 09:53:40 [INFO] [TRAIN] Epoch=32/40, Step=46/66, loss=0.009155, lr=0.002433, time_each_step=0.07s, eta=0:1:40 2020-05-09 09:53:40 [INFO] [TRAIN] Epoch=32/40, Step=48/66, loss=0.006972, lr=0.002425, time_each_step=0.07s, eta=0:1:40 2020-05-09 09:53:40 [INFO] [TRAIN] Epoch=32/40, Step=50/66, loss=0.006389, lr=0.002417, time_each_step=0.07s, eta=0:1:40 2020-05-09 09:53:40 [INFO] [TRAIN] Epoch=32/40, Step=52/66, loss=0.007098, lr=0.002409, time_each_step=0.07s, eta=0:1:40 2020-05-09 09:53:40 [INFO] [TRAIN] Epoch=32/40, Step=54/66, loss=0.010058, lr=0.002401, time_each_step=0.07s, eta=0:1:40 2020-05-09 09:53:40 [INFO] [TRAIN] Epoch=32/40, Step=56/66, loss=0.00447, lr=0.002393, time_each_step=0.07s, eta=0:1:40 2020-05-09 09:53:41 [INFO] [TRAIN] Epoch=32/40, Step=58/66, loss=0.00585, lr=0.002385, time_each_step=0.07s, eta=0:1:40 2020-05-09 09:53:41 [INFO] [TRAIN] Epoch=32/40, Step=60/66, loss=0.00625, lr=0.002377, time_each_step=0.07s, eta=0:1:39 2020-05-09 09:53:41 [INFO] [TRAIN] Epoch=32/40, Step=62/66, loss=0.007664, lr=0.002369, time_each_step=0.08s, eta=0:1:39 2020-05-09 09:53:41 [INFO] [TRAIN] Epoch=32/40, Step=64/66, loss=0.01015, lr=0.002361, time_each_step=0.08s, eta=0:1:39 2020-05-09 09:53:42 [INFO] [TRAIN] Epoch=32/40, Step=66/66, loss=0.004812, lr=0.002353, time_each_step=0.09s, eta=0:1:39 2020-05-09 09:53:42 [INFO] [TRAIN] Epoch 32 finished, loss=0.007734, lr=0.002483 . 2020-05-09 09:53:42 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:08<00:00, 2.16it/s]
2020-05-09 09:53:50 [INFO] [EVAL] Finished, Epoch=32, miou=0.883541, category_iou=[0.99503955 0.77204253], macc=0.995122, category_acc=[0.9981927 0.84173125], kappa=0.868876 . 2020-05-09 09:53:51 [INFO] Model saved in output/deeplab/epoch_32. 2020-05-09 09:53:51 [INFO] Current evaluated best model in eval_dataset is epoch_30, miou=0.8865068559163839 2020-05-09 09:53:55 [INFO] [TRAIN] Epoch=33/40, Step=2/66, loss=0.004894, lr=0.002345, time_each_step=0.27s, eta=0:2:25 2020-05-09 09:53:55 [INFO] [TRAIN] Epoch=33/40, Step=4/66, loss=0.009758, lr=0.002337, time_each_step=0.27s, eta=0:2:24 2020-05-09 09:53:55 [INFO] [TRAIN] Epoch=33/40, Step=6/66, loss=0.007474, lr=0.002329, time_each_step=0.28s, eta=0:2:24 2020-05-09 09:53:55 [INFO] [TRAIN] Epoch=33/40, Step=8/66, loss=0.007161, lr=0.002321, time_each_step=0.28s, eta=0:2:24 2020-05-09 09:53:55 [INFO] [TRAIN] Epoch=33/40, Step=10/66, loss=0.012161, lr=0.002313, time_each_step=0.28s, eta=0:2:23 2020-05-09 09:53:56 [INFO] [TRAIN] Epoch=33/40, Step=12/66, loss=0.007252, lr=0.002305, time_each_step=0.28s, eta=0:2:23 2020-05-09 09:53:56 [INFO] [TRAIN] Epoch=33/40, Step=14/66, loss=0.00548, lr=0.002297, time_each_step=0.28s, eta=0:2:22 2020-05-09 09:53:56 [INFO] [TRAIN] Epoch=33/40, Step=16/66, loss=0.007685, lr=0.002289, time_each_step=0.28s, eta=0:2:21 2020-05-09 09:53:56 [INFO] [TRAIN] Epoch=33/40, Step=18/66, loss=0.009079, lr=0.002281, time_each_step=0.27s, eta=0:2:21 2020-05-09 09:53:56 [INFO] [TRAIN] Epoch=33/40, Step=20/66, loss=0.007093, lr=0.002273, time_each_step=0.27s, eta=0:2:20 2020-05-09 09:53:56 [INFO] [TRAIN] Epoch=33/40, Step=22/66, loss=0.007217, lr=0.002265, time_each_step=0.09s, eta=0:2:12 2020-05-09 09:53:57 [INFO] [TRAIN] Epoch=33/40, Step=24/66, loss=0.00495, lr=0.002257, time_each_step=0.09s, eta=0:2:11 2020-05-09 09:53:57 [INFO] [TRAIN] Epoch=33/40, Step=26/66, loss=0.005274, lr=0.002249, time_each_step=0.09s, eta=0:2:11 2020-05-09 09:53:57 [INFO] [TRAIN] Epoch=33/40, Step=28/66, loss=0.006218, lr=0.002241, time_each_step=0.09s, eta=0:2:11 2020-05-09 09:53:57 [INFO] [TRAIN] Epoch=33/40, Step=30/66, loss=0.006219, lr=0.002233, time_each_step=0.08s, eta=0:2:10 2020-05-09 09:53:57 [INFO] [TRAIN] Epoch=33/40, Step=32/66, loss=0.004733, lr=0.002225, time_each_step=0.08s, eta=0:2:10 2020-05-09 09:53:57 [INFO] [TRAIN] Epoch=33/40, Step=34/66, loss=0.004785, lr=0.002217, time_each_step=0.08s, eta=0:2:10 2020-05-09 09:53:57 [INFO] [TRAIN] Epoch=33/40, Step=36/66, loss=0.01833, lr=0.002209, time_each_step=0.08s, eta=0:2:10 2020-05-09 09:53:58 [INFO] [TRAIN] Epoch=33/40, Step=38/66, loss=0.015783, lr=0.002201, time_each_step=0.07s, eta=0:2:10 2020-05-09 09:53:58 [INFO] [TRAIN] Epoch=33/40, Step=40/66, loss=0.007433, lr=0.002192, time_each_step=0.07s, eta=0:2:9 2020-05-09 09:53:58 [INFO] [TRAIN] Epoch=33/40, Step=42/66, loss=0.006712, lr=0.002184, time_each_step=0.07s, eta=0:2:9 2020-05-09 09:53:58 [INFO] [TRAIN] Epoch=33/40, Step=44/66, loss=0.006075, lr=0.002176, time_each_step=0.07s, eta=0:2:9 2020-05-09 09:53:58 [INFO] [TRAIN] Epoch=33/40, Step=46/66, loss=0.009223, lr=0.002168, time_each_step=0.07s, eta=0:2:9 2020-05-09 09:53:58 [INFO] [TRAIN] Epoch=33/40, Step=48/66, loss=0.005694, lr=0.00216, time_each_step=0.07s, eta=0:2:9 2020-05-09 09:53:58 [INFO] [TRAIN] Epoch=33/40, Step=50/66, loss=0.009911, lr=0.002152, time_each_step=0.07s, eta=0:2:9 2020-05-09 09:53:58 [INFO] [TRAIN] Epoch=33/40, Step=52/66, loss=0.006196, lr=0.002144, time_each_step=0.07s, eta=0:2:8 2020-05-09 09:53:59 [INFO] [TRAIN] Epoch=33/40, Step=54/66, loss=0.011217, lr=0.002136, time_each_step=0.07s, eta=0:2:8 2020-05-09 09:53:59 [INFO] [TRAIN] Epoch=33/40, Step=56/66, loss=0.012932, lr=0.002128, time_each_step=0.07s, eta=0:2:8 2020-05-09 09:53:59 [INFO] [TRAIN] Epoch=33/40, Step=58/66, loss=0.007159, lr=0.00212, time_each_step=0.07s, eta=0:2:8 2020-05-09 09:53:59 [INFO] [TRAIN] Epoch=33/40, Step=60/66, loss=0.005045, lr=0.002112, time_each_step=0.07s, eta=0:2:8 2020-05-09 09:53:59 [INFO] [TRAIN] Epoch=33/40, Step=62/66, loss=0.016721, lr=0.002103, time_each_step=0.07s, eta=0:2:8 2020-05-09 09:53:59 [INFO] [TRAIN] Epoch=33/40, Step=64/66, loss=0.006069, lr=0.002095, time_each_step=0.07s, eta=0:2:8 2020-05-09 09:53:59 [INFO] [TRAIN] Epoch=33/40, Step=66/66, loss=0.006665, lr=0.002087, time_each_step=0.07s, eta=0:2:8 2020-05-09 09:53:59 [INFO] [TRAIN] Epoch 33 finished, loss=0.008054, lr=0.002219 . 2020-05-09 09:53:59 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:04<00:00, 4.67it/s]
2020-05-09 09:54:03 [INFO] [EVAL] Finished, Epoch=33, miou=0.884283, category_iou=[0.99501401 0.77355201], macc=0.995098, category_acc=[0.99842105 0.83308989], kappa=0.869826 . 2020-05-09 09:54:04 [INFO] Model saved in output/deeplab/epoch_33. 2020-05-09 09:54:04 [INFO] Current evaluated best model in eval_dataset is epoch_30, miou=0.8865068559163839 2020-05-09 09:54:06 [INFO] [TRAIN] Epoch=34/40, Step=2/66, loss=0.005922, lr=0.002079, time_each_step=0.19s, eta=0:1:33 2020-05-09 09:54:06 [INFO] [TRAIN] Epoch=34/40, Step=4/66, loss=0.007101, lr=0.002071, time_each_step=0.19s, eta=0:1:33 2020-05-09 09:54:07 [INFO] [TRAIN] Epoch=34/40, Step=6/66, loss=0.015968, lr=0.002063, time_each_step=0.19s, eta=0:1:33 2020-05-09 09:54:07 [INFO] [TRAIN] Epoch=34/40, Step=8/66, loss=0.005625, lr=0.002055, time_each_step=0.19s, eta=0:1:33 2020-05-09 09:54:07 [INFO] [TRAIN] Epoch=34/40, Step=10/66, loss=0.016497, lr=0.002047, time_each_step=0.2s, eta=0:1:33 2020-05-09 09:54:07 [INFO] [TRAIN] Epoch=34/40, Step=12/66, loss=0.009809, lr=0.002039, time_each_step=0.2s, eta=0:1:32 2020-05-09 09:54:07 [INFO] [TRAIN] Epoch=34/40, Step=14/66, loss=0.005504, lr=0.00203, time_each_step=0.2s, eta=0:1:32 2020-05-09 09:54:08 [INFO] [TRAIN] Epoch=34/40, Step=16/66, loss=0.009717, lr=0.002022, time_each_step=0.21s, eta=0:1:32 2020-05-09 09:54:08 [INFO] [TRAIN] Epoch=34/40, Step=18/66, loss=0.004254, lr=0.002014, time_each_step=0.21s, eta=0:1:32 2020-05-09 09:54:08 [INFO] [TRAIN] Epoch=34/40, Step=20/66, loss=0.011984, lr=0.002006, time_each_step=0.21s, eta=0:1:31 2020-05-09 09:54:08 [INFO] [TRAIN] Epoch=34/40, Step=22/66, loss=0.006908, lr=0.001998, time_each_step=0.09s, eta=0:1:26 2020-05-09 09:54:08 [INFO] [TRAIN] Epoch=34/40, Step=24/66, loss=0.006556, lr=0.00199, time_each_step=0.09s, eta=0:1:25 2020-05-09 09:54:08 [INFO] [TRAIN] Epoch=34/40, Step=26/66, loss=0.004875, lr=0.001981, time_each_step=0.09s, eta=0:1:25 2020-05-09 09:54:09 [INFO] [TRAIN] Epoch=34/40, Step=28/66, loss=0.011429, lr=0.001973, time_each_step=0.09s, eta=0:1:25 2020-05-09 09:54:09 [INFO] [TRAIN] Epoch=34/40, Step=30/66, loss=0.006812, lr=0.001965, time_each_step=0.08s, eta=0:1:24 2020-05-09 09:54:09 [INFO] [TRAIN] Epoch=34/40, Step=32/66, loss=0.009232, lr=0.001957, time_each_step=0.08s, eta=0:1:24 2020-05-09 09:54:09 [INFO] [TRAIN] Epoch=34/40, Step=34/66, loss=0.014444, lr=0.001949, time_each_step=0.08s, eta=0:1:24 2020-05-09 09:54:09 [INFO] [TRAIN] Epoch=34/40, Step=36/66, loss=0.008169, lr=0.001941, time_each_step=0.07s, eta=0:1:24 2020-05-09 09:54:09 [INFO] [TRAIN] Epoch=34/40, Step=38/66, loss=0.005977, lr=0.001932, time_each_step=0.07s, eta=0:1:23 2020-05-09 09:54:09 [INFO] [TRAIN] Epoch=34/40, Step=40/66, loss=0.007715, lr=0.001924, time_each_step=0.07s, eta=0:1:23 2020-05-09 09:54:10 [INFO] [TRAIN] Epoch=34/40, Step=42/66, loss=0.007457, lr=0.001916, time_each_step=0.06s, eta=0:1:23 2020-05-09 09:54:10 [INFO] [TRAIN] Epoch=34/40, Step=44/66, loss=0.007434, lr=0.001908, time_each_step=0.06s, eta=0:1:23 2020-05-09 09:54:10 [INFO] [TRAIN] Epoch=34/40, Step=46/66, loss=0.008546, lr=0.0019, time_each_step=0.06s, eta=0:1:23 2020-05-09 09:54:10 [INFO] [TRAIN] Epoch=34/40, Step=48/66, loss=0.006409, lr=0.001891, time_each_step=0.06s, eta=0:1:23 2020-05-09 09:54:10 [INFO] [TRAIN] Epoch=34/40, Step=50/66, loss=0.006212, lr=0.001883, time_each_step=0.06s, eta=0:1:23 2020-05-09 09:54:10 [INFO] [TRAIN] Epoch=34/40, Step=52/66, loss=0.005923, lr=0.001875, time_each_step=0.06s, eta=0:1:22 2020-05-09 09:54:10 [INFO] [TRAIN] Epoch=34/40, Step=54/66, loss=0.009684, lr=0.001867, time_each_step=0.06s, eta=0:1:22 2020-05-09 09:54:10 [INFO] [TRAIN] Epoch=34/40, Step=56/66, loss=0.004945, lr=0.001859, time_each_step=0.06s, eta=0:1:22 2020-05-09 09:54:11 [INFO] [TRAIN] Epoch=34/40, Step=58/66, loss=0.009059, lr=0.00185, time_each_step=0.06s, eta=0:1:22 2020-05-09 09:54:11 [INFO] [TRAIN] Epoch=34/40, Step=60/66, loss=0.007556, lr=0.001842, time_each_step=0.06s, eta=0:1:22 2020-05-09 09:54:11 [INFO] [TRAIN] Epoch=34/40, Step=62/66, loss=0.008243, lr=0.001834, time_each_step=0.06s, eta=0:1:22 2020-05-09 09:54:11 [INFO] [TRAIN] Epoch=34/40, Step=64/66, loss=0.0063, lr=0.001826, time_each_step=0.06s, eta=0:1:22 2020-05-09 09:54:11 [INFO] [TRAIN] Epoch=34/40, Step=66/66, loss=0.012749, lr=0.001817, time_each_step=0.06s, eta=0:1:21 2020-05-09 09:54:11 [INFO] [TRAIN] Epoch 34 finished, loss=0.0079, lr=0.001951 . 2020-05-09 09:54:11 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:04<00:00, 4.60it/s]
2020-05-09 09:54:15 [INFO] [EVAL] Finished, Epoch=34, miou=0.879986, category_iou=[0.99477222 0.76520058], macc=0.99486, category_acc=[0.99842542 0.82320513], kappa=0.864371 . 2020-05-09 09:54:16 [INFO] Model saved in output/deeplab/epoch_34. 2020-05-09 09:54:16 [INFO] Current evaluated best model in eval_dataset is epoch_30, miou=0.8865068559163839 2020-05-09 09:54:18 [INFO] [TRAIN] Epoch=35/40, Step=2/66, loss=0.00738, lr=0.001809, time_each_step=0.2s, eta=0:1:15 2020-05-09 09:54:19 [INFO] [TRAIN] Epoch=35/40, Step=4/66, loss=0.007869, lr=0.001801, time_each_step=0.21s, eta=0:1:16 2020-05-09 09:54:19 [INFO] [TRAIN] Epoch=35/40, Step=6/66, loss=0.007272, lr=0.001793, time_each_step=0.22s, eta=0:1:16 2020-05-09 09:54:19 [INFO] [TRAIN] Epoch=35/40, Step=8/66, loss=0.012723, lr=0.001784, time_each_step=0.22s, eta=0:1:16 2020-05-09 09:54:20 [INFO] [TRAIN] Epoch=35/40, Step=10/66, loss=0.006159, lr=0.001776, time_each_step=0.23s, eta=0:1:16 2020-05-09 09:54:20 [INFO] [TRAIN] Epoch=35/40, Step=12/66, loss=0.010535, lr=0.001768, time_each_step=0.24s, eta=0:1:16 2020-05-09 09:54:20 [INFO] [TRAIN] Epoch=35/40, Step=14/66, loss=0.007885, lr=0.00176, time_each_step=0.25s, eta=0:1:16 2020-05-09 09:54:20 [INFO] [TRAIN] Epoch=35/40, Step=16/66, loss=0.006837, lr=0.001751, time_each_step=0.25s, eta=0:1:15 2020-05-09 09:54:21 [INFO] [TRAIN] Epoch=35/40, Step=18/66, loss=0.008022, lr=0.001743, time_each_step=0.26s, eta=0:1:15 2020-05-09 09:54:21 [INFO] [TRAIN] Epoch=35/40, Step=20/66, loss=0.016412, lr=0.001735, time_each_step=0.27s, eta=0:1:15 2020-05-09 09:54:21 [INFO] [TRAIN] Epoch=35/40, Step=22/66, loss=0.005975, lr=0.001727, time_each_step=0.14s, eta=0:1:9 2020-05-09 09:54:21 [INFO] [TRAIN] Epoch=35/40, Step=24/66, loss=0.005899, lr=0.001718, time_each_step=0.13s, eta=0:1:8 2020-05-09 09:54:22 [INFO] [TRAIN] Epoch=35/40, Step=26/66, loss=0.005336, lr=0.00171, time_each_step=0.13s, eta=0:1:8 2020-05-09 09:54:22 [INFO] [TRAIN] Epoch=35/40, Step=28/66, loss=0.014492, lr=0.001702, time_each_step=0.13s, eta=0:1:7 2020-05-09 09:54:22 [INFO] [TRAIN] Epoch=35/40, Step=30/66, loss=0.004973, lr=0.001693, time_each_step=0.12s, eta=0:1:7 2020-05-09 09:54:22 [INFO] [TRAIN] Epoch=35/40, Step=32/66, loss=0.012491, lr=0.001685, time_each_step=0.12s, eta=0:1:7 2020-05-09 09:54:22 [INFO] [TRAIN] Epoch=35/40, Step=34/66, loss=0.012449, lr=0.001677, time_each_step=0.11s, eta=0:1:6 2020-05-09 09:54:23 [INFO] [TRAIN] Epoch=35/40, Step=36/66, loss=0.005122, lr=0.001668, time_each_step=0.11s, eta=0:1:6 2020-05-09 09:54:23 [INFO] [TRAIN] Epoch=35/40, Step=38/66, loss=0.010361, lr=0.00166, time_each_step=0.1s, eta=0:1:5 2020-05-09 09:54:23 [INFO] [TRAIN] Epoch=35/40, Step=40/66, loss=0.009449, lr=0.001652, time_each_step=0.1s, eta=0:1:5 2020-05-09 09:54:23 [INFO] [TRAIN] Epoch=35/40, Step=42/66, loss=0.009477, lr=0.001643, time_each_step=0.09s, eta=0:1:5 2020-05-09 09:54:23 [INFO] [TRAIN] Epoch=35/40, Step=44/66, loss=0.006715, lr=0.001635, time_each_step=0.08s, eta=0:1:4 2020-05-09 09:54:23 [INFO] [TRAIN] Epoch=35/40, Step=46/66, loss=0.007601, lr=0.001627, time_each_step=0.08s, eta=0:1:4 2020-05-09 09:54:23 [INFO] [TRAIN] Epoch=35/40, Step=48/66, loss=0.006081, lr=0.001618, time_each_step=0.07s, eta=0:1:4 2020-05-09 09:54:23 [INFO] [TRAIN] Epoch=35/40, Step=50/66, loss=0.012736, lr=0.00161, time_each_step=0.07s, eta=0:1:4 2020-05-09 09:54:24 [INFO] [TRAIN] Epoch=35/40, Step=52/66, loss=0.007539, lr=0.001602, time_each_step=0.07s, eta=0:1:4 2020-05-09 09:54:24 [INFO] [TRAIN] Epoch=35/40, Step=54/66, loss=0.008539, lr=0.001593, time_each_step=0.07s, eta=0:1:3 2020-05-09 09:54:24 [INFO] [TRAIN] Epoch=35/40, Step=56/66, loss=0.009808, lr=0.001585, time_each_step=0.07s, eta=0:1:3 2020-05-09 09:54:24 [INFO] [TRAIN] Epoch=35/40, Step=58/66, loss=0.008837, lr=0.001577, time_each_step=0.07s, eta=0:1:3 2020-05-09 09:54:24 [INFO] [TRAIN] Epoch=35/40, Step=60/66, loss=0.004804, lr=0.001568, time_each_step=0.07s, eta=0:1:3 2020-05-09 09:54:24 [INFO] [TRAIN] Epoch=35/40, Step=62/66, loss=0.009051, lr=0.00156, time_each_step=0.07s, eta=0:1:3 2020-05-09 09:54:24 [INFO] [TRAIN] Epoch=35/40, Step=64/66, loss=0.013118, lr=0.001552, time_each_step=0.07s, eta=0:1:3 2020-05-09 09:54:24 [INFO] [TRAIN] Epoch=35/40, Step=66/66, loss=0.011463, lr=0.001543, time_each_step=0.07s, eta=0:1:3 2020-05-09 09:54:24 [INFO] [TRAIN] Epoch 35 finished, loss=0.007988, lr=0.001679 . 2020-05-09 09:54:24 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:05<00:00, 3.38it/s]
2020-05-09 09:54:30 [INFO] [EVAL] Finished, Epoch=35, miou=0.888318, category_iou=[0.9952673 0.78136893], macc=0.995346, category_acc=[0.99830544 0.84751151], kappa=0.874899 . 2020-05-09 09:54:31 [INFO] Model saved in output/deeplab/best_model. 2020-05-09 09:54:31 [INFO] Model saved in output/deeplab/epoch_35. 2020-05-09 09:54:31 [INFO] Current evaluated best model in eval_dataset is epoch_35, miou=0.8883181146731094 2020-05-09 09:54:33 [INFO] [TRAIN] Epoch=36/40, Step=2/66, loss=0.008171, lr=0.001535, time_each_step=0.17s, eta=0:1:19 2020-05-09 09:54:33 [INFO] [TRAIN] Epoch=36/40, Step=4/66, loss=0.009967, lr=0.001526, time_each_step=0.18s, eta=0:1:18 2020-05-09 09:54:34 [INFO] [TRAIN] Epoch=36/40, Step=6/66, loss=0.005522, lr=0.001518, time_each_step=0.18s, eta=0:1:18 2020-05-09 09:54:34 [INFO] [TRAIN] Epoch=36/40, Step=8/66, loss=0.012667, lr=0.00151, time_each_step=0.18s, eta=0:1:18 2020-05-09 09:54:34 [INFO] [TRAIN] Epoch=36/40, Step=10/66, loss=0.010853, lr=0.001501, time_each_step=0.19s, eta=0:1:18 2020-05-09 09:54:34 [INFO] [TRAIN] Epoch=36/40, Step=12/66, loss=0.006373, lr=0.001493, time_each_step=0.19s, eta=0:1:18 2020-05-09 09:54:34 [INFO] [TRAIN] Epoch=36/40, Step=14/66, loss=0.007385, lr=0.001484, time_each_step=0.19s, eta=0:1:17 2020-05-09 09:54:35 [INFO] [TRAIN] Epoch=36/40, Step=16/66, loss=0.009514, lr=0.001476, time_each_step=0.19s, eta=0:1:17 2020-05-09 09:54:35 [INFO] [TRAIN] Epoch=36/40, Step=18/66, loss=0.005855, lr=0.001467, time_each_step=0.2s, eta=0:1:17 2020-05-09 09:54:35 [INFO] [TRAIN] Epoch=36/40, Step=20/66, loss=0.006172, lr=0.001459, time_each_step=0.2s, eta=0:1:17 2020-05-09 09:54:35 [INFO] [TRAIN] Epoch=36/40, Step=22/66, loss=0.008429, lr=0.001451, time_each_step=0.09s, eta=0:1:12 2020-05-09 09:54:35 [INFO] [TRAIN] Epoch=36/40, Step=24/66, loss=0.005699, lr=0.001442, time_each_step=0.09s, eta=0:1:11 2020-05-09 09:54:35 [INFO] [TRAIN] Epoch=36/40, Step=26/66, loss=0.007913, lr=0.001434, time_each_step=0.09s, eta=0:1:11 2020-05-09 09:54:35 [INFO] [TRAIN] Epoch=36/40, Step=28/66, loss=0.007063, lr=0.001425, time_each_step=0.08s, eta=0:1:11 2020-05-09 09:54:36 [INFO] [TRAIN] Epoch=36/40, Step=30/66, loss=0.007089, lr=0.001417, time_each_step=0.08s, eta=0:1:10 2020-05-09 09:54:36 [INFO] [TRAIN] Epoch=36/40, Step=32/66, loss=0.007711, lr=0.001408, time_each_step=0.08s, eta=0:1:10 2020-05-09 09:54:36 [INFO] [TRAIN] Epoch=36/40, Step=34/66, loss=0.009515, lr=0.0014, time_each_step=0.08s, eta=0:1:10 2020-05-09 09:54:36 [INFO] [TRAIN] Epoch=36/40, Step=36/66, loss=0.011593, lr=0.001391, time_each_step=0.07s, eta=0:1:10 2020-05-09 09:54:36 [INFO] [TRAIN] Epoch=36/40, Step=38/66, loss=0.005091, lr=0.001383, time_each_step=0.07s, eta=0:1:9 2020-05-09 09:54:36 [INFO] [TRAIN] Epoch=36/40, Step=40/66, loss=0.006929, lr=0.001374, time_each_step=0.07s, eta=0:1:9 2020-05-09 09:54:36 [INFO] [TRAIN] Epoch=36/40, Step=42/66, loss=0.011238, lr=0.001366, time_each_step=0.07s, eta=0:1:9 2020-05-09 09:54:37 [INFO] [TRAIN] Epoch=36/40, Step=44/66, loss=0.005182, lr=0.001357, time_each_step=0.07s, eta=0:1:9 2020-05-09 09:54:37 [INFO] [TRAIN] Epoch=36/40, Step=46/66, loss=0.008578, lr=0.001349, time_each_step=0.07s, eta=0:1:9 2020-05-09 09:54:37 [INFO] [TRAIN] Epoch=36/40, Step=48/66, loss=0.009196, lr=0.00134, time_each_step=0.07s, eta=0:1:9 2020-05-09 09:54:37 [INFO] [TRAIN] Epoch=36/40, Step=50/66, loss=0.006939, lr=0.001332, time_each_step=0.07s, eta=0:1:9 2020-05-09 09:54:37 [INFO] [TRAIN] Epoch=36/40, Step=52/66, loss=0.007446, lr=0.001323, time_each_step=0.07s, eta=0:1:8 2020-05-09 09:54:37 [INFO] [TRAIN] Epoch=36/40, Step=54/66, loss=0.009595, lr=0.001315, time_each_step=0.07s, eta=0:1:8 2020-05-09 09:54:37 [INFO] [TRAIN] Epoch=36/40, Step=56/66, loss=0.006313, lr=0.001306, time_each_step=0.07s, eta=0:1:8 2020-05-09 09:54:37 [INFO] [TRAIN] Epoch=36/40, Step=58/66, loss=0.007107, lr=0.001297, time_each_step=0.07s, eta=0:1:8 2020-05-09 09:54:38 [INFO] [TRAIN] Epoch=36/40, Step=60/66, loss=0.009252, lr=0.001289, time_each_step=0.07s, eta=0:1:8 2020-05-09 09:54:38 [INFO] [TRAIN] Epoch=36/40, Step=62/66, loss=0.006922, lr=0.00128, time_each_step=0.07s, eta=0:1:8 2020-05-09 09:54:38 [INFO] [TRAIN] Epoch=36/40, Step=64/66, loss=0.007311, lr=0.001272, time_each_step=0.07s, eta=0:1:8 2020-05-09 09:54:38 [INFO] [TRAIN] Epoch=36/40, Step=66/66, loss=0.00553, lr=0.001263, time_each_step=0.07s, eta=0:1:7 2020-05-09 09:54:38 [INFO] [TRAIN] Epoch 36 finished, loss=0.007934, lr=0.001402 . 2020-05-09 09:54:38 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:04<00:00, 4.54it/s]
2020-05-09 09:54:42 [INFO] [EVAL] Finished, Epoch=36, miou=0.889046, category_iou=[0.99533927 0.78275236], macc=0.995416, category_acc=[0.9981862 0.85481696], kappa=0.875805 . 2020-05-09 09:54:43 [INFO] Model saved in output/deeplab/best_model. 2020-05-09 09:54:43 [INFO] Model saved in output/deeplab/epoch_36. 2020-05-09 09:54:43 [INFO] Current evaluated best model in eval_dataset is epoch_36, miou=0.8890458135436641 2020-05-09 09:54:45 [INFO] [TRAIN] Epoch=37/40, Step=2/66, loss=0.010882, lr=0.001255, time_each_step=0.18s, eta=0:0:52 2020-05-09 09:54:46 [INFO] [TRAIN] Epoch=37/40, Step=4/66, loss=0.006168, lr=0.001246, time_each_step=0.18s, eta=0:0:52 2020-05-09 09:54:46 [INFO] [TRAIN] Epoch=37/40, Step=6/66, loss=0.006242, lr=0.001237, time_each_step=0.18s, eta=0:0:52 2020-05-09 09:54:46 [INFO] [TRAIN] Epoch=37/40, Step=8/66, loss=0.007482, lr=0.001229, time_each_step=0.18s, eta=0:0:51 2020-05-09 09:54:46 [INFO] [TRAIN] Epoch=37/40, Step=10/66, loss=0.015322, lr=0.00122, time_each_step=0.19s, eta=0:0:51 2020-05-09 09:54:46 [INFO] [TRAIN] Epoch=37/40, Step=12/66, loss=0.00825, lr=0.001212, time_each_step=0.19s, eta=0:0:51 2020-05-09 09:54:46 [INFO] [TRAIN] Epoch=37/40, Step=14/66, loss=0.004921, lr=0.001203, time_each_step=0.19s, eta=0:0:51 2020-05-09 09:54:47 [INFO] [TRAIN] Epoch=37/40, Step=16/66, loss=0.005444, lr=0.001194, time_each_step=0.2s, eta=0:0:51 2020-05-09 09:54:47 [INFO] [TRAIN] Epoch=37/40, Step=18/66, loss=0.008519, lr=0.001186, time_each_step=0.2s, eta=0:0:50 2020-05-09 09:54:47 [INFO] [TRAIN] Epoch=37/40, Step=20/66, loss=0.004806, lr=0.001177, time_each_step=0.2s, eta=0:0:50 2020-05-09 09:54:47 [INFO] [TRAIN] Epoch=37/40, Step=22/66, loss=0.005616, lr=0.001168, time_each_step=0.09s, eta=0:0:45 2020-05-09 09:54:47 [INFO] [TRAIN] Epoch=37/40, Step=24/66, loss=0.007154, lr=0.00116, time_each_step=0.09s, eta=0:0:45 2020-05-09 09:54:47 [INFO] [TRAIN] Epoch=37/40, Step=26/66, loss=0.00528, lr=0.001151, time_each_step=0.09s, eta=0:0:44 2020-05-09 09:54:48 [INFO] [TRAIN] Epoch=37/40, Step=28/66, loss=0.005532, lr=0.001142, time_each_step=0.09s, eta=0:0:44 2020-05-09 09:54:48 [INFO] [TRAIN] Epoch=37/40, Step=30/66, loss=0.010121, lr=0.001134, time_each_step=0.08s, eta=0:0:44 2020-05-09 09:54:48 [INFO] [TRAIN] Epoch=37/40, Step=32/66, loss=0.005271, lr=0.001125, time_each_step=0.08s, eta=0:0:44 2020-05-09 09:54:48 [INFO] [TRAIN] Epoch=37/40, Step=34/66, loss=0.005299, lr=0.001116, time_each_step=0.08s, eta=0:0:43 2020-05-09 09:54:48 [INFO] [TRAIN] Epoch=37/40, Step=36/66, loss=0.0056, lr=0.001108, time_each_step=0.07s, eta=0:0:43 2020-05-09 09:54:48 [INFO] [TRAIN] Epoch=37/40, Step=38/66, loss=0.007994, lr=0.001099, time_each_step=0.07s, eta=0:0:43 2020-05-09 09:54:48 [INFO] [TRAIN] Epoch=37/40, Step=40/66, loss=0.007798, lr=0.00109, time_each_step=0.07s, eta=0:0:43 2020-05-09 09:54:49 [INFO] [TRAIN] Epoch=37/40, Step=42/66, loss=0.004817, lr=0.001082, time_each_step=0.07s, eta=0:0:42 2020-05-09 09:54:49 [INFO] [TRAIN] Epoch=37/40, Step=44/66, loss=0.005926, lr=0.001073, time_each_step=0.07s, eta=0:0:42 2020-05-09 09:54:49 [INFO] [TRAIN] Epoch=37/40, Step=46/66, loss=0.004676, lr=0.001064, time_each_step=0.07s, eta=0:0:42 2020-05-09 09:54:49 [INFO] [TRAIN] Epoch=37/40, Step=48/66, loss=0.007021, lr=0.001055, time_each_step=0.07s, eta=0:0:42 2020-05-09 09:54:49 [INFO] [TRAIN] Epoch=37/40, Step=50/66, loss=0.006022, lr=0.001047, time_each_step=0.07s, eta=0:0:42 2020-05-09 09:54:49 [INFO] [TRAIN] Epoch=37/40, Step=52/66, loss=0.00681, lr=0.001038, time_each_step=0.07s, eta=0:0:42 2020-05-09 09:54:49 [INFO] [TRAIN] Epoch=37/40, Step=54/66, loss=0.005936, lr=0.001029, time_each_step=0.07s, eta=0:0:42 2020-05-09 09:54:49 [INFO] [TRAIN] Epoch=37/40, Step=56/66, loss=0.016191, lr=0.00102, time_each_step=0.07s, eta=0:0:42 2020-05-09 09:54:50 [INFO] [TRAIN] Epoch=37/40, Step=58/66, loss=0.007969, lr=0.001011, time_each_step=0.07s, eta=0:0:41 2020-05-09 09:54:50 [INFO] [TRAIN] Epoch=37/40, Step=60/66, loss=0.008208, lr=0.001003, time_each_step=0.07s, eta=0:0:41 2020-05-09 09:54:50 [INFO] [TRAIN] Epoch=37/40, Step=62/66, loss=0.005015, lr=0.000994, time_each_step=0.07s, eta=0:0:41 2020-05-09 09:54:50 [INFO] [TRAIN] Epoch=37/40, Step=64/66, loss=0.005178, lr=0.000985, time_each_step=0.07s, eta=0:0:41 2020-05-09 09:54:50 [INFO] [TRAIN] Epoch=37/40, Step=66/66, loss=0.006582, lr=0.000976, time_each_step=0.07s, eta=0:0:41 2020-05-09 09:54:50 [INFO] [TRAIN] Epoch 37 finished, loss=0.00766, lr=0.001118 . 2020-05-09 09:54:50 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:04<00:00, 4.39it/s]
2020-05-09 09:54:54 [INFO] [EVAL] Finished, Epoch=37, miou=0.885993, category_iou=[0.99509832 0.77688677], macc=0.995181, category_acc=[0.99845611 0.83539642], kappa=0.871984 . 2020-05-09 09:54:55 [INFO] Model saved in output/deeplab/epoch_37. 2020-05-09 09:54:55 [INFO] Current evaluated best model in eval_dataset is epoch_36, miou=0.8890458135436641 2020-05-09 09:54:57 [INFO] [TRAIN] Epoch=38/40, Step=2/66, loss=0.007792, lr=0.000967, time_each_step=0.18s, eta=0:0:39 2020-05-09 09:54:57 [INFO] [TRAIN] Epoch=38/40, Step=4/66, loss=0.006691, lr=0.000958, time_each_step=0.18s, eta=0:0:39 2020-05-09 09:54:58 [INFO] [TRAIN] Epoch=38/40, Step=6/66, loss=0.006714, lr=0.00095, time_each_step=0.18s, eta=0:0:39 2020-05-09 09:54:58 [INFO] [TRAIN] Epoch=38/40, Step=8/66, loss=0.012622, lr=0.000941, time_each_step=0.19s, eta=0:0:39 2020-05-09 09:54:58 [INFO] [TRAIN] Epoch=38/40, Step=10/66, loss=0.007521, lr=0.000932, time_each_step=0.19s, eta=0:0:38 2020-05-09 09:54:58 [INFO] [TRAIN] Epoch=38/40, Step=12/66, loss=0.005545, lr=0.000923, time_each_step=0.19s, eta=0:0:38 2020-05-09 09:54:58 [INFO] [TRAIN] Epoch=38/40, Step=14/66, loss=0.004409, lr=0.000914, time_each_step=0.2s, eta=0:0:38 2020-05-09 09:54:59 [INFO] [TRAIN] Epoch=38/40, Step=16/66, loss=0.007432, lr=0.000905, time_each_step=0.21s, eta=0:0:38 2020-05-09 09:54:59 [INFO] [TRAIN] Epoch=38/40, Step=18/66, loss=0.006007, lr=0.000896, time_each_step=0.22s, eta=0:0:38 2020-05-09 09:54:59 [INFO] [TRAIN] Epoch=38/40, Step=20/66, loss=0.006202, lr=0.000887, time_each_step=0.23s, eta=0:0:38 2020-05-09 09:55:00 [INFO] [TRAIN] Epoch=38/40, Step=22/66, loss=0.006321, lr=0.000878, time_each_step=0.12s, eta=0:0:33 2020-05-09 09:55:00 [INFO] [TRAIN] Epoch=38/40, Step=24/66, loss=0.009254, lr=0.00087, time_each_step=0.12s, eta=0:0:33 2020-05-09 09:55:00 [INFO] [TRAIN] Epoch=38/40, Step=26/66, loss=0.006952, lr=0.000861, time_each_step=0.13s, eta=0:0:33 2020-05-09 09:55:00 [INFO] [TRAIN] Epoch=38/40, Step=28/66, loss=0.008918, lr=0.000852, time_each_step=0.13s, eta=0:0:33 2020-05-09 09:55:01 [INFO] [TRAIN] Epoch=38/40, Step=30/66, loss=0.004322, lr=0.000843, time_each_step=0.14s, eta=0:0:33 2020-05-09 09:55:01 [INFO] [TRAIN] Epoch=38/40, Step=32/66, loss=0.015971, lr=0.000834, time_each_step=0.14s, eta=0:0:32 2020-05-09 09:55:01 [INFO] [TRAIN] Epoch=38/40, Step=34/66, loss=0.009883, lr=0.000825, time_each_step=0.14s, eta=0:0:32 2020-05-09 09:55:02 [INFO] [TRAIN] Epoch=38/40, Step=36/66, loss=0.004358, lr=0.000816, time_each_step=0.14s, eta=0:0:32 2020-05-09 09:55:02 [INFO] [TRAIN] Epoch=38/40, Step=38/66, loss=0.008843, lr=0.000807, time_each_step=0.14s, eta=0:0:32 2020-05-09 09:55:02 [INFO] [TRAIN] Epoch=38/40, Step=40/66, loss=0.006354, lr=0.000798, time_each_step=0.14s, eta=0:0:31 2020-05-09 09:55:02 [INFO] [TRAIN] Epoch=38/40, Step=42/66, loss=0.005783, lr=0.000789, time_each_step=0.14s, eta=0:0:31 2020-05-09 09:55:02 [INFO] [TRAIN] Epoch=38/40, Step=44/66, loss=0.006213, lr=0.00078, time_each_step=0.13s, eta=0:0:31 2020-05-09 09:55:03 [INFO] [TRAIN] Epoch=38/40, Step=46/66, loss=0.006021, lr=0.000771, time_each_step=0.12s, eta=0:0:30 2020-05-09 09:55:03 [INFO] [TRAIN] Epoch=38/40, Step=48/66, loss=0.007207, lr=0.000761, time_each_step=0.12s, eta=0:0:30 2020-05-09 09:55:03 [INFO] [TRAIN] Epoch=38/40, Step=50/66, loss=0.006584, lr=0.000752, time_each_step=0.11s, eta=0:0:30 2020-05-09 09:55:03 [INFO] [TRAIN] Epoch=38/40, Step=52/66, loss=0.004555, lr=0.000743, time_each_step=0.1s, eta=0:0:29 2020-05-09 09:55:03 [INFO] [TRAIN] Epoch=38/40, Step=54/66, loss=0.009331, lr=0.000734, time_each_step=0.09s, eta=0:0:29 2020-05-09 09:55:03 [INFO] [TRAIN] Epoch=38/40, Step=56/66, loss=0.005701, lr=0.000725, time_each_step=0.09s, eta=0:0:29 2020-05-09 09:55:03 [INFO] [TRAIN] Epoch=38/40, Step=58/66, loss=0.007235, lr=0.000716, time_each_step=0.08s, eta=0:0:28 2020-05-09 09:55:04 [INFO] [TRAIN] Epoch=38/40, Step=60/66, loss=0.006378, lr=0.000707, time_each_step=0.07s, eta=0:0:28 2020-05-09 09:55:04 [INFO] [TRAIN] Epoch=38/40, Step=62/66, loss=0.009071, lr=0.000698, time_each_step=0.07s, eta=0:0:28 2020-05-09 09:55:04 [INFO] [TRAIN] Epoch=38/40, Step=64/66, loss=0.009247, lr=0.000688, time_each_step=0.06s, eta=0:0:28 2020-05-09 09:55:04 [INFO] [TRAIN] Epoch=38/40, Step=66/66, loss=0.006577, lr=0.000679, time_each_step=0.06s, eta=0:0:28 2020-05-09 09:55:04 [INFO] [TRAIN] Epoch 38 finished, loss=0.007657, lr=0.000826 . 2020-05-09 09:55:04 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:05<00:00, 3.46it/s]
2020-05-09 09:55:09 [INFO] [EVAL] Finished, Epoch=38, miou=0.884936, category_iou=[0.9950983 0.77477304], macc=0.99518, category_acc=[0.99825346 0.84214229], kappa=0.870642 . 2020-05-09 09:55:10 [INFO] Model saved in output/deeplab/epoch_38. 2020-05-09 09:55:10 [INFO] Current evaluated best model in eval_dataset is epoch_36, miou=0.8890458135436641 2020-05-09 09:55:12 [INFO] [TRAIN] Epoch=39/40, Step=2/66, loss=0.00572, lr=0.00067, time_each_step=0.17s, eta=0:0:31 2020-05-09 09:55:12 [INFO] [TRAIN] Epoch=39/40, Step=4/66, loss=0.015622, lr=0.000661, time_each_step=0.18s, eta=0:0:31 2020-05-09 09:55:12 [INFO] [TRAIN] Epoch=39/40, Step=6/66, loss=0.006401, lr=0.000652, time_each_step=0.18s, eta=0:0:31 2020-05-09 09:55:13 [INFO] [TRAIN] Epoch=39/40, Step=8/66, loss=0.006549, lr=0.000642, time_each_step=0.18s, eta=0:0:31 2020-05-09 09:55:13 [INFO] [TRAIN] Epoch=39/40, Step=10/66, loss=0.008347, lr=0.000633, time_each_step=0.19s, eta=0:0:31 2020-05-09 09:55:13 [INFO] [TRAIN] Epoch=39/40, Step=12/66, loss=0.007932, lr=0.000624, time_each_step=0.19s, eta=0:0:30 2020-05-09 09:55:13 [INFO] [TRAIN] Epoch=39/40, Step=14/66, loss=0.014814, lr=0.000615, time_each_step=0.19s, eta=0:0:30 2020-05-09 09:55:13 [INFO] [TRAIN] Epoch=39/40, Step=16/66, loss=0.006238, lr=0.000605, time_each_step=0.19s, eta=0:0:30 2020-05-09 09:55:14 [INFO] [TRAIN] Epoch=39/40, Step=18/66, loss=0.005103, lr=0.000596, time_each_step=0.2s, eta=0:0:30 2020-05-09 09:55:14 [INFO] [TRAIN] Epoch=39/40, Step=20/66, loss=0.005605, lr=0.000587, time_each_step=0.2s, eta=0:0:30 2020-05-09 09:55:14 [INFO] [TRAIN] Epoch=39/40, Step=22/66, loss=0.006391, lr=0.000577, time_each_step=0.09s, eta=0:0:24 2020-05-09 09:55:14 [INFO] [TRAIN] Epoch=39/40, Step=24/66, loss=0.005036, lr=0.000568, time_each_step=0.09s, eta=0:0:24 2020-05-09 09:55:14 [INFO] [TRAIN] Epoch=39/40, Step=26/66, loss=0.010645, lr=0.000558, time_each_step=0.09s, eta=0:0:24 2020-05-09 09:55:14 [INFO] [TRAIN] Epoch=39/40, Step=28/66, loss=0.011293, lr=0.000549, time_each_step=0.09s, eta=0:0:24 2020-05-09 09:55:14 [INFO] [TRAIN] Epoch=39/40, Step=30/66, loss=0.006208, lr=0.00054, time_each_step=0.08s, eta=0:0:23 2020-05-09 09:55:15 [INFO] [TRAIN] Epoch=39/40, Step=32/66, loss=0.010214, lr=0.00053, time_each_step=0.08s, eta=0:0:23 2020-05-09 09:55:15 [INFO] [TRAIN] Epoch=39/40, Step=34/66, loss=0.006168, lr=0.000521, time_each_step=0.08s, eta=0:0:23 2020-05-09 09:55:15 [INFO] [TRAIN] Epoch=39/40, Step=36/66, loss=0.008173, lr=0.000511, time_each_step=0.07s, eta=0:0:23 2020-05-09 09:55:15 [INFO] [TRAIN] Epoch=39/40, Step=38/66, loss=0.007701, lr=0.000502, time_each_step=0.07s, eta=0:0:22 2020-05-09 09:55:15 [INFO] [TRAIN] Epoch=39/40, Step=40/66, loss=0.012319, lr=0.000492, time_each_step=0.07s, eta=0:0:22 2020-05-09 09:55:15 [INFO] [TRAIN] Epoch=39/40, Step=42/66, loss=0.005598, lr=0.000483, time_each_step=0.07s, eta=0:0:22 2020-05-09 09:55:15 [INFO] [TRAIN] Epoch=39/40, Step=44/66, loss=0.007309, lr=0.000473, time_each_step=0.07s, eta=0:0:22 2020-05-09 09:55:16 [INFO] [TRAIN] Epoch=39/40, Step=46/66, loss=0.005214, lr=0.000464, time_each_step=0.07s, eta=0:0:22 2020-05-09 09:55:16 [INFO] [TRAIN] Epoch=39/40, Step=48/66, loss=0.025032, lr=0.000454, time_each_step=0.07s, eta=0:0:22 2020-05-09 09:55:16 [INFO] [TRAIN] Epoch=39/40, Step=50/66, loss=0.005105, lr=0.000444, time_each_step=0.07s, eta=0:0:21 2020-05-09 09:55:16 [INFO] [TRAIN] Epoch=39/40, Step=52/66, loss=0.003989, lr=0.000435, time_each_step=0.07s, eta=0:0:21 2020-05-09 09:55:16 [INFO] [TRAIN] Epoch=39/40, Step=54/66, loss=0.008272, lr=0.000425, time_each_step=0.07s, eta=0:0:21 2020-05-09 09:55:16 [INFO] [TRAIN] Epoch=39/40, Step=56/66, loss=0.008703, lr=0.000415, time_each_step=0.07s, eta=0:0:21 2020-05-09 09:55:16 [INFO] [TRAIN] Epoch=39/40, Step=58/66, loss=0.004372, lr=0.000406, time_each_step=0.07s, eta=0:0:21 2020-05-09 09:55:16 [INFO] [TRAIN] Epoch=39/40, Step=60/66, loss=0.011462, lr=0.000396, time_each_step=0.07s, eta=0:0:21 2020-05-09 09:55:17 [INFO] [TRAIN] Epoch=39/40, Step=62/66, loss=0.006232, lr=0.000386, time_each_step=0.07s, eta=0:0:21 2020-05-09 09:55:17 [INFO] [TRAIN] Epoch=39/40, Step=64/66, loss=0.005689, lr=0.000376, time_each_step=0.07s, eta=0:0:20 2020-05-09 09:55:17 [INFO] [TRAIN] Epoch=39/40, Step=66/66, loss=0.012559, lr=0.000366, time_each_step=0.07s, eta=0:0:20 2020-05-09 09:55:17 [INFO] [TRAIN] Epoch 39 finished, loss=0.007796, lr=0.000522 . 2020-05-09 09:55:17 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:04<00:00, 4.26it/s]
2020-05-09 09:55:21 [INFO] [EVAL] Finished, Epoch=39, miou=0.883195, category_iou=[0.99497601 0.77141424], macc=0.99506, category_acc=[0.99834461 0.83401924], kappa=0.868445 . 2020-05-09 09:55:22 [INFO] Model saved in output/deeplab/epoch_39. 2020-05-09 09:55:22 [INFO] Current evaluated best model in eval_dataset is epoch_36, miou=0.8890458135436641 2020-05-09 09:55:24 [INFO] [TRAIN] Epoch=40/40, Step=2/66, loss=0.006621, lr=0.000357, time_each_step=0.18s, eta=0:0:16 2020-05-09 09:55:24 [INFO] [TRAIN] Epoch=40/40, Step=4/66, loss=0.007679, lr=0.000347, time_each_step=0.19s, eta=0:0:16 2020-05-09 09:55:25 [INFO] [TRAIN] Epoch=40/40, Step=6/66, loss=0.007146, lr=0.000337, time_each_step=0.19s, eta=0:0:16 2020-05-09 09:55:25 [INFO] [TRAIN] Epoch=40/40, Step=8/66, loss=0.008597, lr=0.000327, time_each_step=0.19s, eta=0:0:15 2020-05-09 09:55:25 [INFO] [TRAIN] Epoch=40/40, Step=10/66, loss=0.004891, lr=0.000317, time_each_step=0.19s, eta=0:0:15 2020-05-09 09:55:25 [INFO] [TRAIN] Epoch=40/40, Step=12/66, loss=0.007055, lr=0.000307, time_each_step=0.2s, eta=0:0:15 2020-05-09 09:55:25 [INFO] [TRAIN] Epoch=40/40, Step=14/66, loss=0.006331, lr=0.000297, time_each_step=0.2s, eta=0:0:15 2020-05-09 09:55:25 [INFO] [TRAIN] Epoch=40/40, Step=16/66, loss=0.011922, lr=0.000287, time_each_step=0.2s, eta=0:0:14 2020-05-09 09:55:26 [INFO] [TRAIN] Epoch=40/40, Step=18/66, loss=0.007392, lr=0.000277, time_each_step=0.2s, eta=0:0:14 2020-05-09 09:55:26 [INFO] [TRAIN] Epoch=40/40, Step=20/66, loss=0.007952, lr=0.000266, time_each_step=0.21s, eta=0:0:14 2020-05-09 09:55:26 [INFO] [TRAIN] Epoch=40/40, Step=22/66, loss=0.014121, lr=0.000256, time_each_step=0.09s, eta=0:0:8 2020-05-09 09:55:26 [INFO] [TRAIN] Epoch=40/40, Step=24/66, loss=0.005674, lr=0.000246, time_each_step=0.09s, eta=0:0:8 2020-05-09 09:55:26 [INFO] [TRAIN] Epoch=40/40, Step=26/66, loss=0.006298, lr=0.000236, time_each_step=0.09s, eta=0:0:8 2020-05-09 09:55:26 [INFO] [TRAIN] Epoch=40/40, Step=28/66, loss=0.004821, lr=0.000225, time_each_step=0.08s, eta=0:0:7 2020-05-09 09:55:27 [INFO] [TRAIN] Epoch=40/40, Step=30/66, loss=0.008302, lr=0.000215, time_each_step=0.08s, eta=0:0:7 2020-05-09 09:55:27 [INFO] [TRAIN] Epoch=40/40, Step=32/66, loss=0.010897, lr=0.000204, time_each_step=0.08s, eta=0:0:7 2020-05-09 09:55:27 [INFO] [TRAIN] Epoch=40/40, Step=34/66, loss=0.006845, lr=0.000194, time_each_step=0.07s, eta=0:0:7 2020-05-09 09:55:27 [INFO] [TRAIN] Epoch=40/40, Step=36/66, loss=0.011615, lr=0.000183, time_each_step=0.07s, eta=0:0:6 2020-05-09 09:55:27 [INFO] [TRAIN] Epoch=40/40, Step=38/66, loss=0.011419, lr=0.000172, time_each_step=0.07s, eta=0:0:6 2020-05-09 09:55:27 [INFO] [TRAIN] Epoch=40/40, Step=40/66, loss=0.006959, lr=0.000162, time_each_step=0.07s, eta=0:0:6 2020-05-09 09:55:27 [INFO] [TRAIN] Epoch=40/40, Step=42/66, loss=0.005323, lr=0.000151, time_each_step=0.07s, eta=0:0:6 2020-05-09 09:55:27 [INFO] [TRAIN] Epoch=40/40, Step=44/66, loss=0.004522, lr=0.00014, time_each_step=0.07s, eta=0:0:6 2020-05-09 09:55:28 [INFO] [TRAIN] Epoch=40/40, Step=46/66, loss=0.008302, lr=0.000129, time_each_step=0.07s, eta=0:0:6 2020-05-09 09:55:28 [INFO] [TRAIN] Epoch=40/40, Step=48/66, loss=0.008358, lr=0.000118, time_each_step=0.07s, eta=0:0:6 2020-05-09 09:55:28 [INFO] [TRAIN] Epoch=40/40, Step=50/66, loss=0.007327, lr=0.000107, time_each_step=0.07s, eta=0:0:5 2020-05-09 09:55:28 [INFO] [TRAIN] Epoch=40/40, Step=52/66, loss=0.010632, lr=9.5e-05, time_each_step=0.07s, eta=0:0:5 2020-05-09 09:55:28 [INFO] [TRAIN] Epoch=40/40, Step=54/66, loss=0.008985, lr=8.4e-05, time_each_step=0.07s, eta=0:0:5 2020-05-09 09:55:28 [INFO] [TRAIN] Epoch=40/40, Step=56/66, loss=0.007023, lr=7.2e-05, time_each_step=0.07s, eta=0:0:5 2020-05-09 09:55:28 [INFO] [TRAIN] Epoch=40/40, Step=58/66, loss=0.007071, lr=6e-05, time_each_step=0.07s, eta=0:0:5 2020-05-09 09:55:29 [INFO] [TRAIN] Epoch=40/40, Step=60/66, loss=0.004358, lr=4.8e-05, time_each_step=0.07s, eta=0:0:5 2020-05-09 09:55:29 [INFO] [TRAIN] Epoch=40/40, Step=62/66, loss=0.005111, lr=3.5e-05, time_each_step=0.07s, eta=0:0:5 2020-05-09 09:55:29 [INFO] [TRAIN] Epoch=40/40, Step=64/66, loss=0.004004, lr=2.2e-05, time_each_step=0.07s, eta=0:0:4 2020-05-09 09:55:29 [INFO] [TRAIN] Epoch=40/40, Step=66/66, loss=0.008556, lr=8e-06, time_each_step=0.07s, eta=0:0:4 2020-05-09 09:55:29 [INFO] [TRAIN] Epoch 40 finished, loss=0.007335, lr=0.000193 . 2020-05-09 09:55:29 [INFO] Start to evaluating(total_samples=76, total_steps=19)...
100%|██████████| 19/19 [00:04<00:00, 4.57it/s]
2020-05-09 09:55:33 [INFO] [EVAL] Finished, Epoch=40, miou=0.885201, category_iou=[0.9951117 0.77529023], macc=0.995193, category_acc=[0.99825726 0.84257793], kappa=0.870977 . 2020-05-09 09:55:34 [INFO] Model saved in output/deeplab/epoch_40. 2020-05-09 09:55:34 [INFO] Current evaluated best model in eval_dataset is epoch_36, miou=0.8890458135436641
4.模型预测
使用模型进行预测，同时使用pdx.seg.visualize将结果可视化，可视化结果将保存到./output/deeplab下，其中weight代表原图的权重，即mask可视化结果与原图权重因子。
image_name = 'optic_disc_seg/JPEGImages/H0005.jpg'
result = model.predict(image_name)
pdx.seg.visualize(image_name,rusult,weight=0.4, save_dir='./output/deeplab')可视化结果
PaddleX---dEEPlABv3+语义分割相关推荐
- DeepLabV3+语义分割实战
DeepLabV3+语义分割实战 语义分割是计算机视觉的一项重要任务,本文使用Jittor框架实现了DeepLabV3+语义分割模型. DeepLabV3+论文:https://arxiv.org/p ...
- 基于MindSpore复现Deeplabv3—语义分割
基于MindSpore复现Deeplabv3-语义分割 实验介绍 本实验主要介绍使用MindSpore深度学习框架在PASCAL VOC2012数据集上训练Deeplabv3网络模型.本实验使用了Mi ...
- 憨批的语义分割重制版9——Pytorch 搭建自己的DeeplabV3+语义分割平台
憨批的语义分割重制版9--Pytorch 搭建自己的DeeplabV3+语义分割平台 注意事项 学习前言 什么是DeeplabV3+模型 代码下载 DeeplabV3+实现思路 一.预测部分 1.主干 ...
- 类特征注意机制融合Deeplabv3+语义分割
针对经典语义分割网络在高分辨率遥感图像语义分割中存在的边缘目标分割不准确.不同类型目标分割不一致.预测效率慢等问题,本文提出了一种融合改进的Deeplabv3+网络CFAMNet的分类特征注意机制,用 ...
- win10+tensorflow+cuda+cudnn，实现deeplabv3语义分割自己的数据集
** 参考: ** 1.使用 deeplabv3+ 训练自己的数据集经验总结 2.window环境下进行deeplab_tensorflow实验之第一阶段 3.window环境下进行deeplab_t ...
- 工程（二）——DeeplabV3+语义分割训练自制数据集
目录 1.配置环境 1.1 利用conda新建一个环境并激活 1.2 安装pytorch 1.3 更新bashrc环境 2.安装DeeplabV3 2.1 克隆代码 2.2 下载数据集和预训练模型 3 ...
- 使用语义分割架构的文档扫描仪 DeepLabV3
0 介绍 地址:https://learnopencv.com/deep-learning-based-document-segmentation-using-semantic-segmentatio ...
- 视频教程-DeepLabv3+图像语义分割实战：训练自己的数据集-计算机视觉
DeepLabv3+图像语义分割实战:训练自己的数据集 大学教授,美国归国博士.博士生导师:人工智能公司专家顾问:长期从事人工智能.物联网.大数据研究:已发表学术论文100多篇,授权发明专利10多项 ...
- 【天池赛事】零基础入门语义分割-地表建筑物识别 Task3：网络模型结构发展
[天池赛事]零基础入门语义分割-地表建筑物识别 Task1:赛题理解与 baseline(3 天) – 学习主题:理解赛题内容解题流程 – 学习内容:赛题理解.数据读取.比赛 baseline 构建 ...
- 【深度学习】语义分割-综述(卷积)
这里写目录标题 0.笔记参考 1. 目的 2. 困难点 3. 数据集及评价指标 3.1数据集 3.2评价指标 4.实现架构 5. 模型发展 5.1基于全卷积的对称语义分割模型 5.1.1FCN(201 ...
最新文章
- 关于UIView的userInteractionEnabled属性
- php设置背景为透明,css如何设置背景颜色透明？css设置背景颜色透明度的两种方法介绍...
- 计算机外观类型,知道你的笔记本电脑是什么类型的吗？五大类型派对号入座
- 一千行MySQL学习笔记（十二）
- 洛谷——P2871 [USACO07DEC]手链Charm Bracelet
- 不要把异常当做业务逻辑，这性能可能你无法承受
- javascript 变量作用域
- 崩溃的数据库字符集问题
- oracle往游标中存数据,Oracle数据库：ORACLE11G在存储过程里面遍历游标
- 无聊的python课程_5 个无聊 Python 程序，用 Python 整蛊你的朋友们吧
- 【毕设笔记】轴承振动数据故障诊断软件系统
- opencv下载百度网盘链接及安装
- 如何检查python安装成功_如何检查python是否安装成功
- 喜马拉雅主播工具软件:图片OCR文字识别自动转成mp3高品质语音文件
- HTML常用的标签：
- outlook邮件路径更改_如何在Outlook中更改电子邮件的“答复”地址
- SQL的SUBSTR()函数
- MobaXterm安装
- hive行转列的高级用法later view explode
- 【日影】宇宙兄弟.Uchu.Kyodai.Chi——他们的宇宙梦，我的程序梦