X光安检图像检测挑战赛3.0

Livingbody
1,128 阅读9分钟

携手创作，共同成长！这是我参与「掘金日新计划 · 8 月更文挑战」的第24天，点击查看活动详情

一、X光安检图像检测挑战赛3.0

地址：challenge.xfyun.cn/topic/info?…

1.赛事背景

X光安检是目前在城市轨交、铁路、机场、物流业广泛使用的物检手段。使用人工智能技术，辅助一线安检员进行X光安检判图，可以有效降低因为安检员经验、能力或工作状态造成的错漏检问题。在实际场景中，因待检测物品的多样性、成像角度、重叠遮挡等问题，X光安检图像检测算法研究存在一定挑战。

2.赛事任务

本赛事的任务是：基于科大讯飞提供的真实X光安检图像集构建检测模型，对X光安检图像中的指定类别的物品进行检测。

3.评审规则

3.1 数据说明

此次比赛提供带标注的训练数据，即待检测物品在包裹中的X光图像及其标注文件。

本次比赛标注文件中的类别为8类，包括：

刀（knife）、剪刀（scissors）、打火机（lighter）、优盘（USBFlashDisk）、压力容器（pressure）、带喷嘴塑料瓶（plasticBottleWithaNozzle）、公章（seal）、电池（battery）。

待识别物品的X光成像示意图如图所示。

比赛提供的X光图像及其矩形框标注的文件按照数据来源存放在不同的文件夹中，图像文件采用jpg格式，标注文件采用xml格式，各字段含义参照voc数据集。voc各字段含义对应表为：

  • filename 文件名

  • size 图像尺寸

    • width 图像宽度

    • height 图像高度

    • depth 图像深度，一般为3表示是彩色图像

  • object 图像中的目标，可能有多个

    • name 该目标的标签名称

    • bndbox 该目标的标注框

      • xmin 该目标的左上角宽度方向坐标

      • ymin 该目标的左上角高度方向坐标

      • xmax 该目标的右下角宽度方向坐标

      • ymax 该目标的右下角高度方向坐标

3.2 评估指标

评测方式采用计算mAP（IoU = 0.5）的方式。

首先计算每个类的AP：

（1）根据预测框和标注框的IoU是否达到阈值0.5判断该预测框是真阳性还是假阳性；

（2）根据每个预测框的置信度进行从高到低排序；

（3）在不同置信度阈值下计算精确率和召回率，得到若干组PR值；

（4）绘制PR曲线并计算AP值。

然后计算mAP：把所有类的AP值求平均得到mAP。

3.3 作品提交要求

作品必须健康、合法、无任何不良信息及商业宣传行为，不违反任何中华人民共和国有关法律。须保证原创性，不侵犯任何第三方知识产权或其他权利；一经发现或经权利人指出，主办方将直接取消其参赛资格，科大讯飞保留赛事解释权。

选手需要提交json格式文件，详情见示例。其中，坐标值必须为大于0的正数且不能超过图像的宽高。

按照赛题数据页面2022gamedatasettest1.txt里面的顺序组织json

提交文件需按序排列，首先按图片顺序排列，然后按类别顺序排列，置信度顺序随意。

  • (a) 图片顺序，请按照图片编号顺序。

  • (b) 类别顺序，请参照下列顺序： {'knife': 1, 'scissors': 2, 'lighter': 3, 'USBFlashDisk': 4, 'pressure': 5, 'plasticBottleWithaNozzle': 6, 'seal': 7, 'battery': 8}

二、数据集处理

https://p3-juejin.byteimg.com/tos-cn-i-k3u1fbpfcp/24d34b0345a44534a599bbd0e74ede90~tplv-k3u1fbpfcp-zoom-1.image

1.标注文件整理

# 解压缩数据集
!unzip -qoa data/data165820/round1.zip -d data/
!unzip -qoa data/data165820/round2_test.zip -d data/

!mv data/讯飞研究院-X光安检图像检测挑战赛2022公开数据 data/round1

!mkdir data/round1/train/XML

import glob
import shutil
import os

def move_xml(domain_name):
    xml_dir=f"data/round1/train/{domain_name}/XML"
    filenames=glob.glob(f'{xml_dir}/*.xml')
    print(f"{xml_dir}移动的XML共有{len(filenames)}个。")
    if len(filenames)==0:
        return 
    for filename  in filenames:
        file_path, shortname=os.path.split(filename)
        new_name=domain_name+'_'+shortname
        new_name=os.path.join("data/round1/train/XML", new_name)
        shutil.move(filename, new_name)
    shutil.rmtree(xml_dir)
    print("重命名并移动文件完成。")

move_xml('domain1') 
move_xml('domain2')    
move_xml('domain3')

data/round1/train/domain1/XML移动的XML共有1323个。
重命名并移动文件完成。
data/round1/train/domain2/XML移动的XML共有1383个。
重命名并移动文件完成。
data/round1/train/domain3/XML移动的XML共有1308个。
重命名并移动文件完成。

print(len(glob.glob('data/round1/train/XML/*.xml')))

4014

!mv data/round1/train/XML data/round1/train/Annotations

2.图片文件整理

import glob
import shutil
import os

def move_xml(domain_name):
    jpg_dir=f"data/round1/train/{domain_name}/"
    filenames=glob.glob(f'{jpg_dir}/*.jpg')
    print(f"{jpg_dir}移动的XML共有{len(filenames)}个。")
    if len(filenames)==0:
        return 
    for filename  in filenames:
        file_path, shortname=os.path.split(filename)
        new_name=domain_name+'_'+shortname
        new_name=os.path.join("data/round1/train/JPEGImages", new_name)
        shutil.move(filename, new_name)
    shutil.rmtree(jpg_dir)
    print("重命名并移动文件完成。")

!mkdir data/round1/train/JPEGImages

move_xml('domain1')
move_xml('domain2')
move_xml('domain3')

data/round1/train/domain1/移动的XML共有1323个。
重命名并移动文件完成。
data/round1/train/domain2/移动的XML共有1383个。
重命名并移动文件完成。
data/round1/train/domain3/移动的XML共有1308个。
重命名并移动文件完成。

print(len(glob.glob('data/round1/train/JPEGImages/*.jpg')))

4014

3.PaddleX 安装

!python -m pip install --upgrade -q pip --user
!pip install -q -U paddlex

4.数据集划分

!paddlex --split_dataset --format VOC --dataset_dir data/round1/train/ --val_value 0.2

[08-21 13:52:20 MainThread @logger.py:242] Argv: /opt/conda/envs/python35-paddle120-env/bin/paddlex --split_dataset --format VOC --dataset_dir data/round1/train/ --val_value 0.2
[08-21 13:52:20 MainThread @utils.py:79] WRN paddlepaddle version: 2.3.1. The dynamic graph version of PARL is under development, not fully tested and supported
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/parl/remote/communication.py:38: DeprecationWarning: 'pyarrow.default_serialization_context' is deprecated as of 2.0.0 and will be removed in a future version. Use pickle or the pyarrow IPC functionality instead.
  context = pyarrow.default_serialization_context()
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/__init__.py:107: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
  from collections import MutableMapping
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/rcsetup.py:20: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
  from collections import Iterable, Mapping
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/colors.py:53: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
  from collections import Sized
2022-08-21 13:52:21,696-WARNING: type object 'QuantizationTransformPass' has no attribute '_supported_quantizable_op_type'
2022-08-21 13:52:21,696-WARNING: If you want to use training-aware and post-training quantization, please use Paddle >= 1.8.4 or develop version
2022-08-21 13:52:23 [INFO]	Dataset split starts...
2022-08-21 13:52:23 [INFO]	Dataset split done.
2022-08-21 13:52:23 [INFO]	Train samples: 3212
2022-08-21 13:52:23 [INFO]	Eval samples: 802
2022-08-21 13:52:23 [INFO]	Test samples: 0
2022-08-21 13:52:23 [INFO]	Split files saved in data/round1/train/


!ls data/round1/train/

Annotations  JPEGImages  labels.txt  train_list.txt  val_list.txt

5.图片查看

三、模型训练

1.transforms定义

# 定义训练和验证时的transforms
# API说明：https://github.com/PaddlePaddle/PaddleX/blob/develop/docs/apis/transforms/transforms.md
import paddlex as pdx
from paddlex import transforms as T

train_transforms = T.Compose([
    T.MixupImage(mixup_epoch=250), T.RandomDistort(),
    T.RandomExpand(im_padding_value=[123.675, 116.28, 103.53]), T.RandomCrop(),
    T.RandomDistort(brightness_range=0.5, brightness_prob=0.5, contrast_range=0.5, contrast_prob=0.5, saturation_range=0.5, saturation_prob=0.5, hue_range=18, hue_prob=0.5, random_apply=True, count=4, shuffle_channel=False),
    T.RandomHorizontalFlip(), T.RandomVerticalFlip(), T.BatchRandomResize(
        target_sizes=[320, 352, 384, 416, 448, 480, 512, 544, 576, 608],
        interp='RANDOM'), T.Normalize(
            mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])

eval_transforms = T.Compose([
    T.Resize(
        608, interp='CUBIC'), T.Normalize(
            mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])

[08-21 14:22:29 MainThread @utils.py:79] WRN paddlepaddle version: 2.3.1. The dynamic graph version of PARL is under development, not fully tested and supported


/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/parl/remote/communication.py:38: DeprecationWarning: 'pyarrow.default_serialization_context' is deprecated as of 2.0.0 and will be removed in a future version. Use pickle or the pyarrow IPC functionality instead.
  context = pyarrow.default_serialization_context()
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/__init__.py:107: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
  from collections import MutableMapping
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/rcsetup.py:20: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
  from collections import Iterable, Mapping
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/matplotlib/colors.py:53: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
  from collections import Sized
2022-08-21 14:22:29,871-WARNING: type object 'QuantizationTransformPass' has no attribute '_supported_quantizable_op_type'
2022-08-21 14:22:29,873-WARNING: If you want to use training-aware and post-training quantization, please use Paddle >= 1.8.4 or develop version

2.数据集定义

# 定义训练和验证所用的数据集
# API说明：https://github.com/PaddlePaddle/PaddleX/blob/develop/docs/apis/datasets.md
train_dataset = pdx.datasets.VOCDetection(
    data_dir='data/round1/train/',
    file_list='data/round1/train/train_list.txt',
    label_list='data/round1/train/labels.txt',
    transforms=train_transforms,
    shuffle=True)

eval_dataset = pdx.datasets.VOCDetection(
    data_dir='data/round1/train',
    file_list='data/round1/train/val_list.txt',
    label_list='data/round1/train/labels.txt',
    transforms=eval_transforms,
    shuffle=False)

2022-08-21 14:22:31 [INFO]	Starting to read file list from dataset...
2022-08-21 14:22:35 [INFO]	3212 samples in file data/round1/train/train_list.txt, including 3212 positive samples and 0 negative samples.
creating index...
index created!
2022-08-21 14:22:35 [INFO]	Starting to read file list from dataset...
2022-08-21 14:22:36 [INFO]	802 samples in file data/round1/train/val_list.txt, including 802 positive samples and 0 negative samples.
creating index...
index created!

3.模型定义

# 初始化模型，并进行训练
# 可使用VisualDL查看训练指标，参考https://github.com/PaddlePaddle/PaddleX/blob/develop/docs/visualdl.md
num_classes = len(train_dataset.labels)
model = pdx.det.YOLOv3(num_classes=num_classes, backbone='DarkNet53')

W0821 14:22:36.573086 10280 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 10.1
W0821 14:22:36.577816 10280 gpu_resources.cc:91] device: 0, cuDNN Version: 7.6.

4.模型定义

主要是batch size 选择：

  • batch size: 8， 对应显存 13421
  • batch size:19.37， 对应显存 32000 选择16即可。

# API说明：https://github.com/PaddlePaddle/PaddleX/blob/develop//docs/apis/models/detection.md
# 各参数介绍与调整说明：https://github.com/PaddlePaddle/PaddleX/blob/develop//docs/parameters.md
model.train(
    num_epochs=270,
    train_dataset=train_dataset,
    train_batch_size=16,
    eval_dataset=eval_dataset,
    learning_rate=0.001 / 8,
    warmup_steps=1000,
    warmup_start_lr=0.0,
    save_interval_epochs=10,
    lr_decay_epochs=[216, 243],
    save_dir='output/yolov3_darknet53',
    use_vdl=True)

2022-08-21 14:22:36 [INFO]	Downloading DarkNet53_pretrained.pdparams from https://paddledet.bj.bcebos.com/models/pretrained/DarkNet53_pretrained.pdparams


  0%|          | 291/158704 [00:00<00:56, 2800.01KB/s]100%|██████████| 158704/158704 [00:03<00:00, 50714.52KB/s]


2022-08-21 14:22:39 [INFO]	Loading pretrained model from output/yolov3_darknet53/pretrain/DarkNet53_pretrained.pdparams
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv0.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv0.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv0.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv0.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv0.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv1.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv1.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv1.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv1.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv1.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv2.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv2.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv2.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv2.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv2.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv3.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv3.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv3.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv3.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.conv3.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.route.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.route.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.route.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.route.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.conv_module.route.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.tip.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.tip.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.tip.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.tip.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.0.tip.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_transition.0.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_transition.0.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_transition.0.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_transition.0.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_transition.0.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv0.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv0.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv0.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv0.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv0.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv1.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv1.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv1.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv1.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv1.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv2.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv2.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv2.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv2.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv2.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv3.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv3.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv3.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv3.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.conv3.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.route.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.route.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.route.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.route.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.conv_module.route.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.tip.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.tip.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.tip.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.tip.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.1.tip.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_transition.1.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_transition.1.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_transition.1.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_transition.1.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_transition.1.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv0.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv0.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv0.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv0.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv0.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv1.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv1.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv1.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv1.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv1.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv2.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv2.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv2.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv2.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv2.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv3.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv3.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv3.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv3.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.conv3.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.route.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.route.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.route.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.route.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.conv_module.route.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.tip.conv.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.tip.batch_norm.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.tip.batch_norm.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.tip.batch_norm._mean is not in pretrained model
2022-08-21 14:22:40 [WARNING]	neck.yolo_block.2.tip.batch_norm._variance is not in pretrained model
2022-08-21 14:22:40 [WARNING]	yolo_head.yolo_output.0.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	yolo_head.yolo_output.0.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	yolo_head.yolo_output.1.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	yolo_head.yolo_output.1.bias is not in pretrained model
2022-08-21 14:22:40 [WARNING]	yolo_head.yolo_output.2.weight is not in pretrained model
2022-08-21 14:22:40 [WARNING]	yolo_head.yolo_output.2.bias is not in pretrained model
2022-08-21 14:22:40 [INFO]	There are 260/366 variables loaded into YOLOv3.
2022-08-21 14:23:05 [INFO]	[TRAIN] Epoch=1/270, Step=10/200, loss_xy=7.773049, loss_wh=8.631678, loss_obj=8424.009766, loss_cls=14.103580, loss=8454.517578, lr=0.000001, time_each_step=2.48s, eta=37:16:9
2022-08-21 14:23:23 [INFO]	[TRAIN] Epoch=1/270, Step=20/200, loss_xy=6.570035, loss_wh=6.603613, loss_obj=578.750854, loss_cls=12.434158, loss=604.358643, lr=0.000002, time_each_step=1.81s, eta=27:12:17
2022-08-21 14:23:44 [INFO]	[TRAIN] Epoch=1/270, Step=30/200, loss_xy=6.975683, loss_wh=7.730688, loss_obj=47.920227, loss_cls=14.573313, loss=77.199913, lr=0.000004, time_each_step=2.02s, eta=30:17:17
2022-08-21 14:24:06 [INFO]	[TRAIN] Epoch=1/270, Step=40/200, loss_xy=6.988766, loss_wh=6.911714, loss_obj=39.952698, loss_cls=13.511222, loss=67.364403, lr=0.000005, time_each_step=2.24s, eta=33:32:9
2022-08-21 14:24:31 [INFO]	[TRAIN] Epoch=1/270, Step=50/200, loss_xy=6.662724, loss_wh=7.118166, loss_obj=31.094318, loss_cls=13.199011, loss=58.074219, lr=0.000006, time_each_step=2.55s, eta=38:15:22
2022-08-21 14:24:49 [INFO]	[TRAIN] Epoch=1/270, Step=60/200, loss_xy=5.612492, loss_wh=6.151673, loss_obj=23.123552, loss_cls=12.082365, loss=46.970085, lr=0.000007, time_each_step=1.76s, eta=26:26:55
2022-08-21 14:25:15 [INFO]	[TRAIN] Epoch=1/270, Step=70/200, loss_xy=6.188484, loss_wh=7.226328, loss_obj=22.821095, loss_cls=11.426645, loss=47.662552, lr=0.000009, time_each_step=2.6s, eta=38:56:18
2022-08-21 14:25:39 [INFO]	[TRAIN] Epoch=1/270, Step=80/200, loss_xy=7.388262, loss_wh=8.146477, loss_obj=25.573256, loss_cls=14.415861, loss=55.523857, lr=0.000010, time_each_step=2.42s, eta=36:12:28
2022-08-21 14:26:02 [INFO]	[TRAIN] Epoch=1/270, Step=90/200, loss_xy=6.698905, loss_wh=6.328094, loss_obj=25.467766, loss_cls=13.299914, loss=51.794682, lr=0.000011, time_each_step=2.24s, eta=33:33:32


Exception in thread Thread-7:
Traceback (most recent call last):
  File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/dataloader/dataloader_iter.py", line 620, in _get_data
    data = self._data_queue.get(timeout=self._timeout)
  File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/multiprocessing/queues.py", line 105, in get
    raise Empty
_queue.Empty

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/threading.py", line 926, in _bootstrap_inner
    self.run()
  File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/threading.py", line 870, in run
    self._target(*self._args, **self._kwargs)
  File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/dataloader/dataloader_iter.py", line 534, in _thread_loop
    batch = self._get_data()
  File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/dataloader/dataloader_iter.py", line 636, in _get_data
    "pids: {}".format(len(failed_workers), pids))
RuntimeError: DataLoader 2 workers exit unexpectedly, pids: 10395, 10396




---------------------------------------------------------------------------

SystemError                               Traceback (most recent call last)

/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/dataloader/dataloader_iter.py in __next__(self)
    745                 if _in_legacy_dygraph():
--> 746                     data = self._reader.read_next_var_list()
    747                     data = _restore_batch(data, self._structure_infos.pop(0))


SystemError: (Fatal) Blocking queue is killed because the data reader raises an exception.
  [Hint: Expected killed_ != true, but received killed_:1 == true:1.] (at /paddle/paddle/fluid/operators/reader/blocking_queue.h:166)

​
During handling of the above exception, another exception occurred:

KeyboardInterrupt                         Traceback (most recent call last)

/tmp/ipykernel_10280/328266389.py in <module>
     12     lr_decay_epochs=[216, 243],
     13     save_dir='output/yolov3_darknet53',
---> 14     use_vdl=True)


/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddlex/cv/models/detector.py in train(self, num_epochs, train_dataset, train_batch_size, eval_dataset, optimizer, save_interval_epochs, log_interval_steps, save_dir, pretrain_weights, learning_rate, warmup_steps, warmup_start_lr, lr_decay_epochs, lr_decay_gamma, metric, use_ema, early_stop, early_stop_patience, use_vdl, resume_checkpoint)
    332             early_stop=early_stop,
    333             early_stop_patience=early_stop_patience,
--> 334             use_vdl=use_vdl)
    335 
    336     def quant_aware_train(self,


/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddlex/cv/models/base.py in train_loop(self, num_epochs, train_dataset, train_batch_size, eval_dataset, save_interval_epochs, log_interval_steps, save_dir, ema, early_stop, early_stop_patience, use_vdl)
    331             step_time_tic = time.time()
    332 
--> 333             for step, data in enumerate(self.train_data_loader()):
    334                 if nranks > 1:
    335                     outputs = self.run(ddp_net, data, mode='train')


/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/dataloader/dataloader_iter.py in __next__(self)
    744             else:
    745                 if _in_legacy_dygraph():
--> 746                     data = self._reader.read_next_var_list()
    747                     data = _restore_batch(data, self._structure_infos.pop(0))
    748                 else:


KeyboardInterrupt:
avatar
文章
阅读
粉丝