paddleDetection global_config对象是在何时被赋值的呢？

项目

数据集

课程

比赛

模型库

活动

论坛

访问飞桨官网

项目

数据集

课程

比赛

模型库

活动

论坛

访问飞桨官网

l le19970417 发布于2021-07

想请问paddleDetection 源码中的global_config对象是在何时被赋予了键。

我在workspace.py文件最后一行打印时是空

回到train.py from worksace import global_config

下一句打印时值为：{'ImageFolder': {}, 'COCODataSet': {}, 'VOCDataSet': {}, 'WIDERFaceDataSet': {}, 'KeypointBottomUpCocoDataset': {}, 'KeypointBottomUpCrowdPoseDataset': {}, 'KeypointTopDownCocoDataset': {}, 'KeypointTopDownMPIIDataset': {}, 'MOTDataSet': {}, 'MOTImageFolder': {}, 'MOTVideoDataset': {}, 'VGG': {}, 'ResNet': {}, 'Res5Head': {}, 'DarkNet': {}, 'MobileNet': {}, 'MobileNetV3': {}, 'HRNet': {}, 'BlazeNet': {}, 'GhostNet': {}, 'SENet': {}, 'SERes5Head': {}, 'Res2Net': {}, 'Res2NetC5': {}, 'AnchorGeneratorSSD': {}, 'RCNNBox': {}, 'MultiClassNMS': {}, 'MatrixNMS': {}, 'YOLOBox': {}, 'SSDBox': {}, 'AnchorGrid': {}, 'FCOSBox': {}, 'TTFBox': {}, 'JDEBox': {}, 'MaskMatrixNMS': {}, 'DLA': {}, 'FPN': {}, 'YOLOv3FPN': {}, 'PPYOLOFPN': {}, 'PPYOLOTinyFPN': {}, 'PPYOLOPAN': {}, 'HRFPN': {}, 'TTFFPN': {}, 'CenterNetDLAFPN': {}, 'BlazeNeck': {}, 'AnchorGenerator': {}, 'RPNTargetAssign': {}, 'BBoxAssigner': {}, 'BBoxLibraAssigner': {}, 'MaskAssigner': {}, 'RBoxAssigner': {}, 'ProposalGenerator': {}, 'RPNHead': {}, 'RoIAlign': {}, 'TwoFCHead': {}, 'XConvNormHead': {}, 'BBoxHead': {}, 'MaskFeat': {}, 'MaskHead': {}, 'YOLOv3Head': {}, 'SSDHead': {}, 'FCOSFeat': {}, 'FCOSHead': {}, 'SOLOv2MaskHead': {}, 'SOLOv2Head': {}, 'HMHead': {}, 'WHHead': {}, 'TTFHead': {}, 'CascadeTwoFCHead': {}, 'CascadeXConvNormHead': {}, 'CascadeHead': {}, 'FaceHead': {}, 'S2ANetHead': {}, 'HrHRNetHead': {}, 'YOLOv3Loss': {}, 'IouLoss': {}, 'GIoULoss': {}, 'DIouLoss': {}, 'IouAwareLoss': {}, 'SSDLoss': {}, 'FCOSLoss': {}, 'SOLOv2Loss': {}, 'CTFocalLoss': {}, 'KeyPointMSELoss': {}, 'HrHRNetLoss': {}, 'JDEDetectionLoss': {}, 'JDEEmbeddingLoss': {}, 'JDELoss': {}, 'FairMOTLoss': {}, 'CenterNetHead': {}, 'BaseArch': {}, 'FasterRCNN': {}, 'MaskRCNN': {}, 'YOLOv3': {}, 'CascadeRCNN': {}, 'SSD': {}, 'FCOS': {}, 'SOLOv2': {}, 'TTFNet': {}, 'S2ANet': {}, 'HigherHRNet': {}, 'HrHRNetPostProcess': {}, 'TopDownHRNet': {}, 'KalmanFilter': {}, 'BaseTrack': {}, 'STrack': {}, 'Track': {}, 'JDETracker': {}, 'DeepSORTTracker': {}, 'JDE': {}, 'DeepSORT': {}, 'FairMOT': {}, 'CenterNet': {}, 'BlazeFace': {}, 'BBoxPostProcess': {}, 'MaskPostProcess': {}, 'FCOSPostProcess': {}, 'S2ANetBBoxPostProcess': {}, 'JDEBBoxPostProcess': {}, 'CenterNetPostProcess': {}, 'JDEEmbeddingHead': {}, 'PCBPyramid': {}, 'FairMOTEmbeddingHead': {}, 'TrainReader': {}, 'EvalReader': {}, 'TestReader': {}, 'EvalMOTReader': {}, 'TestMOTReader': {}, 'LearningRate': {}, 'OptimizerBuilder': {}, 'Pruner': {}, 'QAT': {}, 'DistillYOLOv3Loss': {}}

实在没想通望大神指点

全部评论(2)

JavaRoom

#2 回复于2021-07

打印的位置不对。

workspace.py里，应该再load_config方法里打印

def load_config(file_path):
    """
    Load config from file.

    Args:
        file_path (str): Path of the config file to be loaded.

    Returns: global config
    """
    _, ext = os.path.splitext(file_path)
    assert ext in ['.yml', '.yaml'], "only support yaml files for now"

    # load config from file and merge into global config
    cfg = _load_config_with_base(file_path)
    cfg['filename'] = os.path.splitext(os.path.split(file_path)[-1])[0]
    merge_config(cfg)

    return global_config

train.py

# ignore warning log
import warnings
warnings.filterwarnings('ignore')
import random
import numpy as np

import paddle
from paddle.distributed import ParallelEnv

from ppdet.core.workspace import load_config, merge_config, create

JavaRoom

#3 回复于2021-07

你再试试，有情况再讨论

提issue

需求/bug反馈？一键提issue告诉我们

提pr

发现bug？如果您知道修复办法，欢迎提pr直接参与建设飞桨~