ValueError: Annotation must have data_list and metainfo keys

[690696306]

Prerequisite

• I have searched Issues and Discussions but cannot get the expected help.
• The bug has not been fixed in the latest version(https://github.com/open-mmlab/mmpose).

Environment

py10_cuda118_torch200

Reproduces the problem - code sample

Train config file

base = ['../../../base/default_runtime.py']

runtime

max_epochs = 270
stage2_num_epochs = 30
base_lr = 4e-3

train_cfg = dict(max_epochs=max_epochs, val_interval=10)
randomness = dict(seed=21)

optimizer

optim_wrapper = dict(
type='OptimWrapper',
optimizer=dict(type='AdamW', lr=base_lr, weight_decay=0.05),
paramwise_cfg=dict(
norm_decay_mult=0, bias_decay_mult=0, bypass_duplicate=True))

learning rate

param_scheduler = [
dict(
type='LinearLR',
start_factor=1.0e-5,
by_epoch=False,
begin=0,
end=1000),
dict(
type='CosineAnnealingLR',
eta_min=base_lr * 0.05,
begin=max_epochs // 2,
end=max_epochs,
T_max=max_epochs // 2,
by_epoch=True,
convert_to_iter_based=True),
]

automatically scaling LR based on the actual training batch size

auto_scale_lr = dict(base_batch_size=512)

codec settings

codec = dict(
type='SimCCLabel',
input_size=(288, 384),
sigma=(6., 6.93),
simcc_split_ratio=2.0,
normalize=False,
use_dark=False)

model settings

model = dict(
type='TopdownPoseEstimator',
data_preprocessor=dict(
type='PoseDataPreprocessor',
mean=[123.675, 116.28, 103.53],
std=[58.395, 57.12, 57.375],
bgr_to_rgb=True),
backbone=dict(
scope='mmdet',
type='CSPNeXt',
arch='P5',
expand_ratio=0.5,
deepen_factor=1.,
widen_factor=1.,
out_indices=(4, ),
channel_attention=True,
norm_cfg=dict(type='SyncBN'),
act_cfg=dict(type='SiLU'),
init_cfg=dict(
type='Pretrained',
prefix='backbone.',
checkpoint='https://download.openmmlab.com/mmpose/v1/projects/'
'rtmposev1/cspnext-m_udp-aic-coco_210e-256x192-f2f7d6f6_20230130.pth' # noqa: E501
)),
head=dict(
type='RTMCCHead',
in_channels=1024,
out_channels=133,
input_size=codec['input_size'],
in_featuremap_size=tuple([s // 32 for s in codec['input_size']]),
simcc_split_ratio=codec['simcc_split_ratio'],
final_layer_kernel_size=7,
gau_cfg=dict(
hidden_dims=256,
s=128,
expansion_factor=2,
dropout_rate=0.,
drop_path=0.,
act_fn='SiLU',
use_rel_bias=False,
pos_enc=False),
loss=dict(
type='KLDiscretLoss',
use_target_weight=True,
beta=10.,
label_softmax=True),
decoder=codec),
test_cfg=dict(flip_test=True, ))

base dataset settings

dataset_type = 'MyCustomDataset'
#data_mode = 'topdown'
data_root = 'data/coco/'

backend_args = dict(backend='local')

backend_args = dict(

backend='petrel',

path_mapping=dict({

f'{data_root}': 's3://openmmlab/datasets/detection/coco/',

f'{data_root}': 's3://openmmlab/datasets/detection/coco/'

}))

pipelines

train_pipeline = [
dict(type='LoadImage', backend_args=backend_args),
dict(type='GetBBoxCenterScale'),
dict(type='RandomFlip', direction='horizontal'),
dict(type='RandomHalfBody'),
dict(
type='RandomBBoxTransform', scale_factor=[0.6, 1.4], rotate_factor=80),
dict(type='TopdownAffine', input_size=codec['input_size']),
dict(type='mmdet.YOLOXHSVRandomAug'),
dict(
type='Albumentation',
transforms=[
dict(type='Blur', p=0.1),
dict(type='MedianBlur', p=0.1),
dict(
type='CoarseDropout',
max_holes=1,
max_height=0.4,
max_width=0.4,
min_holes=1,
min_height=0.2,
min_width=0.2,
p=1.0),
]),
dict(type='GenerateTarget', encoder=codec),
dict(type='PackPoseInputs')
]
val_pipeline = [
dict(type='LoadImage', backend_args=backend_args),
dict(type='GetBBoxCenterScale'),
dict(type='TopdownAffine', input_size=codec['input_size']),
dict(type='PackPoseInputs')
]

train_pipeline_stage2 = [
dict(type='LoadImage', backend_args=backend_args),
dict(type='GetBBoxCenterScale'),
dict(type='RandomFlip', direction='horizontal'),
dict(type='RandomHalfBody'),
dict(
type='RandomBBoxTransform',
shift_factor=0.,
scale_factor=[0.75, 1.25],
rotate_factor=60),
dict(type='TopdownAffine', input_size=codec['input_size']),
dict(type='mmdet.YOLOXHSVRandomAug'),
dict(
type='Albumentation',
transforms=[
dict(type='Blur', p=0.1),
dict(type='MedianBlur', p=0.1),
dict(
type='CoarseDropout',
max_holes=1,
max_height=0.4,
max_width=0.4,
min_holes=1,
min_height=0.2,
min_width=0.2,
p=0.5),
]),
dict(type='GenerateTarget', encoder=codec),
dict(type='PackPoseInputs')
]

data loaders

train_dataloader = dict(
batch_size=32,
num_workers=10,
persistent_workers=True,
sampler=dict(type='DefaultSampler', shuffle=True),
dataset=dict(
type=dataset_type,
data_root=data_root,
#data_mode=data_mode,
ann_file='annotations/coco_wholebody_train_v1.0.json',
data_prefix=dict(img='train2017/'),
pipeline=train_pipeline,
metainfo=dict(from_file='configs/base/datasets/coco_wholebody.py'),
))
val_dataloader = dict(
batch_size=32,
num_workers=10,
persistent_workers=True,
drop_last=False,
sampler=dict(type='DefaultSampler', shuffle=False, round_up=False),
dataset=dict(
type=dataset_type,
data_root=data_root,
#data_mode=data_mode,
ann_file='annotations/coco_wholebody_val_v1.0.json',
data_prefix=dict(img='val2017/'),
test_mode=True,
#bbox_file='data/coco/person_detection_results/'
#'COCO_val2017_detections_AP_H_56_person.json',
pipeline=val_pipeline,
metainfo=dict(from_file='configs/base/datasets/coco_wholebody.py'),
))
test_dataloader = val_dataloader

hooks

#default_hooks = dict(
#checkpoint=dict(
#save_best='coco-wholebody/AP', rule='greater', max_keep_ckpts=1))

custom_hooks = [
dict(
type='EMAHook',
ema_type='ExpMomentumEMA',
momentum=0.0002,
update_buffers=True,
priority=49),
dict(
type='mmdet.PipelineSwitchHook',
switch_epoch=max_epochs - stage2_num_epochs,
switch_pipeline=train_pipeline_stage2)
]

evaluators

#val_evaluator = dict(
#type='CocoWholeBodyMetric',
#ann_file=data_root + 'annotations/coco_wholebody_val_v1.0.json')

val_evaluator = [
dict(type='PCKAccuracy', thr=0.2),
dict(type='AUC'),
dict(type='EPE'),
]

test_evaluator = val_evaluator

Custom datasets

from mmengine.dataset import BaseDataset
from mmpose.registry import DATASETS

import copy
import os.path as osp
from typing import Optional

import numpy as np

@DATASETS.register_module(name='MyCustomDataset')
class MyCustomDataset(BaseDataset):

METAINFO: dict = dict(
    from_file='configs/_base_/datasets/coco_wholebody.py')

Reproduces the problem - command or script

python tools/train.py /home/lab/Anaconda_project/mmpose/configs/wholebody_2d_keypoint/rtmpose/coco-wholebody/rtmpose-l_8xb32-270e_coco-wholebody-384x288.py --work-dir /home/lab/Anaconda_project/mmpose/weights

Reproduces the problem - error message

/home/lab/Anaconda_project/mmpose/mmpose/datasets/transforms/common_transforms.py:656: UserWarning: Blur is not pixel-level transformations. Please use with caution.
warnings.warn(
/home/lab/Anaconda_project/mmpose/mmpose/datasets/transforms/common_transforms.py:656: UserWarning: MedianBlur is not pixel-level transformations. Please use with caution.
warnings.warn(
/home/lab/Anaconda_project/mmpose/mmpose/datasets/transforms/common_transforms.py:656: UserWarning: CoarseDropout is not pixel-level transformations. Please use with caution.
warnings.warn(
Traceback (most recent call last):
File "/home/lab/Anaconda_project/mmpose/tools/train.py", line 162, in
main()
File "/home/lab/Anaconda_project/mmpose/tools/train.py", line 158, in main
runner.train()
File "/home/lab/.local/lib/python3.10/site-packages/mmengine/runner/runner.py", line 1728, in train
self._train_loop = self.build_train_loop(
File "/home/lab/.local/lib/python3.10/site-packages/mmengine/runner/runner.py", line 1527, in build_train_loop
loop = EpochBasedTrainLoop(
File "/home/lab/.local/lib/python3.10/site-packages/mmengine/runner/loops.py", line 44, in init
super().init(runner, dataloader)
File "/home/lab/.local/lib/python3.10/site-packages/mmengine/runner/base_loop.py", line 26, in init
self.dataloader = runner.build_dataloader(
File "/home/lab/.local/lib/python3.10/site-packages/mmengine/runner/runner.py", line 1370, in build_dataloader
dataset = DATASETS.build(dataset_cfg)
File "/home/lab/.local/lib/python3.10/site-packages/mmengine/registry/registry.py", line 570, in build
return self.build_func(cfg, *args, **kwargs, registry=self)
File "/home/lab/.local/lib/python3.10/site-packages/mmengine/registry/build_functions.py", line 121, in build_from_cfg
obj = obj_cls(**args) # type: ignore
File "/home/lab/.local/lib/python3.10/site-packages/mmengine/dataset/base_dataset.py", line 247, in init
self.full_init()
File "/home/lab/.local/lib/python3.10/site-packages/mmengine/dataset/base_dataset.py", line 298, in full_init
self.data_list = self.load_data_list()
File "/home/lab/.local/lib/python3.10/site-packages/mmengine/dataset/base_dataset.py", line 440, in load_data_list
raise ValueError('Annotation must have data_list and metainfo '
ValueError: Annotation must have data_list and metainfo keys

Additional information

Hi,I want to use my customize data to train ,and I follow the doc to build file.But when I run train.py , I keep getting [ValueError: Annotation must have data_list and metainfo keys] error.

I thought it might be the problem with my config file,so I changed all the file path to the original coco_wholebody config but remain with dataset_type = 'MyCustomDataset'

out_channels=133,
ann_file='annotations/coco_wholebody_train_v1.0.json',
data_prefix=dict(img='train2017/'),
ann_file='annotations/coco_wholebody_val_v1.0.json',
data_prefix=dict(img='val2017/'),
metainfo=dict(from_file='configs/base/datasets/coco_wholebody.py')

but I still get this error.The way to fix this is to change the dataset_type = 'MyCustomDataset' to dataset_type ='CocoWholeBodyDataset'.But this way I can not train the custom dataset.

Does any one have this problem before?