• https://mmaction2.readthedocs.io/en/latest/advanced_guides/customize_dataset.html

이거 보셈

• Custom Dataset 생성법
  • BaseDataset 상속
    • load_annotations(self), evaluate(self, results, metrics, logger), dump_results(self, results, out) 메소드를 overwrite
  • 위 세 가지 형태로 변환
• 위 과정 수행 후, configuration 수정

After the data pre-processing, the users need to further modify the config files to use the dataset. Here is an example of using a custom dataset in rawframe format.

In configs/task/method/my_custom_config.py:

# dataset settings
dataset_type = 'RawframeDataset'
data_root = 'path/to/your/root'
data_root_val = 'path/to/your/root_val'
ann_file_train = 'data/custom/custom_train_list.txt'
ann_file_val = 'data/custom/custom_val_list.txt'
ann_file_test = 'data/custom/custom_val_list.txt'

data = dict(
    videos_per_gpu=32,
    workers_per_gpu=2,
    train=dict(
        type=dataset_type,
        ann_file=ann_file_train,
        ...),
    val=dict(
        type=dataset_type,
        ann_file=ann_file_val,
        ...),
    test=dict(
        type=dataset_type,
        ann_file=ann_file_test,
        ...))

예시

import copy
import os.path as osp

import mmcv

from .base import BaseDataset
from .builder import DATASETS

@DATASETS.register_module()
class MyDataset(BaseDataset):

    def __init__(self,
                 ann_file,
                 pipeline,
                 data_prefix=None,
                 test_mode=False,
                 filename_tmpl='img_{:05}.jpg'):
        super(MyDataset, self).__init__(ann_file, pipeline, test_mode)

        self.filename_tmpl = filename_tmpl

    def load_annotations(self):
        video_infos = []
        with open(self.ann_file, 'r') as fin:
            for line in fin:
                if line.startswith("directory"):
                    continue
                frame_dir, total_frames, label = line.split(',')
                if self.data_prefix is not None:
                    frame_dir = osp.join(self.data_prefix, frame_dir)
                video_infos.append(
                    dict(
                        frame_dir=frame_dir,
                        total_frames=int(total_frames),
                        label=int(label)))
        return video_infos

    def prepare_train_frames(self, idx):
        results = copy.deepcopy(self.video_infos[idx])
        results['filename_tmpl'] = self.filename_tmpl
        return self.pipeline(results)

    def prepare_test_frames(self, idx):
        results = copy.deepcopy(self.video_infos[idx])
        results['filename_tmpl'] = self.filename_tmpl
        return self.pipeline(results)

    def evaluate(self,
                 results,
                 metrics='top_k_accuracy',
                 topk=(1, 5),
                 logger=None):
        pass

dataset_A_train = dict(
    type='MyDataset',
    ann_file=ann_file_train,
    pipeline=train_pipeline
)

Use mixed datasets for training

MMAction2 also supports to mix dataset for training. Currently it supports to repeat dataset.

Repeat dataset

We use RepeatDataset as wrapper to repeat the dataset. For example, suppose the original dataset as Dataset_A, to repeat it, the config looks like the following

dataset_A_train= dict(
			type='RepeatDataset',
			times=N,
			dataset=dict(# This is the original config of Dataset_Atype='Dataset_A',
			...
			pipeline=train_pipeline
	)
)

새로운 모델 mmengine에 상속하는 방법