3.3 KiB

Raw Blame History

Tutorial 3: Custom Data Pipelines

Design of Data pipelines

Following typical conventions, we use Dataset and DataLoader for data loading with multiple workers. Dataset returns a dict of data items corresponding to the arguments of models' forward method.

The data preparation pipeline and the dataset is decomposed. Usually a dataset defines how to process the annotations and a data pipeline defines all the steps to prepare a data dict. A pipeline consists of a sequence of operations. Each operation takes a dict as input and also output a dict for the next transform.

The operations are categorized into data loading, pre-processing and formatting.

Here is an pipeline example for ResNet-50 training on ImageNet.

img_norm_cfg = dict(
    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='RandomResizedCrop', size=224),
    dict(type='RandomFlip', flip_prob=0.5, direction='horizontal'),
    dict(type='Normalize', **img_norm_cfg),
    dict(type='ImageToTensor', keys=['img']),
    dict(type='ToTensor', keys=['gt_label']),
    dict(type='Collect', keys=['img', 'gt_label'])
]
test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='Resize', size=256),
    dict(type='CenterCrop', crop_size=224),
    dict(type='Normalize', **img_norm_cfg),
    dict(type='ImageToTensor', keys=['img']),
    dict(type='ToTensor', keys=['gt_label']),
    dict(type='Collect', keys=['img', 'gt_label'])
]

For each operation, we list the related dict fields that are added/updated/removed. At the end of the pipeline, we use Collect to only retain the necessary items for forward computation.

Data loading

LoadImageFromFile

add: img, img_shape, ori_shape

Pre-processing

Resize

add: scale, scale_idx, pad_shape, scale_factor, keep_ratio
update: img, img_shape

RandomFlip

add: flip
update: img

RandomCrop

update: img, pad_shape

Normalize

add: img_norm_cfg
update: img

Formatting

ToTensor

update: specified by keys.

ImageToTensor

update: specified by keys.

Transpose

update: specified by keys.

Collect

remove: all other keys except for those specified by keys

Extend and use custom pipelines

Write a new pipeline in any file, e.g., my_pipeline.py. It takes a dict as input and return a dict.

from mmcls.datasets import PIPELINES

@PIPELINES.register_module()
class MyTransform(object):

    def __call__(self, results):
        results['dummy'] = True
        # apply transforms on results['img']
        return results

Import the new class.
```
from .my_pipeline import MyTransform
```

Use it in config files.

img_norm_cfg = dict(
    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='RandomResizedCrop', size=224),
    dict(type='RandomFlip', flip_prob=0.5, direction='horizontal'),
    dict(type='MyTransform'),
    dict(type='Normalize', **img_norm_cfg),
    dict(type='ImageToTensor', keys=['img']),
    dict(type='ToTensor', keys=['gt_label']),
    dict(type='Collect', keys=['img', 'gt_label'])
]

3.3 KiB Raw Blame History