mmedit.engine.runner

Package Contents

Classes

EditTestLoop	Test loop for MMEditing models. This class support evaluate:
EditValLoop	Validation loop for MMEditing models. This class support evaluate:
EditLogProcessor	EditLogProcessor inherits from mmengine.runner.LogProcessor and

class mmedit.engine.runner.EditTestLoop(runner, dataloader, evaluator, fp16=False)

Bases: mmengine.runner.base_loop.BaseLoop

Test loop for MMEditing models. This class support evaluate:

1. Metrics (metric) on a single dataset (e.g. PSNR and SSIM on DIV2K dataset)

2. Different metrics on different datasets (e.g. PSNR on DIV2K and SSIM and PSNR on SET5)

Use cases:

Case 1: metrics on a single dataset

>>> # add the following lines in your config
>>> # 1. use `EditTestLoop` instead of `TestLoop` in MMEngine
>>> val_cfg = dict(type='EditTestLoop')
>>> # 2. specific EditEvaluator instead of Evaluator in MMEngine
>>> test_evaluator = dict(
>>>     type='EditEvaluator',
>>>     metrics=[
>>>         dict(type='PSNR', crop_border=2, prefix='Set5'),
>>>         dict(type='SSIM', crop_border=2, prefix='Set5'),
>>>     ])
>>> # 3. define dataloader
>>> test_dataloader = dict(...)

Case 2: different metrics on different datasets

>>> # add the following lines in your config
>>> # 1. use `EditTestLoop` instead of `TestLoop` in MMEngine
>>> Test_cfg = dict(type='EditTestLoop')
>>> # 2. specific a list EditEvaluator
>>> # do not forget to add prefix for each metric group
>>> div2k_evaluator = dict(
>>>     type='EditEvaluator',
>>>     metrics=dict(type='SSIM', crop_border=2, prefix='DIV2K'))
>>> set5_evaluator = dict(
>>>     type='EditEvaluator',
>>>     metrics=[
>>>         dict(type='PSNR', crop_border=2, prefix='Set5'),
>>>         dict(type='SSIM', crop_border=2, prefix='Set5'),
>>>     ])
>>> # define evaluator config
>>> test_evaluator = [div2k_evaluator, set5_evaluator]
>>> # 3. specific a list dataloader for each metric groups
>>> div2k_dataloader = dict(...)
>>> set5_dataloader = dict(...)
>>> # define dataloader config
>>> test_dataloader = [div2k_dataloader, set5_dataloader]

参数

• runner (Runner) – A reference of runner.

• dataloader (Dataloader or dict or list) – A dataloader object or a dict to build a dataloader a list of dataloader object or a list of config dicts.

• evaluator (Evaluator or dict or list) – A evaluator object or a dict to build the evaluator or a list of evaluator object or a list of config dicts.

property total_length: int

_build_dataloaders(dataloader: DATALOADER_TYPE) → List[torch.utils.data.DataLoader]

Build dataloaders.

参数

dataloader (Dataloader or dict or list) – A dataloader object or a dict to build a dataloader a list of dataloader object or a list of config dict.

返回

List of dataloaders for compute metrics.

返回类型

List[Dataloader]

_build_evaluators(evaluator: EVALUATOR_TYPE) → List[mmengine.evaluator.Evaluator]

Build evaluators.

参数

evaluator (Evaluator or dict or list) – A evaluator object or a dict to build the evaluator or a list of evaluator object or a list of config dicts.

返回

List of evaluators for compute metrics.

返回类型

List[Evaluator]

run()

Launch validation. The evaluation process consists of four steps.

1. Prepare pre-calculated items for all metrics by calling self.evaluator.prepare_metrics().

2. Get a list of metrics-sampler pair. Each pair contains a list of metrics with the same sampler mode and a shared sampler.

3. Generate images for the each metrics group. Loop for elements in each sampler and feed to the model as input by calling self.run_iter().

4. Evaluate all metrics by calling self.evaluator.evaluate().

run_iter(idx, data_batch: dict, metrics: Sequence[mmengine.evaluator.BaseMetric])

Iterate one mini-batch and feed the output to corresponding metrics.

参数

• idx (int) – Current idx for the input data.

• data_batch (dict) – Batch of data from dataloader.

• metrics (Sequence[BaseMetric]) – Specific metrics to evaluate.

class mmedit.engine.runner.EditValLoop(runner, dataloader: DATALOADER_TYPE, evaluator: EVALUATOR_TYPE, fp16: bool = False)

Bases: mmengine.runner.base_loop.BaseLoop

Validation loop for MMEditing models. This class support evaluate:

1. Metrics (metric) on a single dataset (e.g. PSNR and SSIM on DIV2K dataset)

2. Different metrics on different datasets (e.g. PSNR on DIV2K and SSIM and PSNR on SET5)

Use cases:

Case 1: metrics on a single dataset

>>> # add the following lines in your config
>>> # 1. use `EditValLoop` instead of `ValLoop` in MMEngine
>>> val_cfg = dict(type='EditValLoop')
>>> # 2. specific EditEvaluator instead of Evaluator in MMEngine
>>> val_evaluator = dict(
>>>     type='EditEvaluator',
>>>     metrics=[
>>>         dict(type='PSNR', crop_border=2, prefix='Set5'),
>>>         dict(type='SSIM', crop_border=2, prefix='Set5'),
>>>     ])
>>> # 3. define dataloader
>>> val_dataloader = dict(...)

Case 2: different metrics on different datasets

>>> # add the following lines in your config
>>> # 1. use `EditValLoop` instead of `ValLoop` in MMEngine
>>> val_cfg = dict(type='EditValLoop')
>>> # 2. specific a list EditEvaluator
>>> # do not forget to add prefix for each metric group
>>> div2k_evaluator = dict(
>>>     type='EditEvaluator',
>>>     metrics=dict(type='SSIM', crop_border=2, prefix='DIV2K'))
>>> set5_evaluator = dict(
>>>     type='EditEvaluator',
>>>     metrics=[
>>>         dict(type='PSNR', crop_border=2, prefix='Set5'),
>>>         dict(type='SSIM', crop_border=2, prefix='Set5'),
>>>     ])
>>> # define evaluator config
>>> val_evaluator = [div2k_evaluator, set5_evaluator]
>>> # 3. specific a list dataloader for each metric groups
>>> div2k_dataloader = dict(...)
>>> set5_dataloader = dict(...)
>>> # define dataloader config
>>> val_dataloader = [div2k_dataloader, set5_dataloader]

参数

• runner (Runner) – A reference of runner.

• dataloader (Dataloader or dict or list) – A dataloader object or a dict to build a dataloader a list of dataloader object or a list of config dicts.

• evaluator (Evaluator or dict or list) – A evaluator object or a dict to build the evaluator or a list of evaluator object or a list of config dicts.

property total_length: int

_build_dataloaders(dataloader: DATALOADER_TYPE) → List[torch.utils.data.DataLoader]

Build dataloaders.

参数

dataloader (Dataloader or dict or list) – A dataloader object or a dict to build a dataloader a list of dataloader object or a list of config dict.

返回

List of dataloaders for compute metrics.

返回类型

List[Dataloader]

_build_evaluators(evaluator: EVALUATOR_TYPE) → List[mmengine.evaluator.Evaluator]

Build evaluators.

参数

evaluator (Evaluator or dict or list) – A evaluator object or a dict to build the evaluator or a list of evaluator object or a list of config dicts.

返回

List of evaluators for compute metrics.

返回类型

List[Evaluator]

run()

Launch validation. The evaluation process consists of four steps.

1. Prepare pre-calculated items for all metrics by calling self.evaluator.prepare_metrics().

2. Get a list of metrics-sampler pair. Each pair contains a list of metrics with the same sampler mode and a shared sampler.

3. Generate images for the each metrics group. Loop for elements in each sampler and feed to the model as input by calling self.run_iter().

4. Evaluate all metrics by calling self.evaluator.evaluate().

run_iter(idx, data_batch: dict, metrics: Sequence[mmengine.evaluator.BaseMetric])

Iterate one mini-batch and feed the output to corresponding metrics.

参数

• idx (int) – Current idx for the input data.

• data_batch (dict) – Batch of data from dataloader.

• metrics (Sequence[BaseMetric]) – Specific metrics to evaluate.

class mmedit.engine.runner.EditLogProcessor(window_size=10, by_epoch=True, custom_cfg: Optional[List[dict]] = None, num_digits: int = 4, log_with_hierarchy: bool = False)

Bases: mmengine.runner.LogProcessor

EditLogProcessor inherits from mmengine.runner.LogProcessor and overwrites self.get_log_after_iter().

This log processor should be used along with mmedit.engine.runner.EditValLoop and mmedit.engine.runner.EditTestLoop.

_get_dataloader_size(runner, mode) → int

Get dataloader size of current loop. In EditValLoop and EditTestLoop, we use total_length instead of len(dataloader) to denote the total number of iterations.

参数

• runner (Runner) – The runner of the training/validation/testing

• mode (str) – Current mode of runner.

返回

The dataloader size of current loop.

返回类型

int