mirrors
/
mmsegmentation
mirror of https://github.com/open-mmlab/mmsegmentation.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

[Enhancement] Remove batch inference assertion (#3210) 

Thanks for your contribution and we appreciate it a lot. The following
instructions would make your pull request more healthy and more easily
get feedback. If you do not understand some items, don't worry, just
make the pull request and seek help from maintainers.

## Motivation

https://github.com/open-mmlab/mmsegmentation/issues/3181
https://github.com/open-mmlab/mmsegmentation/issues/2965
https://github.com/open-mmlab/mmsegmentation/issues/2644
https://github.com/open-mmlab/mmsegmentation/issues/1645
https://github.com/open-mmlab/mmsegmentation/issues/1444
https://github.com/open-mmlab/mmsegmentation/issues/1370
https://github.com/open-mmlab/mmsegmentation/issues/125

## Modification

Remove the assertion at data_preprocessor

## BC-breaking (Optional)

Does the modification introduce changes that break the
backward-compatibility of the downstream repos?
If so, please describe how it breaks the compatibility and how the
downstream projects should modify their code to keep compatibility with
this PR.

## Use cases (Optional)

If this PR introduces a new feature, it is better to list some use cases
here, and update the documentation.

## Checklist

1. Pre-commit or other linting tools are used to fix the potential lint
issues.
2. The modification is covered by complete unit tests. If not, please
add more unit test to ensure the correctness.
3. If the modification has potential influence on downstream projects,
this PR should be tested with downstream projects, like MMDet or
MMDet3D.
4. The documentation has been modified accordingly, like docstring or
example tutorials.
pull/3223/head
谢昕辰 2023-07-20 09:45:04 +08:00 committed by GitHub
parent 20fa12912f
commit 857f854b61
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 10 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
mmseg/models/data_preprocessor.py
View File

@ -132,9 +132,9 @@ class SegDataPreProcessor(BaseDataPreprocessor):
inputs, data_samples = self.batch_augments( inputs, data_samples = self.batch_augments(
inputs, data_samples) inputs, data_samples)
else: else:
assert len(inputs) == 1, ( img_size = inputs[0].shape[1:]
'Batch inference is not support currently, ' assert all(input_.shape[1:] == img_size for input_ in inputs), \
'as the image size might be different in a batch') 'The image size in a batch should be the same.'
# pad images when testing # pad images when testing
if self.test_cfg: if self.test_cfg:
inputs, padded_samples = stack_batch( inputs, padded_samples = stack_batch(

8
mmseg/models/segmentors/encoder_decoder.py
View File

@ -1,8 +1,10 @@
# Copyright (c) OpenMMLab. All rights reserved. # Copyright (c) OpenMMLab. All rights reserved.
import logging
from typing import List, Optional from typing import List, Optional
import torch.nn as nn import torch.nn as nn
import torch.nn.functional as F import torch.nn.functional as F
from mmengine.logging import print_log
from torch import Tensor from torch import Tensor
from mmseg.registry import MODELS from mmseg.registry import MODELS
@ -330,7 +332,11 @@ class EncoderDecoder(BaseSegmentor):
f'Only "slide" or "whole" test mode are supported, but got ' \ f'Only "slide" or "whole" test mode are supported, but got ' \
f'{self.test_cfg["mode"]}.' f'{self.test_cfg["mode"]}.'
ori_shape = batch_img_metas[0]['ori_shape'] ori_shape = batch_img_metas[0]['ori_shape']
assert all(_['ori_shape'] == ori_shape for _ in batch_img_metas) if not all(_['ori_shape'] == ori_shape for _ in batch_img_metas):
print_log(
'Image shapes are different in the batch.',
logger='current',
level=logging.WARN)
if self.test_cfg.mode == 'slide': if self.test_cfg.mode == 'slide':
seg_logit = self.slide_inference(inputs, batch_img_metas) seg_logit = self.slide_inference(inputs, batch_img_metas)
else: else: