• Docs >
  • Module code >
  • mmedit.models.backbones.encoder_decoders.two_stage_encoder_decoder

Source code for mmedit.models.backbones.encoder_decoders.two_stage_encoder_decoder

# Copyright (c) OpenMMLab. All rights reserved.
import torch
import torch.nn as nn
from mmcv.cnn import constant_init, normal_init
from mmcv.runner import auto_fp16, load_checkpoint
from mmcv.utils.parrots_wrapper import _BatchNorm

from mmedit.models.builder import build_backbone, build_component
from mmedit.models.registry import BACKBONES
from mmedit.utils import get_root_logger

[docs]@BACKBONES.register_module() class DeepFillEncoderDecoder(nn.Module): """Two-stage encoder-decoder structure used in DeepFill model. The details are in: Generative Image Inpainting with Contextual Attention Args: stage1 (dict): Config dict for building stage1 model. As DeepFill model uses Global&Local model as baseline in first stage, the stage1 model can be easily built with `GLEncoderDecoder`. stage2 (dict): Config dict for building stage2 model. return_offset (bool): Whether to return offset feature in contextual attention module. Default: False. """ def __init__(self, stage1=dict( type='GLEncoderDecoder', encoder=dict(type='DeepFillEncoder'), decoder=dict(type='DeepFillDecoder', in_channels=128), dilation_neck=dict( type='GLDilationNeck', in_channels=128, act_cfg=dict(type='ELU'))), stage2=dict(type='DeepFillRefiner'), return_offset=False): super().__init__() self.stage1 = build_backbone(stage1) self.stage2 = build_component(stage2) self.return_offset = return_offset # support fp16 self.fp16_enabled = False
[docs] @auto_fp16() def forward(self, x): """Forward function. Args: x (torch.Tensor): This input tensor has the shape of (n, 5, h, w). In channel dimension, we concatenate [masked_img, ones, mask] as DeepFillv1 models do. Returns: tuple[torch.Tensor]: The first two item is the results from first \ and second stage. If set `return_offset` as True, the offset \ will be returned as the third item. """ input_x = x.clone() masked_img = input_x[:, :3, ...] mask = input_x[:, -1:, ...] x = self.stage1(x) stage1_res = x.clone() stage1_img = stage1_res * mask + masked_img * (1. - mask) stage2_input = torch.cat([stage1_img, input_x[:, 3:, ...]], dim=1) stage2_res, offset = self.stage2(stage2_input, mask) if self.return_offset: return stage1_res, stage2_res, offset return stage1_res, stage2_res
# TODO: study the effects of init functions
[docs] def init_weights(self, pretrained=None): """Init weights for models. Args: pretrained (str, optional): Path for pretrained weights. If given None, pretrained weights will not be loaded. Defaults to None. """ if isinstance(pretrained, str): logger = get_root_logger() load_checkpoint(self, pretrained, strict=False, logger=logger) elif pretrained is None: for m in self.modules(): if isinstance(m, nn.Conv2d): normal_init(m, 0, 0.02) elif isinstance(m, (_BatchNorm, nn.InstanceNorm2d)): constant_init(m, 1) else: raise TypeError('pretrained must be a str or None but' f' got {type(pretrained)} instead.')
Read the Docs v: v0.13.0
On Read the Docs
Project Home

Free document hosting provided by Read the Docs.