czczup
commented
1 year ago 你好，权重载入的提示是不影响的，
RuntimeError: The size of tensor a (257) must match the size of tensor b (1025) at non-singleton dimension 3报这个错是因为输入图像的分辨率不对，请问你推理用的图片是数据集里的吗，还是自己的数据？ 我在postdom的config里写的配置是：
test_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='MultiScaleFlipAug',
img_scale=(2048, 512),
# img_ratios=[0.5, 0.75, 1.0, 1.25, 1.5, 1.75],
flip=False,
transforms=[
dict(type='Resize', keep_ratio=True),
dict(type='ResizeToMultiple', size_divisor=32),
dict(type='RandomFlip'),
dict(type='Normalize', **img_norm_cfg),
dict(type='ImageToTensor', keys=['img']),
dict(type='Collect', keys=['img']),
])
]这要求图像的长宽比必须小于2048/512=4，才能保证短边是512（因为BEiT要求512x512的固定大小的输入）。
所以只需把img_scale=(2048, 512),改成img_scale=(9999, 512),即可
AlanCSU
wobukun123
37s
首先很感谢作者您对于图像分割的贡献以及开源代码，但是我在预测Potsdam数据集的时候存在一些小问题，运行代码如下： python segmentation/image_demo.py segmentation/configs/potsdam/mask2former_beit_adapter_large_512_80k_potsdam_ss.py D:\PyCharm_Projects\ViT-Adapter-main\segmentation\pretrained_model\beit_large_patch16_224_pt22k_ft22k.pth D:/downloads/Potsdam/Potsdam/myOutputs/images/2_10_0_0.png
但是这里运行时报错显示： unexpected key in source state_dict: model missing keys in source state_dict: backbone.cls_token, backbone.level_embed, backbone.patch_embed.proj.weight, backbone.patch_embed.proj.bias, backbone.blocks.0.gamma_1, backbone.blocks.0.gamma_2, ... 等一系列backbone,decode_head开头的一系列权重载入
其次还存在这个问题：
Traceback (most recent call last): File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\image_demo.py", line 58, in
main()
File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\image_demo.py", line 45, in main
result = inference_segmentor(model, args.img)
File "d:\downloads\mmsegmentation-master\mmsegmentation-master\mmseg\apis\inference.py", line 102, in inference_segmentor
result = model(return_loss=False, rescale=True, data)
File "E:\conda\lib\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, *kwargs)
File "E:\conda\lib\site-packages\mmcv\runner\fp16_utils.py", line 116, in new_func
return old_func(args, kwargs)
File "d:\downloads\mmsegmentation-master\mmsegmentation-master\mmseg\models\segmentors\base.py", line 110, in forward
return self.forward_test(img, img_metas, kwargs)
File "d:\downloads\mmsegmentation-master\mmsegmentation-master\mmseg\models\segmentors\base.py", line 94, in forward_test
return self.aug_test(imgs, img_metas, kwargs)
File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\mmseg_custom\models\segmentors\encoder_decoder_mask2former.py", line 276, in aug_test
seg_logit = self.inference(imgs[0], img_metas[0], rescale)
File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\mmseg_custom\models\segmentors\encoder_decoder_mask2former.py", line 240, in inference
seg_logit = self.slide_inference(img, img_meta, rescale)
File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\mmseg_custom\models\segmentors\encoder_decoder_mask2former.py", line 180, in slide_inference
crop_seg_logit = self.encode_decode(crop_img, img_meta)
File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\mmseg_custom\models\segmentors\encoder_decoder_mask2former.py", line 73, in encode_decode
x = self.extract_feat(img)
File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\mmseg_custom\models\segmentors\encoder_decoder_mask2former.py", line 65, in extract_feat
x = self.backbone(img)
File "E:\conda\lib\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, kwargs)
File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\mmseg_custom\models\backbones\beit_adapter.py", line 115, in forward
x, c, cls = layer(x, c, cls, self.blocks[indexes[0]:indexes[-1] + 1],
File "E:\conda\lib\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, *kwargs)
File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\mmseg_custom\models\backbones\adapter_modules.py", line 222, in forward
x = blk(x, H, W)
File "E:\conda\lib\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(input, kwargs)
File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\mmseg_custom\models\backbones\base\beit.py", line 186, in forward
x = _inner_forward(x)
File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\mmseg_custom\models\backbones\base\beit.py", line 179, in _inner_forward
x = x + self.drop_path(self.gamma_1 self.attn(self.norm1(x), rel_pos_bias=rel_pos_bias))
File "E:\conda\lib\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(input, **kwargs)
File "D:\PyCharm_Projects\ViT-Adapter-main\segmentation\mmseg_custom\models\backbones\base\beit.py", line 136, in forward
attn = attn + relative_position_bias.unsqueeze(0)
RuntimeError: The size of tensor a (257) must match the size of tensor b (1025) at non-singleton dimension 3
如果您有时间可以给我一些指点，我将非常感谢