inferencer([['demo_test/t0.jpg', 'demo_test/t1.jpg']], show=True, out_dir='demo_test')
报错如下:
Loads checkpoint by local backend from path: pretrained/ChangerEx_r18-512x512_40k_levircd_20221223_120511.pth
C:\Users\vc.conda\envs\openc\lib\site-packages\mmseg\models\decode_heads\decode_head.py:120: UserWarning: For binary segmentation, we suggest usingout_channels = 1 to define the outputchannels of segmentor, and use thresholdto convert seg_logits into a predictionapplying a threshold
warnings.warn('For binary segmentation, we suggest using'
C:\Users\vc.conda\envs\openc\lib\site-packages\mmseg\models\builder.py:36: UserWarning: build_loss would be deprecated soon, please use mmseg.registry.MODELS.build()
warnings.warn('build_loss would be deprecated soon, please use '
C:\Users\vc.conda\envs\openc\lib\site-packages\mmseg\models\losses\cross_entropy_loss.py:250: UserWarning: Default avg_non_ignore is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set avg_non_ignore=True.
warnings.warn(
C:\Users\vc.conda\envs\openc\lib\site-packages\mmseg\structures\sampler\builder.py:11: UserWarning: `build_pi war warnings.warn( 08/25 00:02:27 - mmengine - WARNING - Failed to search registry with scope "opencd" in the "function" registry tree. As a workaround, the current "function" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "opencd" is a correct scope, or whether the registry is initialized. C:\Users\vc\.conda\envs\openc\lib\site-packages\mmengine\visualization\visualizer.py:196: UserWarning: Failed to add <class 'opencd.visualization.cd_vis_backend.CDLocalVisBackend'>, please provide thesave_dir` argument.
你好,根据您提供的推理代码进行单个图像对的推理,但是得到的结果是全黑色,想问一下如何解决: from opencd.apis import OpenCDInferencer inferencer = OpenCDInferencer(model='configs/changer/changer_ex_r18_512x512_40k_levircd.py', weights='pretrained/ChangerEx_r18-512x512_40k_levircd_20221223_120511.pth', classes=('unchanged', 'changed'), palette=[[0, 0, 0], [255, 255, 255]])
inferencer([['t0.jpg', 't1.jpg']], show=True, out_dir='demo_test')
inferencer([['demo_test/t0.jpg', 'demo_test/t1.jpg']], show=True, out_dir='demo_test') 报错如下: Loads checkpoint by local backend from path: pretrained/ChangerEx_r18-512x512_40k_levircd_20221223_120511.pth C:\Users\vc.conda\envs\openc\lib\site-packages\mmseg\models\decode_heads\decode_head.py:120: UserWarning: For binary segmentation, we suggest using
out_channels = 1
to define the outputchannels of segmentor, and usethreshold
to convertseg_logits
into a predictionapplying a threshold warnings.warn('For binary segmentation, we suggest using' C:\Users\vc.conda\envs\openc\lib\site-packages\mmseg\models\builder.py:36: UserWarning:build_loss
would be deprecated soon, please usemmseg.registry.MODELS.build()
warnings.warn('build_loss
would be deprecated soon, please use ' C:\Users\vc.conda\envs\openc\lib\site-packages\mmseg\models\losses\cross_entropy_loss.py:250: UserWarning: Defaultavg_non_ignore
is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, setavg_non_ignore=True
. warnings.warn( C:\Users\vc.conda\envs\openc\lib\site-packages\mmseg\structures\sampler\builder.py:11: UserWarning:`build_pi war warnings.warn( 08/25 00:02:27 - mmengine - WARNING - Failed to search registry with scope "opencd" in the "function" registry tree. As a workaround, the current "function" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "opencd" is a correct scope, or whether the registry is initialized. C:\Users\vc\.conda\envs\openc\lib\site-packages\mmengine\visualization\visualizer.py:196: UserWarning: Failed to add <class 'opencd.visualization.cd_vis_backend.CDLocalVisBackend'>, please provide the
save_dir` argument.