Qiankun Liu

这种情况感觉不太对。建议换成inference_inpainting.py这个脚本进行测试，相对比较简单且灵活，具体可以看看readme。

@cahya-wirawan Thanks for your interesting! I have not implemented the inference function for a single image yet. In fact, you can simply put the image and mask into different directories,...

Hi @cahya-wirawan PUT can produce diverse results, but not all produced results are of high quality. You can try to generate more results. In addition, there are some hyper-parameters for...

Hi @cahya-wirawan , `filter_ratio` in the inference script controlls the value of `K`. If it is an integer number and greater than 1, it is the value of `K`. If...

Hi @dongdongdong233 , Thanks for your interests. Sorry for this mistake. The `relative_path` is just used to save the completed results. You can simply modify the config file like the...

Hi @dongdongdong233 ， P-VQVAE is just and auto-encoder, which can not inpaint the images but only reconstruct the images. If you want to inpaint an image, you need to train...

From the logs you provided, it seems that the images are not located correctly, which means 0 images need to be inpainted. You can have a debug about the class...

@zhangbaijin ， You can have a look at [this 12 #issue comment](https://github.com/liuqk3/PUT/issues/12#issuecomment-1299430997). Or you can modify 'inference.py' to load masks from another directory.

@Lecxxx , I think the error is thrown by the reconstruction function. Currently, the pretrained P-VQVAE is not seperately provided. You can try the inpainting function.

@Lecxxx @Hgy12345 While inpainting with inference.py or inference_inpainting.py, the argument `--func inference_inpainting` is required. Please read readme.md carefully.