Argument interpolation should be a InterpolationMode or a corresponding Pillow integer constant

在学习使用Stable Diffusion (v1.8.0)，尝试查看生成图片的提示词，就使用图生图中的Interrogate Clip, 但遇到问题。

load checkpoint from D:\DiskH\aiwork\sd-webui-aki-v4\models\BLIP\model_base_caption_capfilt_large.pth
None
*** Error interrogating
    Traceback (most recent call last):
      File "D:\DiskH\aiwork\sd-webui-aki-v4\modules\interrogate.py", line 195, in interrogate
        caption = self.generate_caption(pil_image)
      File "D:\DiskH\aiwork\sd-webui-aki-v4\modules\interrogate.py", line 175, in generate_caption
        gpu_image = transforms.Compose([
      File "D:\DiskH\aiwork\sd-webui-aki-v4\py310\lib\site-packages\torchvision\transforms\transforms.py", line 95, in __call__
        img = t(img)
      File "D:\DiskH\aiwork\sd-webui-aki-v4\py310\lib\site-packages\torch\nn\modules\module.py", line 1511, in _wrapped_call_impl
        return self._call_impl(*args, **kwargs)
      File "D:\DiskH\aiwork\sd-webui-aki-v4\py310\lib\site-packages\torch\nn\modules\module.py", line 1520, in _call_impl
        return forward_call(*args, **kwargs)
      File "D:\DiskH\aiwork\sd-webui-aki-v4\py310\lib\site-packages\torchvision\transforms\transforms.py", line 354, in forward
        return F.resize(img, self.size, self.interpolation, self.max_size, self.antialias)
      File "D:\DiskH\aiwork\sd-webui-aki-v4\py310\lib\site-packages\torchvision\transforms\functional.py", line 439, in resize
        raise TypeError(
    TypeError: Argument interpolation should be a InterpolationMode or a corresponding Pillow integer constant

---

查了一堆的文件，Pillow 也升级到最新版本，但没有解决问题。最后，实属无奈只能查看一下代码。

sd-webui-aki-v4\modules\interrogate.py

    def generate_caption(self, pil_image):
        gpu_image = transforms.Compose([
            transforms.Resize((blip_image_eval_size, blip_image_eval_size), interpolation=InterpolationMode.BILINEAR),
            transforms.ToTensor(),
            transforms.Normalize((0.48145466, 0.4578275, 0.40821073), (0.26862954, 0.26130258, 0.27577711))
        ])(pil_image).unsqueeze(0).type(self.dtype).to(devices.device_interrogate)

        with torch.no_grad():
            caption = self.blip_model.generate(gpu_image, sample=False, num_beams=shared.opts.interrogate_clip_num_beams, min_length=shared.opts.interrogate_clip_min_length, max_length=shared.opts.interrogate_clip_max_length)

        return caption[0]

在这个文件的前端添加引用

import torchvision.transforms.functional as F

    def generate_caption(self, pil_image):
        gpu_image = transforms.Compose([
            transforms.Resize((blip_image_eval_size, blip_image_eval_size), interpolation=F.InterpolationMode.BILINEAR),
            transforms.ToTensor(),
            transforms.Normalize((0.48145466, 0.4578275, 0.40821073), (0.26862954, 0.26130258, 0.27577711))
        ])(pil_image).unsqueeze(0).type(self.dtype).to(devices.device_interrogate)

        with torch.no_grad():
            caption = self.blip_model.generate(gpu_image, sample=False, num_beams=shared.opts.interrogate_clip_num_beams, min_length=shared.opts.interrogate_clip_min_length, max_length=shared.opts.interrogate_clip_max_length)

        return caption[0]

修改后，这个问题就解决了。
祝大家生活快乐！