当前环境信息:
在mac上安装好stable diffusion后,能做图片生成了之后,遇到一些图片需要做提示词反推,这个时候需要下载一个插件,参考:
https://gitcode.***/ranting8323/stable-diffusion-webui-wd14-tagger
安装完成后,可以在这个位置使用,这里是给了一个图片来反推提示词。
使用过程中遇到了一些错误记录如下:
Loading wd14-vit-v2-git model file from SmilingWolf/wd-v1-4-vit-tagger-v2
'(MaxRetryError("HTTPSConnectionPool(host='huggingface.co', port=443): Max retries exceeded with url: /SmilingWolf/wd-v1-4-vit-tagger-v2/resolve/main/model.onnx (Caused by ConnectTimeoutError(<urllib3.connection.HTTPSConnection object at 0x1622b***40>, 'Connection to huggingface.co timed out. (connect timeout=10)'))"), '(Request ID: 008f823b-7a19-4899-bbed-3bacbe0bfee2)')' thrown while requesting HEAD https://huggingface.co/SmilingWolf/wd-v1-4-vit-tagger-v2/resolve/main/model.onnx
这个错误是访问huggingface被拦截,需要科学上网。
还有一个错误,安装onnxruntime失败:
RuntimeError: Couldn't install onnxruntime.
***mand: "/Users/xxxx/aigc/stable-diffusion-webui/venv/bin/python3" -m pip install onnxruntime-gpu --prefer-binary
Error code: 1
stdout: Looking in indexes: https://mirrors.aliyun.***/pypi/simple/
我这台电脑的解决方案是,修改webui的代码,如下:
文件名:interrogator.py
def load(self) -> None:
model_path, tags_path = self.download()
# only one of these packages should be installed at a time in any one environment
# https://onnxruntime.ai/docs/get-started/with-python.html#install-onnx-runtime
# TODO: remove old package when the environment changes?
from launch import is_installed, run_pip
if not is_installed('onnxruntime'):
package = os.environ.get(
'ONNXRUNTIME_PACKAGE',
# 'onnxruntime-gpu'
'onnxruntime'
)
修改代码后重新启动 webui,就能正常的反推提示词了;
将反推的提示词重新生成图片,效果如下: