About how to run rembg using a self trained model on Windows

[LINHYYY]

trafficstars

When I trained my U2net model and wanted to use it to run Rembg, I initially used the method described by the author in README.md rembg i -m u2net_custom -x '{"model_path": "~/.u2net/u2net.onnx"}' path/to/input.png path/to/output.png

But I encountered the following error: Traceback (most recent call last): File "D:\Anaconda3\envs\rembg\lib\runpy.py", line 197, in _run_module_as_main return _run_code(code, main_globals, None, File "D:\Anaconda3\envs\rembg\lib\runpy.py", line 87, in run_code exec(code, run_globals) File "D:\Anaconda3\envs\rembg\Scripts\rembg.exe_main.py", line 4, in from rembg.cli import main File "D:\Anaconda3\envs\rembg\lib\site-packages\rembg\cli.py", line 16, in _main() File "D:\Anaconda3\envs\rembg\lib\site-packages\click\core.py", line 1157, in call return self.main(*args, **kwargs) File "D:\Anaconda3\envs\rembg\lib\site-packages\click\core.py", line 1078, in main rv = self.invoke(ctx) File "D:\Anaconda3\envs\rembg\lib\site-packages\click\core.py", line 1688, in invoke return _process_result(sub_ctx.command.invoke(sub_ctx)) File "D:\Anaconda3\envs\rembg\lib\site-packages\click\core.py", line 1434, in invoke return ctx.invoke(self.callback, **ctx.params) File "D:\Anaconda3\envs\rembg\lib\site-packages\click\core.py", line 783, in invoke return __callback(*args, **kwargs) File "D:\Anaconda3\envs\rembg\lib\site-packages\rembg\commands\i_command.py", line 108, in i_command output.write(remove(input.read(), session=new_session(model, **kwargs), **kwargs)) File "D:\Anaconda3\envs\rembg\lib\site-packages\rembg\session_factory.py", line 44, in new_session return session_class(model_name, sess_opts, providers, *args, **kwargs) File "D:\Anaconda3\envs\rembg\lib\site-packages\rembg\sessions\u2net_custom.py", line 39, in init raise ValueError("model_path is required") ValueError: model_path is required

I have found that this issue has also been mentioned in many issues, such as：issues #629 issues #636

It seems to be caused by the inconsistency between Linux system commands and Windows system commands. The author mentioned Linux commands in README.md, but after my attempts, I found that the correct command for Windows is: rembg i -m u2net_custom -x "{\"model_path\": \"C:/your model path/your model name.onnx\"}" ./images/input.jpg ./images/output.jpg I have successfully used the model I trained myself, it need to pay attention to the '\' '/' issue between the model path and the image path. I hope it can help people who are also facing this problem.