zongshuai zhang
zongshuai zhang
有人解决了吗
谢谢大佬，标题文本有了
> pip install -e . install camke https://blog.csdn.net/qq_45505081/article/details/130573708 and then build locally.
Can any body reproduce this bug?
> [#729 (comment)](https://github.com/FunAudioLLM/CosyVoice/issues/729#issuecomment-2545399338) > > * [CosyVoice2-0.5B does not have spk2info.pt #729](https://github.com/FunAudioLLM/CosyVoice/issues/729) > > It appears that you're encountering an issue with the absence of the spk2info.pt file in the...
> 报错的意思应该是说inputs里面没有 'x2paddle_images'这个字段，inputs中包含的字段为'images', 'scale_fcator', 'im_shape'这三个，我看你的onnx model里面只有images一个输入，但是转换成的paddle格式中却包含x2paddle_images这个名字的输入，你可以用netron可视化一下model.pdmodel看一下转换成的paddle模型中输入的名字 我也是先把yolov8的模型转化为onnx格式，然后用x2paddle将模型转化为paddle格式，也出现了KeyError: 'x2paddle_images'这个错误 import paddle from paddle.inference import Config, create_predictor config = Config("weights/yolov8/model.pdmodel", "weights/yolov8/model.pdiparams") predictor = create_predictor(config) input_names = predictor.get_input_names() print("Model Input Names:", input_names) 这是我用来测试模型的，其结果为Model...
我按照GPU文档里的第十条10. Enable CUDA Acceleration for OCR `python -m pip install paddlepaddle-gpu==3.0.0b1 -i https://www.paddlepaddle.org.cn/packages/stable/cu118/` 就可以运行了
>  我拿到首包要12秒左右，你们也是这么慢吗 inference one shot是不是没加stream=True