wenmeng zhou
wenmeng zhou
Only deployed models could be used to execute process of loading. Please deploy model before loading
use flow model deploy and use the model_version and model_id returned to load. But I failed in load model with following error: "retcode": 123, "retmsg": "model initialization error, please check...
can you attach the exported checkpoint and give the model info
> > can you attach the exported checkpoint and give the model info > > 请问有比较好的解决方法嘛 看起来你是直接用的训练保存的checkpoint做预测，而没有进行export模型？
The yolox baseline implemented in EasyCV is the first version of Megvii/YOLOX，almost the same version with mmdet/YOLOX， we have noticed there is almost 0.5 mAP gap for yolox-s and yolox-l...
https://modelscope.cn/docs/Configuration%E8%AF%A6%E8%A7%A3#lr_scheduler
模型导出后评测的方式请贴一下？
这个需求在规划中，预计本月开发支持
Could not find cuda drivers on your machine, GPU will not be used. 您的环境cuda没有找到
您好，目前的pipeline主要考虑单场景的跑通，您可以考虑继承或直接修改pipeline对应的__call__函数完成你的定制化需求 https://github.com/modelscope/modelscope/blob/master/modelscope/pipelines/cv/face_detection_pipeline.py#L54