1
我有一个预先训练好的MobileNetSSd.caffemodelmodel
文件MobileNetSSd.prototxt
。现在我想将这些部署到管道中的 Azure ML
首先,我尝试使用以下代码注册这两个模型:
#Register the model
caffemodel = Model.register(model_path = "mobilenet_ssd/MobileNetSSD.caffemodel", #local path
model_name = "caffemodel",
tags = {"severity": "1"},
description = "object detection caffemodel",
workspace = ws)
prototxt = Model.register(model_path = "mobilenet_ssd/MobileNetSSD.prototxt",
model_name = "prototxt",
tags = {"severity": "1"},
description = "object detection prototxt",
workspace = ws)
文件注册成功,Home/Models
目录下可见
现在我需要阅读这些模型,但我无法阅读它们。
我尝试使用以下代码,但它给了我一个错误,如下所述:
from azureml.core.model import Model
global caffemodel
global prototxt
caffemodel = Model.get_model_path('caffemodel')
prototxt = Model.get_model_path('prototxt')
#错误:
826 # Probing azureml-models/<name>
827 if not os.path.exists(candidate_model_path):
--> 828 return Model._get_model_path_local_from_root(model_name)
829 else:
830 # Probing azureml-models/<name> exists, probing version
/anaconda/envs/azureml_py36/lib/python3.6/site-packages/azureml/core/model.py in _get_model_path_local_from_root(model_name)
870 raise ModelNotFoundException("Model {} not found in cache at {} or in current working directory {}. "
871 "For more info, set logging level to DEBUG.".format(model_name, MODELS_DIR,
--> 872 os.getcwd()))
873
874 @staticmethod
ModelNotFoundException: ModelNotFoundException:
Message: Model caffemodel not found in cache at azureml-models or in current working directory /mnt/batch/tasks/shared/LS_root/mounts/clusters/jsing2091/code/Users/jsing209/Computer_Vision/Realtime_People_Counting. For more info, set logging level to DEBUG.
InnerException None
ErrorResponse
{
"error": {
"message": "Model caffemodel not found in cache at azureml-models or in current working directory /mnt/batch/tasks/shared/LS_root/mounts/clusters/jsing2091/code/Users/jsing209/Computer_Vision/Realtime_People_Counting. For more info, set logging level to DEBUG."
}
}
我不知道为什么它找不到模型......我有点迷茫,如果有人可以帮我阅读这些模型,我将不胜感激