重新训练和要使用的Blob令牌 [英] Retraining and blob token to use

问题描述

嗨！
我有一个天蓝色的再培训问题.

我一直在按照指示进行再培训. C#，并使用了代码用于"OverwriteModel"的示例，这是在示例中提出的.
(例如，在https://docs.microsoft.com/zh-CN/azure/machine-learning/studio/retrain-a-classic-web-service中)

我非常确定BaseLocation，RelativeLocation，endpointURL，patchURL等是正确的(在"Microsoft Azure机器学习Web服务"中检索). 但是:我不确定我使用的是正确的SAS Blob令牌.现在，我在上传新训练数据的结果中使用令牌.

上传新数据后，容器中的所有内容似乎都更新了blob. RetrainOutput.ilearner似乎已更新(但是它是二进制文件，所以我无法读取).

在OverwriteModel模型中，它看起来像这样.

--- 8< --------------------------------------------- -------------------------------------------------- --------
使用(var request = new HttpRequestMessage(new HttpMethod("PATCH")，RetrainingEndpointPatchUrl))
                {
                    request.Content = new StringContent(JsonConvert.SerializeObject(resourceLocations)，System.Text.Encoding.UTF8，"application/json");
                    HttpResponseMessage response =等待客户端.SendAsync(请求);

                   如果(！response.IsSuccessStatusCode)
                    {
                       等待WriteFailedResponse(response);                        
                    }

                    //通过成功的响应在这里做您想做的事情.
                }

-------------------------------------------------- ---------------------------------------------> 8- --------

响应是该请求.内容为空:请求失败，状态码为:NoContent"

我唯一的猜测是SAS blob令牌不正确(这是上面的resourceLocations的一部分)，但是那我不应该再遇到另一个错误吗?

Hi!
I have a retraining problem in azure.

I have been following the instruction to make a retraining from  C# and used the code  for "OverwriteModel", that is proposed in the examples for doing this.
(e.g in https://docs.microsoft.com/en-us/azure/machine-learning/studio/retrain-a-classic-web-service)

I am pretty sure the BaseLocation, RelativeLocation, endpointURL, patchURL etc are correct (Retrieved in "Microsoft Azure Machine Learning Web Services"). 
However : I not so sure I use the correct SAS blob token. As it is now I use the token from the result where I upload the new training data.

After uploaded the new data everything seems fine in the container updated the blobs. The RetrainOutput.ilearner seems to be updated (however it is binary file so I cant read it).

In the OverwriteModel modelu it looks something like this.

--- 8<-------------------------------------------------------------------------------------------------------
using (var request = new HttpRequestMessage(new HttpMethod("PATCH"), RetrainingEndpointPatchUrl))
                {
                    request.Content = new StringContent(JsonConvert.SerializeObject(resourceLocations), System.Text.Encoding.UTF8, "application/json");
                    HttpResponseMessage response = await client.SendAsync(request);

                    if (!response.IsSuccessStatusCode)
                    {
                        await WriteFailedResponse(response);                        
                    }

                    // Do what you want with a successful response here.
                }

----------------------------------------------------------------------------------------------- >8 ----------

The response is that request.Content is empty: "The request failed with status code: NoContent"

My only guess is that the SAS blob token is incorrect (which is a part of the resourceLocations above), but should I not get an another error then?

要使用什么Blob令牌?以及如何找回它?

What blob token to use? And how to retrieve it?

我希望我足够清楚.

最好的问候，
马格努斯R

Best Regards,
Magnus R

无内容"；是正确的行为.有关Blob令牌的问题仍然存在.

"No Content" is correct behavior. Question is still regarding blob token.

//MR

推荐答案

Magnus

Hi Magnus,

以您的"

还可以提供您在响应中收到的HTTP响应吗?例如:401、403等...

Also, can you provide the HTTP response you received in the response? ie: 401, 403, etc...

谢谢.

詹姆斯

这篇关于重新训练和要使用的Blob令牌的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！