This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(284.8, 78%, 37%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EMW%3C/text%3E%3C/svg%3E)
Hi all,
I created a classifier by trained SVM and PCA as well as normalization transformation. And use Request/Response Excel to test.
And if I only input for one row, all the PCA would be "0"
but if I put more than one row together then the result would change and both behavior doesn't make sense.
I read some other questions below and I wonder if something wrong in "Normalization" calculating that it didn't count by referring "Applying transformation" but simply normalizing by data input.
https://social.msdn.microsoft.com/Forums/azure/en-US/bb800d7e-2fbf-4874-a525-aae9ace5df4f/azure-ml-webservice-same-sample-file-but-different-predictions?forum=MachineLearning
https://social.msdn.microsoft.com/Forums/en-US/21938a5a-ae9c-4559-82f0-95889b359206/normalization-and-predictive-web-service?forum=MachineLearning
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(22.400000000000002, 11%, 12%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ERM%3C/text%3E%3C/svg%3E)
@Michael WANG Thanks for the question.Can you please publish your experiment to the azure AI gallery and please share the link to the same to check. We have forwarded to the product team to check on this.
https://gallery.azure.ai/
@Ramr-msft I have published to gallery at https://gallery.azure.ai/Experiment/First-elec-screw-driver-on-2020-8-25-Predictive-Exp Please check if you can access it.
@Michael WANG Thanks, Can you please add more details about the what’s the transformation method that you have applied.
Please follow the detailed introduction to normalize data in classic studio: https://video2.skills-academy.com/en-us/azure/machine-learning/studio-module-reference/normalize-data?redirectedfrom=MSDN
@Ramr-msft I read it before and wonder it helps for my experienced issue here. So I upload the training model here
https://gallery.cortanaintelligence.com/Experiment/First-elec-screw-driver-on-2020-8-25
And data can't be uploaded so please check if it works.
@Michael WANG ·Thanks for the details. We are checking on cause for the difference in results, will update on the same.
@Michael WANG Thanks. We would recommend to raise a Azure support desk ticket from Help+Support blade from Azure portal if you have a support plan for your subscription. This will help you to share the details securely and work with an engineer who can provide more insights about the issue.
If you do not have a valid support plan we can enable a onetime support request for your subscription if you can email us at AzCommunity[at]microsoft[dot]com with the URL of this thread and your subscription id.
@Ramr-msft I sent the email with title "[Microsoft Q&A] ramr-msft mentioned you in a comment" with the relevant information required so please help to check, thank you.