when I am publishing the model with designer and referring in the data-flow or power-bi it is working fine. when I am trying to get the rest endpoint from the automated ML. it is showing this screen and I am not clear what to be input in this. I tried providing columns and values. it is not working. it is throwing following error.
any idea what is wrong here.
Expression.Error: We cannot convert the value "[Function]" to type Function.DetailsValue = [Function]
Related
I tried to create a custom named entity recognition model in Azure. I uploaded data, labled it, trained the model and it works pretty good. The only problem I have is, that the data type is not as I hoped. Thanks for your help
E.g. when I extract a birthdate and the entity is "tomorrow", the data type is not a date but a String. My question is, how to create a date out of the "tomorrow"?
I am using semantic search in azs, api version is 21-04-30, analyers i have selected is engish microsoft while creating index.
When i am typing any question in query string with searchMode=all, i am getting nothing as a result, when i am typing searchMode=any, then it is giving me result but not accurate. Can anyone help me?
(https://i.stack.imgur.com/Lm2IC.png)
Please check this image
I am writing query in question form with all search mode, but not getting expected result. if i am removing query keywords such as what , it is giving me result but not while framing question. Please help.
I am trying to query data using the term aggregation in AWS OpenSearch to get the group by the result using 1 field. The query below working in their Dev tools successfully as expected. Attaching a screenshot below.
But when I try to do this same in the query in Node.js it's not giving a bucket array result. I search a lot but I found nothing different than my code. I have attached a screenshot and the result below. In the screenshot, I am able to see 16 as the total result but not a bucket result. Please help me to figure out what I am doing wrong here.
Referring to the batch testing document from Microsoft, I have imported the LUIS app provided from this document and import this this JSON batch file to LUIS batch testing panel and try to run the test.
The testing result of 'Intents' is as expected, however, for the 'Entities', it is expected some of the entities will pass but at the end all entities are failed:
Entities Fail Screenshot
Is it a bug in LUIS or anything going wrong?
Updates: The issue above is fixed but some batch testing function are still not working
After the fix have been released, I have tried it again. I found that it works fine with no entity, simple entity and hierarchical entity. However, for List Entity and Regex Entity still not working on Microsoft sample. Entity list and entity list in batch testing in Microsoft sample are not match. Besides, in my own Luis project, the list entity is showing in the batch testing entity list but the result is strange (caption in this screen cap).
By the way, if you look at the tutorial, it is stated that:
"Only the machine-learned entities are used: simple, hierarchical (parent-only), and composite entities. Do not add non-machine-learned entities because they are always found either through regular expressions, or explicit text matches"
And List Entities are not machine-learned entities.
When I try to run a report in COGNOS Report Studio, I get error : ORA-00918: column ambiguously defined
Now there is no way to get the runtime sql and test it out against the oracle db. So I am left groping around.
My question is...when we develop the model in framework manager, we do not write our own sql. Just specify the tables and columns and joins. So the error should never come because this error come when you forget to prefix a column name with the table alias.
I agree it is odd. Start removing data items from the query until it works. Try to narrow it down to a specific data item within a table. See what SQL is generated without the offending field, that should give you a hint as to what is going on.