Stack Overflow
Stack Overflow

Questions tagged [watson-openscale]

Any questions which are meant for the (earlier named) product IBM AI OpenScale, should use the tag watson-openscale. The product was rebranded to be called IBM Watson OpenScale.

IBM AI OpenScale was recently rebranded as the IBM Watson OpenScale in this recent official announcement. We currently have the ai-openscale tag that is used on most of the questions for the service. As with any rebranding, it takes a while for everyone to get accustomed to it. We would like watson-openscale to be the master tag and ai-openscale to be a synonym tag. As I understand it, if someone added the ai-openscale tag to a question, it would then be automatically published as watson-openscale.

21 questions
2
votes
1 answer

Do I have to provision a paid database in IBM cloud in order to use Watson OpenScale's features?

Watson OpenScale provides free internal database in the tutorial. However, if I want to monitor my own models, do I need to provision a paid database in IBM cloud to do the payload logging? If not, does the free internal database also support the…
Bufan Zeng
  • 33
  • 4
1
vote
2 answers

Payload logging on AWS SageMaker

How can I enable monitoring and log payload for a model deployed on AWS SageMaker? I am using a classification model and will be outputting predicted class and confidence. How should i configure this in UI or sdk?
Rohan Vaidyanathan
  • 43
  • 7
1
vote
1 answer

What model types are supported by OpenScale?

Does OpenScale support all types of model that can be deployed to Watson Machine Learning? Or just a subset of them? Among the supported types of model, will the explainability feature work for all of them?
Bufan Zeng
  • 33
  • 4
0
votes
1 answer

can't configure Watson OpenScale on IBM Cloud

I can’t create APIClient for watson openscale. I get message: AuthorizationError: You are not authorized to access AI OpenScale instance *** I am following the notebook below and stopped at ‘Configure OpenScale’…
ertogrul
  • 23
  • 1
  • 8
0
votes
0 answers

Evaluating a dataset on Openscale returns "column needs to be double" error

I'm trying to evaluate this dataset and keep getting an error saying that a column needs to be double, but when I check it it's a float, which should be the same thing in Python 3 (?) import pandas as pd df =…
guipleite
  • 58
  • 1
  • 8
0
votes
1 answer

Unable to configure a local WML provider in CP4D 2.5

I have OpenScale and WML installed on CP4D 2.5. When I try to configure OpenScale to use the local WML provider I receive the following error: Error There was an error while saving the configuration. Failed to connect to ML provider. Please check…
Miguel Troncoso
  • 21
  • 4
0
votes
1 answer

Cant correctly config quality and drift monitors

I have really strange problem So, I created model via Modeler Flow using Spark Environment There was a .csv file, that consists of data about bank's clients and this model had to predict is there any risk to give these people the loan. After that,…
TiHan
  • 35
  • 5
0
votes
1 answer

Unable to add transaction_id column in model for using on watson openscale

To view transactions on Watson OpenScale, a transaction id field is required. I cannot add this field while training the model and cannot add this field after the model has been trained. How can I add this column to view transactions on Watson…
Richa Srivastava
  • 1
  • 1
0
votes
1 answer

Question about interpretation of importance in contrastive explanation section of the explanation task response

I'm wondering how to interpret the "importance" values in the contrastive explanation part when getting an explanation task response. See https://cloud.ibm.com/apidocs/ai-openscale#get-explanation-task > entity > contrastive_explanation >…
71ML
  • 3
  • 2
0
votes
1 answer

Can not update payload table with error message "com.ibm.analytics.aios.utils.clients.db.JsonToSQLConversionError: db.json.convert"

I'm trying to use Openscale to check the explainibiltiy of my image classification model(Keras:2.2.4、tensorflow：1.11) So far, I have finished the configuration and able to see the explainability of my first scoring request. However, when I tried to…
Hsu HanPang
  • 16
  • 2
0
votes
1 answer

prediction explanation for specific customer

I am thinking thru implementation pattern around churn problem. Scenario I am thinking about is pretty simple - for customers with propensity to churn to identify features of the account that could be changed/different plans offered, etc to try to…
Alex Baryudin
  • 1
  • 1
0
votes
1 answer

Can Watson OpenScale monitor models that are serving using AWS Lambda Function?

H2O Driverless AI serves its models using AWS Lambda Function. I'm wondering if there's a possible integration between H2O and Watson OpenScale.
Bufan Zeng
  • 33
  • 4
0
votes
1 answer

Fairness Monitor Doesn't Allow Configuration of "Maximum Records"

When configuring accuracy monitor, one can specify min records and max records for metric computation; however, when configuring fairness monitor, there is only min records, and effectively it seems to be the fixed number of rows for fairness…
Wendy
  • 89
  • 5
0
votes
1 answer

How do I implement payload analytics in Watson OpenScale?

How do I get access to the payload data in Watson OpenScale? What languages are supported for querying the data (e.g. SQL)? Will OpenScale provide a UI for payload data visualization in the future?
Bufan Zeng
  • 33
  • 4
0
votes
1 answer

Where does the de-biased model get trained in Watson OpenScale?

Does the model get trained in Watson Machine Learning, some other 3rd party engines or micro services in OpenScale? After the model is trained, where does it get served? Can I inspect/modify the model?
Bufan Zeng
  • 33
  • 4
1
2