I am not using WML via API, but directly deploying CP model and data to WML, at this point.
Here is what I have done:
Once I upgraded my subscription, I updated the plan info of resources. Then, I created a new project "Test", uploaded csv inputs and CP code, and ran the model. It still ran with "Lite" plan (worker=1). I wish I can share my screen with IBM WML UI expert. Again, I could not find the place where I can specify the capacity (vCPU and GRAM).
[2021-03-19T11:36:15Z, INFO] ! Workers = 1
------------------------------
Andy Ham
------------------------------
Original Message:
Sent: Fri March 19, 2021 03:47 AM
From: ALEX FLEISCHER
Subject: WML plan "Lite" to "Standard v1" with "Pay-As-You-Go"
Hi,
in https://www.ibm.com/cloud/blog/announcements/decision-optimization-now-available-in-watson-machine-learning-service
You can follow the link to a notebook https://dataplatform.cloud.ibm.com/exchange/public/entry/view/50fa9246181026cd7ae2a5bc7e4ac7bd
where you will see
client.deployments.ConfigurationMetaNames.HARDWARE_SPEC: {'name': 'S', 'nodes': 1}
The 'S' means small. And if you need a larger instance you will either choose M or L
regards
NB:
nodes will help you run many jobs in //
See https://medium.com/@AlainChabrier/do-for-wml-throughput-and-latency-5ca382ca320e
------------------------------
[Alex] [Fleischer]
[EMEA CPLEX Optimization Technical Sales]
[IBM]
Original Message:
Sent: Thu March 18, 2021 02:00 PM
From: Andy Ham
Subject: WML plan "Lite" to "Standard v1" with "Pay-As-You-Go"
I have upgraded my plan to "Standard v1" and ran a CP model. The model still ran with "1 Worker" under "Lite" plan. I searched all configurations on WML, but I could not figure out how to specify the plan and select the capacity. I want to try "16 vCPU + 64GB RAM". I tried the chat screen, but only sales personal is available. Can you please help me?
#DecisionOptimization