1. Pre-requisite
In order to monitor your externally deployed model on Picsellia you have to configure:
- A
Model
on Picsellia - A
ModelVersion
related - Add a
LabelMap
for yourModelVersion
- A
Deployment
in monitoring mode with yourModelVersion
attached.
1. Variables
Let's say you have a PyTorch Classification Model called Silicon-Valley-Classification with a version called Deployed predicting the classes HOT-DOG & NOT-HOT-DOG with the Labelmap
:
{
'0': 'HOT-DOG',
'1': 'NOT-HOT-DOG'
}
2. Configuration
A. Create your Picsellia Model
Model
![](https://files.readme.io/1d984c7-image.png)
![](https://files.readme.io/2110141-image.png)
B. Create your ModelVersion
ModelVersion
![](https://files.readme.io/0f3220e-image.png)
You need to have the exact same order in the
Labelmap
of your customModelVersion
and theLabelmap
configured on Picsellia
Now you should see a new version in your ModelVersion
overview:
![](https://files.readme.io/d69aca7-image.png)
It's time to deploy your ModelVersion
![](https://files.readme.io/8645727-image.png)
You should select the Monitoring only mode
![](https://files.readme.io/c8dc9de-image.png)
And now you should be good to go:
![](https://files.readme.io/6f2a14d-image.png)
Updated 8 months ago