FAQ & Troubleshooting

I have questions about...
​Code-level integration with Kolena​
​Test case best practices​
​Test suite best practices​
​Model best practices​
...or something else:
Do I have to share my data with Kolena to use it?
No. Kolena doesn't store your image data, only URLs pointing to its location in a cloud bucket that you own.
While onboarding your team, we'll discuss what access restrictions are necessary for your data and select the right integration solution. As one example, as a part of the integration we might restrict access to images registered with Kolena to only users on your corporate VPN.
We support a variety of integration patterns depending on your organization's requirements and security stance. Get in touch with us to discuss details!
Do I have to upload my models to Kolena to test them?
No. Kolena never sees your models, only metadata describing them and any inferences they produce during testing. Testing is always performed on your infrastructure.
What types of data does Kolena support?
Kolena supports testing any data type by building a workflow.
In-platform data visualization is supported for most computer vision data types.
For built-in workflows, Kolena currently supports the following computer vision data types:
RGB and grayscale static imagery in PNG, JPEG, TIFF and other standard formats
Multispectral imagery (>3 channels)
Domain-specific image formats such as NITF and DICOM​
Exotic imagery types such as infrared, SAR, x-ray, and more
Don't see your data type here? Contact us!
Which cloud platforms does Kolena support?
Kolena integrates seamlessly with AWS (S3) GCP (GCS), and Azure (Blob Storage). Kolena also supports custom, internal, or local data storage solutions.
​Contact us for details.
Which machine learning frameworks does Kolena support?
Kolena is framework-agnostic. You can test models of any form on Kolena.
Last modified 8mo ago