This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(115.19999999999999, 63%, 21%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EA%3C/text%3E%3C/svg%3E)
I want to know if it shows same results as Predict API would show? basically what are the Quick Test predictions based on?
Is there any official detailed information about how Quick Test function works? my Quick Test results are very different from the offline TensorFlow models that I exported, that's why I'm asking.
Thank you
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(278.4, 1%, 36%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EYM%3C/text%3E%3C/svg%3E)
Hello @Ashi
Thanks for reaching out to us here, in most of the situations, they are different since the quick test is only used when you want to check the latest iteration's performance.
The Quick Test uses the most recently trained iteration of your model only, like the below screenshot, the iteration 4 will be used:
https://learn.microsoft.com/en-us/azure/cognitive-services/custom-vision-service/test-your-model#test-your-model
The Predict API uses the iteration you selected the best performance one to publish:
https://learn.microsoft.com/en-us/azure/cognitive-services/custom-vision-service/use-prediction-api#publish-your-trained-iteration
Besides that, I think the guidance of how to improve the model you may need: https://learn.microsoft.com/en-us/azure/cognitive-services/custom-vision-service/getting-started-improving-your-classifier
I hope this helps, please let me know if you need futher help.
Regards,
Yutong
-Please kindly accept the answer if you feel helpful to help the community, thanks!
Hello @Ashi
Thanks for your follow up, just double confirmed with the product manager, the quick answer is YES.
Please let me know if you need further assistance, hope this helps.
Regards,
Yutong
Thank you for your reply, but I'm afraid it doesn't answer my question. When using Quick Test there's actually a drop down menu for selecting the iterations, so it doesn't have to be the last one when you are testing.
My question specifically is if Quick test with iteration X would show same results as using prediction API with a published iteration X.
Thank you