Amazon rekognition custom labels LabelsDetection() returns empty value

0

I trained a model in rekognition and run in aws cli. But ı call the method return empty list like this: "CustomLabels": [] How can I solve ıt. Thank you.

cagatay
已提問 2 年前檢視次數 425 次
2 個答案
0

Hi,

To make a better debug, could you please let me know what is the script of the aws cli you called or provide the return response with empty value? Also, please let me know the arn of you model.

Thanks,

AWS
Jady
已回答 2 年前
  • Hi, aws rekognition detect-custom-labels
    --project-version-arn "arn:aws:rekognition:us-east-1:680335478196:project/dugme/version/dugme.2022-08-25T11.39.37/1661416777799"
    --image '{"S3Object": {"Bucket": "dugme-model","Name": "NXT_12_Aug_2022_14_11_00_853.jpg"}}'
    --region us-east-1 This is aws's advice script for use model. And this is my model's arn: arn:aws:rekognition:us-east-1:680335478196:project/dugme/version/dugme.2022-08-25T11.39.37/1661416777799

    Thank you for answer

0

As discussed further on this page, Rekognition applies confidence thresholding on returned labels - using assumed thresholds by default unless you specifically configure something different. Probably for this test image your model is not confidently detecting any labels, so all detections are being thresholded out.

You can try specifying MinConfidence in your DetectCustomLabels API call to 0 to return all labels regardless of detected confidence?

AWS
專家
Alex_T
已回答 2 年前

您尚未登入。 登入 去張貼答案。

一個好的回答可以清楚地回答問題並提供建設性的意見回饋,同時有助於提問者的專業成長。

回答問題指南