You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Describe the bug
I have recently started using Amazon Rekoginition in paralel with Compreface, and haven't been able to train the model when Double-Take returns unknown face recognition.
As an example, when I select an image an apply a label, I jump to the Train tab of Double-Take, and wait until the page loads (while training is in process) and get an error message saying Error 500. Slightly after that (couple of minutes) I get multiple Error 502 messages.
This is my Double-Take configuration for Rekognition:
rekognition:
aws_access_key_id: !secret aws_access_key
aws_secret_access_key: !secret aws_secret_access_key
aws_region: eu-west-1 #this is in Ireland, default is us-east-1
collection_id: double-take
opencv_face_required: true
Version of Double Take v1.13.11.8
Expected behavior
Would expect that by applying the label to an image, that the model would be trained and would start getting subsequent images recognized by the model
Screenshots
Select image and apply label
Check training status
Request fails
Images appear as trained, but new (or tarined) images don't show the label applied
Hardware
OS: running double-take in Home Assistant (HACS)
Browser: Chrome
Additional context
Cannot find any useful info in the logs (including a reference to those error messages)
The text was updated successfully, but these errors were encountered:
Describe the bug
I have recently started using Amazon Rekoginition in paralel with Compreface, and haven't been able to train the model when Double-Take returns unknown face recognition.
As an example, when I select an image an apply a label, I jump to the Train tab of Double-Take, and wait until the page loads (while training is in process) and get an error message saying Error 500. Slightly after that (couple of minutes) I get multiple Error 502 messages.
This is my Double-Take configuration for Rekognition:
Version of Double Take
v1.13.11.8
Expected behavior
Would expect that by applying the label to an image, that the model would be trained and would start getting subsequent images recognized by the model
Screenshots
Hardware
Additional context
Cannot find any useful info in the logs (including a reference to those error messages)
The text was updated successfully, but these errors were encountered: