🚦
Status
Understand the 'Status' definition and meanings.
Note that 'status' feature exists in NER, Text classification, Sentiment Analysis projects.
In the Status column on Dashboard menu page, you can see one of the four status values below by hovering the mouse on the icon beside the Status title.
Dashboard > Status column
The value of the option in this state depends on the inference state of the model and whether it has been checked by the user.
Inference is not done yet. Once inferring is done, the status changes to 2, 3, or 4.
Inference is done and either it's not confirmed by an agent or the training is turned off for the phrase.
Inference is complete, and the agent-authenticated result is accurate. There are two possible scenarios in this state.
1) The status is accurate when the inference result and verification result are the same, and no changes have been made to the row. That is, if the verification result is changed to be the same as the inference result or if the verification result is already the same as the inference result, the status will change to accurate when the training/test toggle is turned on.
2) It is accurate when the uploaded data contains tag information and the inference result matches the tag(s).
Inference is complete, and it has been confirmed that the result is incorrect by the agent. There are two possible scenarios for the status.
1) It is inaccurate when the inference result and the verified result are different, and changes have been made to the row. When the verified result is changed to be different from the inference result or when the verified result is already different from the inference result, the status will change to inaccurate when the training/test toggle is turned on.
2) It is inaccurate when the uploaded data contains tag information, and the inference result is different from the tag(s) when the training toggle is turned on.