Train a new model.
Path Parameters
- project_id integer required
ID of the project
- application/json
Request Body required
- display_name string
Display name of the model
- language string
Possible values: [
en
,de
,ar
]Language of the model
- name string
internal name. It should be unique for a workspace
- base_id integer
ID of the Base Model used for training
- annotation_ids integer[]
Possible values:
>= 1
IDs of the labeling sessions to be used to create test and train sets. If this parameter is given, trainset_annotation_ids and testset_annotation_ids will be ignored
- trainset_annotation_ids integer[]
Possible values:
>= 1
IDs of the labeling sessions to be used as source for the trainset
- testset_annotation_ids integer[]
Possible values:
>= 1
IDs of the labeling sessions to be used as source for the testset
- 201
- 400
- 401
- 403
- 500
Model training was started
- application/json
- Schema
- Example (from schema)
Schema
- base string
the name of the model used to create the modal that's being changed
- created date-time
creation date
- display_name string
a friendly name.
- uuid uuid
model unique identifier. Should be a UUID value
- id integer
Possible values:
>= 1
The ID of the model
- language string
Possible values: [
en
,de
,ar
] - long_description string
a longer description which proper describes the model
- name string
internal name. It should be unique for a workspace
- owner string
- progress float
- project_id integer
The ID of the project
- score float
model score. As this is a result of a training session, this value shouldn't be touched
- short_description string
a short description to help users quickly identify the model
- status string
Possible values: [
CREATED
,QUEUED
,TRAINING
,FAILED
,READY
,ACTIVE
,PUBLISHED
]the model status
- type string
Possible values: [
ABSA
,CLASS
,LABEL
,CLASSLABEL
]model type (1: BASE, 2: ABSA, 3: LABEL, 4: CLASS, 5: CLASSLABEL))
{
"base": "Hotel Pre-Trained",
"created": "2023-11-02",
"display_name": "My model 1",
"uuid": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
"id": 0,
"language": "en",
"long_description": "My model 1 can predict anything I want",
"name": "My model 1",
"owner": "John Doe",
"progress": 6.78,
"project_id": 1,
"score": 0.68,
"short_description": "This is a very short description of My Model 1",
"status": "CREATED",
"type": "ABSA"
}
The body sent is invalid. Check the error message for more details
User is not authenticated. This means token is either not present or invalid
User is read-only or does not have the minimal project role "MANAGER"
Something is wrong with the API.