4.3.2 NER Training Service

This topic describes the payload details for NER model training service.

Input Payload

Table 4-9 Input Payload

name	in	Type	Required	Remarks
trainCorpusPath	formData	string	true	Training Corpus path.
modelType	formData	File	true	Type of model being trained.
runRef	formData	string	true	Unique running reference number.

• name: " trainCorpusPath"

  in: "formData"

  type: string

  required: true

• name: " modelType"

  in: "formData"

  type: string

  value for NER model training: "nlpNer"

  required: true

• name: " runRef "

  in: "formData"

  type: string

  required: true

Output Payload

  {
   "data": {
    "timeTaken": 0,
    "corpusSize": 0,
    "precision": 0,
    "recall": 0,
    "f1score": 0,
    "model_fold_performances": [
    {
     "fold_no": 0,
     "eval_metric": "string",
     "value": 0
    }
   ],
   "tag_perfomances": [
    {
     "name": "string",
     "precision": 0,
     "recall": 0,
     "f1score": 0
    }
   ]
  }
 }

Output payload data model definition

ModelTrainParamsDTO:

type: object

properties:

data:

type: object

properties:

timeTaken:

type: number

corpusSize:

type: number

precision:

type: number

recall:

type: number

f1score:

type: number

model_fold_performances:

type: array

items:

$ref: "#/definitions/ModelFoldPerfromancesDTO"

tag_perfomances:

type: array

items:

$ref: "#/definitions/MltbTagPerfomancesDTO"

ModelTrainParamsDTO:

type: object

properties:

fold_no:

type: number

eval_metric:

type: string

value:

type: number

ModelTrainParamsDTO:

type: object

properties:

name:

type: string

precision:

type: number

recall:

type: number

f1score:

type: number