Sign in

v1

latestOpenAPI 3.0.12026-07-14453659.5 KB
Language Training

Train a Dataset

Trains a dataset and creates a model.

post/v2/language/train

Response

Training Status

algorithmstring

Algorithm used to create the model. Returned only when the modelType is image-detection.

createdAtstring date-time

Date and time that the model was created.

datasetIdinteger required

ID of the dataset trained to create the model.

datasetVersionIdinteger required

Not available yet

epochsinteger

Number of epochs used during training.

failureMsgstring

Reason the dataset training failed. Returned only if the training status is FAILED.

languagestring required

Model language inherited from the dataset language. For image datasets, default is N/A. For text datasets, default is en_US.

learningRatenumber double

Learning rate used during training.

modelIdstring required

ID of the model. Contains letters and numbers.

modelTypestring

Type of data from which the model was created.

namestring required

Name of the model.

objectstring

Object returned; in this case, training.

progressnumber required

How far the dataset training has progressed. Values are between 0�1.

queuePositioninteger

Where the training job is in the queue. This field appears in the response only if the status is QUEUED.

status'QUEUED' | 'RUNNING' | 'SUCCEEDED' | 'FAILED' | 'KILLED' | 'FAILED_WITH_RETRIES' required

Status of the model.

trainParamsstring

Training parameters passed into the request.

trainStatsstring

Returns null when you train a dataset. Training statistics are returned when the status is SUCCEEDED or FAILED.

updatedAtstring date-time

Date and time that the model was last updated.

Example response

{
  "algorithm": "object-detection",
  "datasetId": 57,
  "epochs": 20,
  "failureMsg": "To train a dataset and create a model, the dataset must contain at least 100 examples per label for test set",
  "language": "en_US",
  "learningRate": 0.0001,
  "modelId": "2KXJEOM3N562JBT4P7OX7VID2Q",
  "name": "My Model - Version1",
  "object": "training",
  "progress": 0.7,
  "queuePosition": 1,
  "trainParams": "{\"trainSplitRatio\":0.7}"
}