Trainedmodels

Trained models are models that you train and are owned by a project.

For a list of methods for this resource, see the end of this page.

Resource representations

{
  "kind": "prediction#training",
  "id": string,
  "selfLink": string,
  "created": datetime,
  "trainingComplete": datetime,
  "modelInfo": {
    "numberInstances": long,
    "modelType": string,
    "numberLabels": long,
    "classificationAccuracy": string,
    "classWeightedAccuracy": string,
    "meanSquaredError": string
  },
  "trainingStatus": string
}
Property name Value Description Notes
kind string What kind of resource this is - "prediction#training".
id string The unique name for the predictive model.
created datetime Insert time of the model (as a RFC 3339 timestamp).
trainingComplete datetime Training completion time (as a RFC 3339 timestamp).
modelInfo object Model metadata.
modelInfo.numberInstances long Number of valid data instances used in the trained model.
modelInfo.modelType string Type of predictive model (CLASSIFICATION or REGRESSION).
modelInfo.numberLabels long Number of class labels in the trained model (CLASSIFICATION models only).
modelInfo.classificationAccuracy string A number between 0.0 and 1.0, where 1.0 is 100% accurate. This is an estimate, based on the amount and quality of the training data, of the estimated prediction accuracy. You can use this is a guide to decide whether the results are accurate enough for your needs. This estimate will be more reliable if your real input data is similar to your training data (CLASSIFICATION models only).
modelInfo.classWeightedAccuracy string Estimated accuracy of model taking utility weights into account (CLASSIFICATION models only).
modelInfo.meanSquaredError string An estimated mean squared error. The can be used to measure the quality of the predicted model (REGRESSION models only).
trainingStatus string The current status of the training job. This can be one of following:
  • RUNNING - Only returned when retraining a model; for a new model, a trainedmodels.get call will return HTTP 200 before training is complete.
  • DONE
  • ERROR
  • ERROR: NO VALID DATA INSTANCES
  • ERROR: TRAINING JOB NOT FOUND
  • ERROR: TRAINING TIME LIMIT EXCEEDED
  • ERROR: TRAINING SYSTEM CAPACITY EXCEEDED
  • ERROR: TRAINING DATA FILE SIZE LIMIT ERROR
  • ERROR: STORAGE LOCATION IS INVALID

Methods

analyze
Get analysis of the model and the data the model was trained on.
delete
Delete a trained model.
get
Check training status of your model.
insert
Train a Prediction API model.
list
List available models.
predict
Submit model id and request a prediction.
update
Add new data to a trained model.

Send feedback about...

Prediction API Documentation