predict_lgbm

This document has been machine translated.

Apply a trained LGBM model to perform prediction.

The execution method uses JSON-RPC v2.0.

Example request

To run predictions with the LGBM model, specify predict_lgbm as the method in the parameter prov.process of the Provenance API.

An example JSON-RPC request is shown below.

{
  "jsonrpc": "2.0",
  "method": "prov.process",
  "params": {
    "method": "predict_lgbm",
    "params": {
      "output_ddc": "ddc:result",
      "model_ddc": "ddc:model",
      "input_ddc": "ddc:test_data",
      "param_json": "{ (パラメータの項目を参照) }",
      "no_exec": true
    },
  },
  "id": "provenance_jsonrpc_id"
}

Parameter

The following parameters can be specified for predict_lgbm.

parameter name	content
output_ddc	output ddc name
output_mode	output mode (overwrite, append, or error)
model_ddc	model ddc name
input_ddc	input ddc name
param_json	predicted parameters

The param_json should be set to the same content as that specified during model training. However, you may omit the contents of lgbm.stressor.

Translated with www.DeepL.com/Translator (free version)

Input data

The ddc specified for model_ddc must have the following schema Normally, this should be the ddc output by the LGBM model train (train_lgbm).

column name	contents	remarks
method	learning method	fixed to lgbm string
train_data	input data name	the real table name will be listed instead of ddc
param_json	parameter	parameter string when learning
fixed to Max string	location_code	id representing the measurement location
target	value of target.column	value of `target.column
prediction_time	the value of target.prediction_time	the value of target.prediction_time
model_path`	model file path	the model file entity is stored on the compute node

The ddc specified in input_ddc must have the following schema

column name	contents	notes
start_datetime	start time	required column in event table
end_datetime	required column of event table
(measurement point id)	id representing the measurement point
(Attribute 1)	any attribute value (numeric type)
(Attribute 2)	any attribute value (numeric type)
(...)	Any attribute value (numeric type)

Translated with www.DeepL.com/Translator (free version)

Output data

The schema of ddc output to output_ddc is as follows

column name	contents	notes
start_datetime	start_time
end_datetime	end_datetime
location	location information
base_datetime	base time	base time + predicted time is the start time
location_id	id representing the point of measurement
value	Predicted result

Return value

Output ddc information