Grid Search (ML_GRID)

This function exhaustively searches for the optimal combination of hyperparameter values for a machine learning model.

Grid search is a hyperparameter tuning technique used to find the optimal combination of hyperparameter values for a machine learning model. It works by systematically searching through a predefined grid of hyperparameter values, evaluating the model's performance using cross-validation at each point in the grid. Grid search helps to identify the hyperparameter configuration that yields the best performance, enhancing the model's accuracy and generalization ability.

Sample Request

This request is performing a grid search for hyperparameter tuning on the "RandomForest" model. It searches for the best combination of hyperparameters "n_estimators", "random_state", and "criterion" by evaluating the model's performance with different values provided in the "param_grid."

{
    "project_id": 13,
    "parent_id": 3,
    "block_id": 4,
    "function_code": "ML_GRID",
    "args": {
        "model_name": "RandomForest",
        "param_grid": [
            {
                "n_estimators": [
                    5,
                    10,
                    60,
                    100
                ]
            },
            {
                "random_state": [
                    0,
                    42,
                    60
                ],
                "criterion": [
                    "gini",
                    "entropy"
                ]
            }
        ]
    }
}

Grid Search

POST https://autogon.ai/api/v1/engine/start

Request Body

Name	Type	Description
project_id*	int	ID of the current project
parent_id*	int	ID of the previous block
block_id*	int	ID of the current block
function_code*	String	Function code for the current block
model_name*	String	Name of the pre-trained model to be used for analysis
param_grid	object	set of hyperparameter values that the grid search will exhaustively explore to find the optimal combination of hyperparameters

200: OK StateManagement object

{
    "status": "true",
    "message": {
        "id": 1,
        "project": 12,
        "block_id": 10,
        "parent_id": 11,
        "dataset_url": "",
        "x_value_url": "",
        "y_value_url": "",
        "x_train_url": "",
        "y_train_url": "",
        "x_test_url": "",
        "y_test_url": "",
        "output": "{'confusion_matrix': '', 'accuracy': 0.9}"
    }
}

Python

// Some code

Node

const project_id = 1
const parent_id = 7
const block_id = 8
    
LogisticRegressionMetrics= await client.logistic_regression_metrics(project_id, parent_id, block_id, {
    model_name: "SimpleModel",

});