💡Functions

All functions in PyCaret

Initialize

setup

This function initializes the experiment in PyCaret and prepares the transformation pipeline based on all the parameters passed in the function. The setup function must be called before executing any other function. It only requires two parameters: data and target. All the other parameters are optional.

Train

compare_models

This function trains and evaluates the performance of all the models available in the model library using cross-validation. The output of this function is a scoring grid with average cross-validated scores.

create_model

This function trains and evaluates the performance of a given model using cross-validation. The output of this function is a scoring grid with cross-validated scores along with mean and standard deviation.

Optimize

tune_model

This function tunes the hyperparameters of a given model. The output of this function is a scoring grid with cross-validated scores of the best model. Search spaces are pre-defined with the flexibility to provide your own. The search algorithm can be random, bayesian, and a few others with the ability to scale on large clusters.

ensemble_model

This function ensembles a given model. The output of this function is a scoring grid with cross-validated scores of the ensembled model. Two methods Bagging or Boosting can be used for ensembling.

blend_models

This function trains a Soft Voting / Majority Rule classifier for given models in a list. The output of this function is a scoring grid with cross-validated scores of a Voting Classifier or Regressor.

stack_models

This function trains a meta-model over given models in a list. The output of this function is a scoring grid with cross-validated scores of a Stacking Classifier or Regressor.

optimize_threshold

This function optimizes the probability threshold for a given model. It iterates over performance metrics at different probability thresholds and returns a plot with performance metrics on the y-axis and threshold on the x-axis.

calibrate_model

This function calibrates the probability of a given model using isotonic or logistic regression. The output of this function is a scoring grid with cross-validated scores of calibrated classifier.

Analyze

plot_model

This function analyzes the performance of a trained model on the hold-out set. It may require re-training the model in certain cases.

evaluate_model

This function uses ipywidgets to display a basic user interface for analyzing the performance of a trained model.

interpret_model

This function analyzes the predictions generated from a trained model. Most plots in this function are implemented based on the SHAP (Shapley Additive exPlanations).

dashboard

This function generates the interactive dashboard for a trained model. The dashboard is implemented using the ExplainerDashboard project.

check_fairness

This function provides fairness-related metrics between different groups in the dataset for a given model. There are many approaches to evaluate fairness but this function uses the approach known as group fairness, which asks: which groups of individuals are at risk for experiencing harm.

get_leaderboard

This function returns the leaderboard of all models trained in the current setup.

assign_model

This function assigns labels to the training dataset using the trained model. It is only available for unsupervised modules.

Deploy

predict_model

This function generates the label using a trained model. When unseen data is not passed, it predicts the label and score on the holdout set.

finalize_model

This function refits a given model on the entire dataset.

save_model

This function saves the ML pipeline as a pickle file for later use.

load_model

This function loads a previously saved pipeline.

save_experiment

This function saves an experiment to a pickle file.

load_experiment

This function loads an experiment back into Python from a pickle file.

check_drift

This function generates a drift report file using the evidently library.

deploy_model

This function deploys the entire ML pipeline on the cloud.

convert_model

This function transpiles the trained machine learning model's decision function in different programming languages such as Python, C, Java, Go, C#, etc.

create_api

This function takes an input model and creates a POST API for inference. It only creates the API and doesn't run it automatically.

create_docker

This function creates a Dockerfile and requirements.txt for deploying API.

create_app

This function creates a basic gradio app for inference.

Others

pull

Returns the last printed scoring grid.

models

Return a table containing all the models available in the imported module of the model library.

get_config

This function retrieves the global variables created by the setup function.

set_config

This function resets the global variables.

get_metrics

Returns the table of all available metrics used for cross-validation.

add_metric

Adds a custom metric to the metric container for cross-validation.

remove_metric

Removes a custom metric from the metric container.

automl

This function returns the best model from all the models in the current setup.

get_logs

Returns a table of experiment logs. Only works when log_experiment = True when initializing the setup function.

get_current_experiment

Obtain the current experiment object.

set_current_experiment

Set the current experiment to be used with the functional API.