Man pages for jeroenvdhoven/datapiper
datapiper
| build_docker | Build a docker image out of a model function |
| build_model_package | Builds a package from your pipeline |
| create_stats | Calculates stats based on custom functions on the response... |
| create_stats_predict | Uses previous statistics results to generate columns for a... |
| feature_categorical_filter_predict | Filters categorical variables for new datasets |
| feature_finder | Add features one-by-one to find good small sets of features |
| feature_interactions_predict | Computes interaction effects for a new dataset |
| feature_one_hot_encode_predict | Apply one-hot encoding |
| feature_transformer_post_predict | Uses the results of 'pipe_feature_transformer' on new... |
| feature_transformer_predict | Uses the results of 'pipe_feature_transformer' on new... |
| find_best_models | Select top models from the find_model function |
| find_expand_results | Expands the params column of the result of 'find_model' |
| find_model | Find fitting models and test them using given metrics on the... |
| find_model_through_bayes | Hyper parameter search using bayesian optimisation |
| find_template_formula_and_data | A convenient wrapper function for find_model for models that... |
| find_xgb | A convenient wrapper function for find_model for xgboost |
| flatten_pipeline | Flattens a pipeline so it does not contain any more... |
| greedy_max_independent_set | Finds a maximum independent set using greedy search |
| high_correlation_features | Determines which columns are too highly correlated. |
| impute_model | Generate a model to impute missing data in a column |
| impute_predict_all | Use the models from 'impute_all' to impute the selected... |
| invoke | Generic function to apply either a pipe or pipeline to new... |
| invoke.pipe | Applies a pipe to new data |
| invoke.pipeline | Applies a pipeline to new data |
| is.pipe | Tests if an object inherits from pipe |
| is.pipeline | Tests if an object inherits from pipeline |
| model_trainer | Wrapper function for model inputs to 'find_model' |
| NA_indicators_predict | Indicate which fields are NA |
| pipe | Creates a pipe object out of a function and a list of... |
| pipe_categorical_filter | Remove values from categorical variables that do not occur... |
| pipe_check | Create a pipeline step that learns what the data looks like |
| pipe_clustering | Add cluster labels to a training set |
| pipe_create_stats | Generic function for creating statistics on the response... |
| pipe_dplyr | Wrapper function to turn a dplyr function into a pipeline... |
| pipe_feature_interactions | Generates permutation interaction effects between sets of... |
| pipe_feature_transformer | Applies different transformations to each numeric feature and... |
| pipe_function | Wrapper for putting a single function into a pipeline |
| pipe_impute | Impute multiple missing columns using lm, mean, or xgboost,... |
| pipeline | Creates a pipeline out of a set of pipes |
| pipe_mutate | Applies mutate in a pipeline |
| pipe_NA_indicators | Indicate which fields are NA |
| pipe_one_hot_encode | Train one-hot encoding |
| pipe_pca | Apply PCA to a subset of the columns in a dataset |
| pipe_range_classifier | Generates features for regression problems through... |
| pipe_remove_high_correlation_features | Removes highly correlated features whilst keeping as many as... |
| pipe_remove_single_value_columns | Remove all columns that have only a single value |
| pipe_scaler | Rescales data to standardised ranges |
| pipe_select | Applies select in a pipeline |
| plot_high_correlations | Plots highly correlated features as a graph |
| preserve_columns_predict | Only keep previously selected columns |
| range_predict | Use generated models and scales to create features for a new... |
| segment | A simple wrapper for creating a pipe segment |
| standard_column_names | Standardises column names into an oft-acceptable format |
| test_docker | Test your docker image |
| train_pipeline | Create a train/test pipeline from individual functions |
