Data management

---

Data sets

ATOM is designed to work around one single dataset: the one with which atom is initialized. This is the dataset you want to explore, transform, and use for model training and validation. ATOM differentiates three different data sets:

• The training set is usually the largest of the data sets. As the name suggests, this set is used to train the pipeline. During hyperparameter tuning, only the training set is used to fit and evaluate the estimator in every call. The training set in the current branch can be accessed through the train attribute. It's features and target can be accessed through X_train and y_train respectively.
• The test set is used to evaluate the models in the pipeline. The model scores on this set give an indication on how the model performs on new data. The test set can be accessed through the test attribute. It's features and target can be accessed through X_test and y_test respectively.
• The holdout set is an optional, separate set that should only be used to evaluate the final model's performance. Create this set when you are going to use the test set for an intermediate validation step. The holdout set is immediately set apart during initialization and is not considered part of atom's dataset (the dataset attribute only returns the train and test sets). The holdout set is left untouched until predictions are made on it, i.e. it does not undergo any pipeline transformations. The holdout set is stored in the trainer's holdout attribute. It's features and target can not be accessed separately. See here an example that shows how to use the holdout data set.

The data can be provided in different formats. If the data sets are not specified beforehand, you can input the features and target separately or together:

• X
• X, y

Remember to use the y parameter to indicate the target column in X when using the first option. If not specified, the last column in X is used as target. In both these cases, the size of the sets are defined using the test_size and holdout_size parameters. Note that the splits are made after the subsample of the dataset with the n_rows parameter (when not left to its default value).

If you already have the separate data sets, provide them using one of the following formats:

• train, test
• train, test, holdout
• X_train, X_test, y_train, y_test
• X_train, X_test, X_holdout, y_train, y_test, y_holdout
• (X_train, y_train), (X_test, y_test)
• (X_train, y_train), (X_test, y_test), (X_holdout, y_holdout)

The input data is always converted internally to a pandas dataframe, if it isn't one already. The column names should always be strings. If they are not, atom changes their type at initialization. If no column names are provided, default names are given of the form feature n, where n stands for the n-th feature.

Indexing

By default, atom resets the dataframe's index after initialization and after every transformation in the pipeline. To avoid this, specify the index parameter. If the dataset has an 'identifier' column, it is useful to use it as index for two reasons:

• An identifier doesn't usually contain any useful information on the target column, and should therefore be removed before training.
• Predictions of specific rows can be accessed through their index.

Sparse matrices

If atom is initialized using a scipy sparse matrix, it is converted internally to a dataframe of sparse columns. Read more about pandas' sparse data structures here. The same conversion takes place when a transformer returns a sparse matrix, like, for example, the Vectorizer.

Note that ATOM considers a dataset to be sparse if any of the columns is sparse. A dataset can only benefit from sparsity when all its columns are sparse, hence mixing sparse and non-sparse columns is not recommended and can cause estimators to decrease their training speed or even crash. Use the shrink method to convert dense features to sparse and the available_models method to check which models have native support for sparse matrices.

Click here to see an example that uses sparse data.

Branches

You might want to compare how a model performs on a dataset transformed through multiple pipelines, each using different transformers. For example, on one pipeline with an undersampling strategy and the other with an oversampling strategy. To be able to do this, ATOM has the branching system.

The branching system helps the user to manage multiple pipelines within the same atom instance. Every pipeline is stored in a branch, which can be accessed through the branch property. A branch contains a copy of the dataset, and all transformers and models that are fitted on that specific dataset. Transformers and models called from atom use the dataset in the current branch, as well as data attributes such as atom.dataset. Use the branch's __repr__ to get an overview of the transformers in the branch. It's not allowed to change the data in a branch after fitting a model with it. Doing this would cause unexpected model behaviour and break down the plotting methods. Instead, create a new branch for every unique pipeline.

By default, atom starts with one branch called "master". To start a new branch, set a new name to the property, e.g. atom.branch = "undersample". This will create a new branch from the current one. To create a branch from any other branch type "_from_" between the new name and the branch from which to split, e.g. atom.branch = "oversample_from_master" will create branch "oversample" from branch "master", even if the current branch is "undersample". To switch between existing branches, just type the name of the desired branch, e.g. atom.branch = "master" brings you back to the master branch. Note that every branch contains a unique copy of the whole dataset! Creating many branches can cause memory issues for large datasets.

See the Imbalanced datasets or Feature engineering examples for branching use cases.

Warning

Always create a new branch if you want to change the dataset after fitting a model!

Figure 1. Diagram of a possible branch system to compare an oversampling with an undersampling pipeline.

The branch class has the following methods.

delete	Delete the branch from the atom instance.
rename	Change the name of the branch.
status	Get an overview of the pipeline and models in the branch.

method delete() [source]

Delete the branch and all the models in it. Same as executing del atom.branch.

method rename(name) [source]

Change the name of the branch.

Parameters:

name: str New name for the branch. Can not be empty nor equal to an existing branch.

method status() [source]

Get an overview of the pipeline and models in the branch. This method prints the same information as the __repr__ and also saves it to the logger.

Memory considerations

An atom instance stores one copy of the dataframe in each branch, and one copy of the initial dataset with which the instance is initialized (this copy is necessary to avoid data leakage during hyperparameter tuning and for some specific methods like cross_validate and reset). This initial copy is created as soon as there are no branches in the initial state (usually after calling the first data transformation) and it's stored in an internal branch called og (original). The og branch is not accessible by the user.

Apart from the dataset itself, the model's predictions (e.g. predict_proba_train), metric scores and shap values are also stored as attributes of the model to avoid having to recalculate them every time they are needed. This data can occupy a considerable amount of memory for large datasets. You can delete all these attributes using the clear method in order to free some memory before saving the class.

Data transformations

Performing data transformations is a common requirement of many datasets before they are ready to be ingested by a model. ATOM provides various classes to apply data cleaning and feature engineering transformations to the data. This tooling should be able to help you apply most of the typically needed transformations to get the data ready for modelling. For further fine-tuning, it's also possible to transform the data using custom transformers (see the add method) or through a function (see the apply method). Remember that all transformations are only applied to the dataset in the current branch.

AutoML

Automated machine learning (AutoML) automates the selection, composition and parameterization of machine learning pipelines. Automating the machine learning process makes it more user-friendly and often provides faster, more accurate outputs than hand-coded algorithms. ATOM uses the TPOT package for AutoML optimization. TPOT uses a genetic algorithm to intelligently explore thousands of possible pipelines in order to find the best one for your data. Such an algorithm can be started through the automl method. The resulting data transformers and final estimator are merged with atom's pipeline (check the pipeline and models attributes after the method finishes running).

Warning

AutoML algorithms aren't intended to run for only a few minutes. If left to its default parameters, the method can take a very long time to finish!