DESlib: A Dynamic ensemble selection library in Python
Authors: Rafael M. O. Cruz, Luiz G. Hafemann, Robert Sabourin, George D. C. Cavalcanti
JMLR 2020 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The library also provides static ensemble methods which are often used as baseline comparisons for dynamic ensembles. Tables 1 and 2 list the implemented DS and baseline methods, respectively. |
| Researcher Affiliation | Academia | Laboratoire d imagerie, vision et intelligence artificielle (LIVIA) École de Technologie Supérieure (ÉTS) Université du Québec Montreal, Canada Centro de Informática Universidade Federal de Pernambuco Recife, Brazil |
| Pseudocode | Yes | In the example below, we show how to use the library, with a given Training (X_train, y_train), and Testing (X_test, y_test) datasets. The META-DES (Cruz et al., 2015a) technique is used in this example: from sklearn.ensemble import RandomForestClassifier from deslib.des.meta_des import METADES # Train a pool of 10 classifiers pool_classifiers = RandomForestClassifier(n_estimators=10) pool_classifiers.fit(X_train, y_train) # Initialize the DS model metades = METADES(pool_classifiers) # Fit the dynamic selection model metades.fit(X_dsel, y_dsel) # Predict new examples: metades.predict(X_test) |
| Open Source Code | Yes | DESlib is an open-source python library providing the implementation of several dynamic selection techniques. ... Documentation, code and examples can be found on its Git Hub page: https://github.com/scikit-learn-contrib/DESlib. |
| Open Datasets | No | The paper provides an example of how to use the library with placeholder datasets (X_train, y_train, etc.) but does not specify any concrete, publicly available datasets used for evaluation or experimentation in the paper itself, nor does it provide links or citations to such datasets. |
| Dataset Splits | No | The paper provides a code example that uses 'Training (X_train, y_train), and Testing (X_test, y_test) datasets' as placeholders, but it does not specify any actual dataset split percentages, counts, or methodologies for these or any other datasets. |
| Hardware Specification | No | The paper does not provide any specific hardware details such as GPU models, CPU specifications, or memory amounts used for running experiments. It only describes the software library and its functionality. |
| Software Dependencies | No | The library is written in pure python, working on any platform, and depends on the following python packages: scikit-learn, numpy and scipy. The paper lists software dependencies but does not specify their version numbers. |
| Experiment Setup | Yes | In the example below, we show how to use the library, with a given Training (X_train, y_train), and Testing (X_test, y_test) datasets. The META-DES (Cruz et al., 2015a) technique is used in this example: from sklearn.ensemble import RandomForestClassifier from deslib.des.meta_des import METADES # Train a pool of 10 classifiers pool_classifiers = RandomForestClassifier(n_estimators=10) |