Scikit-learn: Machine Learning in Python
By Fabian Pedregosa, G. Varoquaux, Gaël Varoquaux, Alexandre Gramfort, V. Michel, Bertrand Thirion, Olivier Grisel, Mathieu Blondel, Gilles Louppe, Peter Prettenhofer and 8 more. 2011
91,720 citations recorded by the source corpus.
Scikit-learn is a Python module integrating a wide range of state-of-the-art machine learning algorithms for medium-scale supervised and unsupervised problems. This package focuses on bringing machine learning to non-specialists using a general-purpose high-level language. Emphasis is put on ease of use, performance, documentation, and API consistency. It has minimal dependencies and is distributed under the simplified BSD license, encouraging its use in both academic and commercial settings. Source code, binaries, and documentation can be downloaded from http://scikit-learn.sourceforge.net.
Subfields: Artificial Intelligence, Evaluation & Benchmarks
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Yoshua Bengio
- Mohit Bansal
- Percy Liang
- Quoc V. Le
- Jianfeng Gao
- Luke S. Zettlemoyer
- Sergey Levine
- Kyunghyun Cho
In Evaluation & Benchmarks
- Mohit Bansal
- Frank Hutter
- Yoshua Bengio
- Yejin Choi
- Hannaneh Hajishirzi
- Luke S. Zettlemoyer
- Kyunghyun Cho
- Matt Gardner
Citations in the corpus
- Pylearn2: a machine learning research library (2013, methodology): “This is different from other notable machine learning libraries, such as scikit-learn [39] or the learning algorithms provided as part of OpenCV [7], the STAIR Vision Library [23], etc.”
- Inductive Representation Learning on Large Graphs (2017, neutral): “For the feature only model and to make predictions on the embeddings output from the unsupervised models, we used the logistic SGDClassifier from the scikit-learn Python package [26], with all default settings.”
- You Autocomplete Me: Poisoning Vulnerabilities in Neural Code Completion (2020, methodology): “We train a decision tree classifier using Scikit-Learn [43]’s DecisionTree class, which implements a variant of the CART decision tree learning algorithm.”
- A Sober Look at the Unsupervised Learning of Disentangled Representations and their Evaluation (2020, methodology): “Finally, we train a gradient boosting classifier from Scikit-learn with default parameters.”
- Array programming with NumPy (2020, background): “2) builds on top of this foundation to provide several, widely used technique specific libraries [16, 17, 23], that in turn underlay numerous domain specific projects [24, 25, 26, 27, 28, 29].”
- VirTex: Learning Visual Representations from Textual Annotations (2020, methodology): “rming SVMs are typically trained with cost C 2f0:01;0:1;1:0;10:0g. Based on this observation, we only train SVMs on these four cost values. We use the same hyperparameters as [22]: we use scikitlearn [106] package with LIBLINEAR [107] backend, default parameters are: LinearSVC(penalty=‘l2’, class we”
- Variational Graph Auto-Encoders (2016, methodology): “For SC, we use the implementation from [11] with an embedding dimension of 128.”
- On Mutual Information Maximization for Representation Learning (2019, positive): “, 2014), as implemented in scikit-learn (Pedregosa et al., 2011).”
- Invariant Information Clustering for Unsupervised Image Classification and Segmentation (2018, neutral): “tation) and IIC* (semi-supervised overclustering) results are shown, together with the ground truth segmentation (GT). COCO-Stuff-3 COCO-Stuff Potsdam-3 Potsdam Random CNN 37.3 19.4 38.2 28.3 K-means [42]y 52.2 14.1 45.7 35.3 SIFT [38]z 38.1 20.2 38.2 28.5 Doersch 2015 [16]z 47.5 23.1 49.6 37.2 Isol”
- Adversarial Example Games (2020, methodology): “We illustrate the idea that the optimal generator and the F -entropy [57] with linear and polynomial (of degree 3 and 5) features.”
- GAIT: A Geometric Approach to Information Theory (2019, methodology): “Note that when using K-means, the resulting allocation of mass from points in the target measure to the nearest centroid can result in a highly unbalanced distribution, shown in the bar plot in orange.”