Kymatio: Scattering Transforms in Python
Authors: Mathieu Andreux, Tomás Angles, Georgios Exarchakis, Roberto Leonarduzzi, Gaspar Rochette, Louis Thiry, John Zarka, Stéphane Mallat, Joakim Andén, Eugene Belilovsky, Joan Bruna, Vincent Lostanlen, Muawiz Chaudhary, Matthew J. Hirn, Edouard Oyallon, Sixin Zhang, Carmine Cella, Michael Eickenberg
JMLR 2020 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Several examples are provided with the code, illustrating the power of Kymatio. These include image reconstruction and generation from scattering (Angles and Mallat, 2018), hybrid scattering and CNN training on CIFAR and MNIST (Oyallon et al., 2018), regression of molecular properties on QM7/QM9 using solid harmonic scattering (Eickenberg et al., 2017), and classifying recordings of spoken digits. |
| Researcher Affiliation | Academia | Mathieu Andreux EMAIL Tom as Angles EMAIL Georgios Exarchakis EMAIL Roberto Leonarduzzi EMAIL Gaspar Rochette EMAIL Louis Thiry EMAIL John Zarka EMAIL Ecole normale sup erieure, CNRS, PSL Research University, 45, rue d Ulm, 75005 Paris, France St ephane Mallat EMAIL Ecole normale sup erieure, CNRS, PSL Research University, 45, rue d Ulm, 75005 Paris, France Coll ege de France, 11, place Marcelin-Berthelot 75231 Paris, France Flatiron Institute, 162 5th Avenue, New York, NY 10010, USA Joakim And en EMAIL Flatiron Institute, 162 5th Avenue, New York, NY 10010, USA Eugene Belilovsky EMAIL Mila, Universit e de Montr eal, 6666 St Urbain Street, Montreal, Quebec H2S 3H1, Canada Joan Bruna EMAIL Vincent Lostanlen EMAIL New York University, 70 Washington Square South, New York, NY 10012, USA Muawiz Chaudhary EMAIL Western Washington University, 516 High Street, Bellingham, WA 98225, USA Matthew J. Hirn EMAIL Michigan State University, 426 Auditorium Road East Lansing, MI 48824, USA Edouard Oyallon EMAIL CNRS, LIP6, Sorbonne University, 4 place Jussieu, 75252 Paris, France Sixin Zhang EMAIL Peking University, No. 5 Yiheyuan Road, Haidian District, Beijing 100871, China Carmine Cella EMAIL University of California, Berkeley, 101 Sproul Hall, Berkeley, CA 94720, USA Michael Eickenberg EMAIL Flatiron Institute, 162 5th Avenue, New York, NY 10010, USA |
| Pseudocode | No | The paper describes the mathematical definition and implementation details of the scattering transform, but it does not provide any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Source code, documentation, and examples are available under a BSD license at https://www.kymat.io. |
| Open Datasets | Yes | Several examples are provided with the code, illustrating the power of Kymatio. These include image reconstruction and generation from scattering (Angles and Mallat, 2018), hybrid scattering and CNN training on CIFAR and MNIST (Oyallon et al., 2018), regression of molecular properties on QM7/QM9 using solid harmonic scattering (Eickenberg et al., 2017), and classifying recordings of spoken digits. |
| Dataset Splits | No | The paper mentions several datasets (CIFAR, MNIST, QM7/QM9) in the context of demonstrating the Kymatio software's capabilities through examples, but it does not provide specific details regarding training, validation, or test splits for these datasets. |
| Hardware Specification | No | The transforms are implemented on both CPUs and GPUs, the latter offering a significant speedup over the former. However, no specific CPU or GPU models or detailed hardware specifications are provided. |
| Software Dependencies | No | This article presents Kymatio, a scattering transform implementation that is user-friendly, well-documented, fast, and compatible with existing automatic differentiation libraries. It brings together transforms in 1D, 2D, and 3D under a unified application programming interface (API). [...] Frontends are provided for many frameworks, including Num Py, scikit-learn, Py Torch, and Tensor Flow/Keras, allowing for seamless integrating scattering transforms in a variety of pipelines. In particular, the Py Torch, and Tensor Flow/Keras frontends allow for inclusion into many deep learning workflows. However, no specific version numbers for these software dependencies are provided. |
| Experiment Setup | No | The paper describes the Kymatio software package and mentions its use in various applications, but it does not provide specific experimental setup details such as hyperparameter values, training configurations, or system-level settings for these applications. |