Learning Random Forests on the GPU
نویسندگان
چکیده
Random Forests are a popular and powerful machine learning technique, with several fast multi-core CPU implementations. Since many other machine learning methods have seen impressive speedups from GPU implementations, applying GPU acceleration to random forests seems like a natural fit. Previous attempts to use GPUs have relied on coarse-grained task parallelism and have yielded inconclusive or unsatisfying results. We introduce CudaTree, a GPU Random Forest implementation which adaptively switches between data and task parallelism. We show that, for larger datasets, this algorithm is faster than highly tuned multi-core CPU implementations.
منابع مشابه
CURFIL: Random Forests for Image Labeling on GPU
Random forests are popular classifiers for computer vision tasks such as image labeling or object detection. Learning random forests on large datasets, however, is computationally demanding. Slow learning impedes model selection and scientific research on image features. We present an open-source implementation that significantly accelerates both random forest learning and prediction for image ...
متن کاملCURFIL: A GPU Library for Image Labeling with Random Forests
Random forests are popular classifiers for computer vision tasks such as image labeling or object detection. Learning random forests on large datasets, however, is computationally demanding. Slow learning impedes model selection and scientific research on image features. We present an open-source implementation that significantly accelerates both random forest learning and prediction for image ...
متن کاملRandom forests algorithm in podiform chromite prospectivity mapping in Dolatabad area, SE Iran
The Dolatabad area located in SE Iran is a well-endowed terrain owning several chromite mineralized zones. These chromite ore bodies are all hosted in a colored mélange complex zone comprising harzburgite, dunite, and pyroxenite. These deposits are irregular in shape, and are distributed as small lenses along colored mélange zones. The area has a great potential for discovering further chromite...
متن کاملInteractive Texture Segmentation using Random Forests and Total Variation
Common methods for interactive texture segmentation rely on probability maps based on low dimensional features such as e.g. intensity or color, that are usually modeled using basic learning algorithms such as histograms or Gaussian Mixture Models. The use of low level features allows for fast generation of these hypotheses but limits applicability to a small class of images. We address this pro...
متن کاملFault Locating in High Voltage Transmission Lines Based on Harmonic Components of One-end Voltage Using Random Forests
In this paper, an approach is proposed for accurate locating of single phase faults in transmission lines using voltage signals measured at one-end. In this method, harmonic components of the voltage signals are extracted through Discrete Fourier Transform (DFT) and are normalized by a transformation. The proposed fault locator, which is designed based on Random Forests (RF) algorithm, is train...
متن کامل