#SVM MNIST digit classification
MNIST digit classification with scikit-learn and Support Vector Machine (SVM) algorithm.
The goal of this project is to prapare the baseline for MNIST classification, you can easily import and run SVM classification.
The file svm_mnist_classification.py presents simple svm with RBF
Params:
- SVM params C = 1 , gamma = 0.0001
- accuracy:
- trainning time:
- Nystroem kernel approximation
- Fourier kernel approximation
- Random Kitchen Sinks
- Download code from GitHub: https://github.com/ksirg/svm_mnist_digit_classification
- Run project at PLON.io: https://plon.io/explore/svm-mnist-handwritten-digit/USpQjoNcO8QHlmG6T
- http://scikit-learn.org/stable/modules/kernel_approximation.html#kernel-approximation
- Random features for large-scale kernel machines Rahimi, A. and Recht, B. - Advances in neural information processing 2007,
- "Random Fourier approximations for skewed multiplicative histogram kernels" - Lecture Notes for Computer Sciencd (DAGM)
- Efficient additive kernels via explicit feature maps Vedaldi, A. and Zisserman, A. - Computer Vision and Pattern Recognition 2010
- Generalized RBF feature maps for Efficient Detection Vempati, S. and Vedaldi, A. and Zisserman, A. and Jawahar, CV - 2010