search

rois-codh / kmnist

Repository for Kuzushiji-MNIST, Kuzushiji-49, and Kuzushiji-Kanji
Creative Commons Attribution Share Alike 4.0 International
693 stars 94 forks source link

KMNIST kNN #10

Closed dzisandy closed 4 years ago

dzisandy commented 5 years ago

Adding PCA dimension reduction produces better Test Accuracy x_train = x_train.reshape(-1, 784) x_test = x_test.reshape(-1, 784) After this code rows: pca = PCA(n_components= 60, random_state= 0 ) x_train = pca.fit_transform(x_train) x_test = pca.transform(x_test) The resulting is Test accuracy: 0.9398

hardmaru commented 5 years ago

Thanks, that’s a good datapoint to know!

On Tue, Mar 26, 2019 at 8:25 AM Andrei Dzis notifications@github.com wrote:

Closed #10 https://github.com/rois-codh/kmnist/issues/10.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/rois-codh/kmnist/issues/10#event-2228155951, or mute the thread https://github.com/notifications/unsubscribe-auth/AGBoHlSaYoK1E5J_aVbVI3pW5nCGBo2vks5vaVr2gaJpZM4cJ4xh .

mxbi commented 4 years ago

Added, it would be good to also have data points for MNIST and K-49

dzisandy commented 4 years ago

@mxbi Thanks for credit!

mxbi commented 4 years ago

See also #13 for K-49 and MNIST code/details