Instantly share code, notes, and snippets.

View H infinity smallest eigs.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
View censored normal moments.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
View patch_biblatex.sty
% biblatex authoryear with natbib does a million stray commas in versions < 3.5
% see: https://tex.stackexchange.com/q/312539/
% this works around it for versions 3.3 and 3.4
\ifthenelse{\equal{\abx@version}{3.4}}{%
\DeclareDelimFormat[cbx@textcite]{nameyeardelim}{\addspace}
}{\ifthenelse{\equal{\abx@version}{3.3}}{%
% Workaround following http://golatex.de/neue-biber-version-ploetzlich-komma-bei-zitierung-t17005.html
% (in German...), which changes \nameyeardelim / \nonameyeardelim to \addspace
% in the citation style's textcite macro.
% sigh....
View alphanumeric-hybrid.bbx
% This is an unholy mashup of the numeric.bbx and alphabetic.bbx files from
% the standard biblatex distribution, made so we can switch between numeric
% style for the main body and alphabetic style for the appendix.
\ProvidesFile{alphanumeric-hybrid.bbx}
[\abx@bbxid]
\RequireBibliographyStyle{standard}
\ExecuteBibliographyOptions{labelalpha,labelnumber,sorting=anyt}
% To use the alphabetic section in one refsection, \setbool{bbx:alpha}{true} at the start;
View two-sample tests via set classification.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
View two-sample tests via set classification.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
View mkl dgemm_batch.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
View autoencoder_stuff.py
from __future__ import division
import json
import numpy as np
import keras
from keras.datasets import mnist
from keras import backend as K, layers as L
from keras.models import load_model, Model, model_from_config
View fix_paths.sh
#!/bin/bash
CUDA_LIB_DIR=/usr/local/cuda/lib
CUDA_VERSION=7.5
CUDA_LIBS="cublas cudart curand"
CUDNN_LIB_DIR=/usr/local/cuda/cudnn-3/lib
CUDNN_VERSION=7.0
CUDNN_LIBS="cudnn"