issues
search
accosmin
/
nano
C++ library [machine learning & numerical optimization] - superseeded by libnano
MIT License
1
stars
0
forks
source link
Zero initialization
#192
Closed
accosmin
closed
6 years ago
accosmin
commented
6 years ago
add model_t::zero() to initialize its parameters to zero.
configurable batch_trainer_t to specify the initialization method (zero or random).
refactor experimentation scripts: use bash, export only .json & .csv & .log files from the experiments
generic gnuplot-based scripts to plot the .csv files (using lines, boxes etc.)