Genetic algorithm to train the weights of a neural
2016-08-23
0 0 0
no vote
Other
Earn points
Takes a text file as input that contains the x and y coordinates of a series of points on a curve. The code then uses a genetic algorithm to train a neural network to approximate the same curve.
Runs from the python command line; an example input command would look like: python nn_weights_1.py -r 10 -n 100 -g 500 -t 3600.0 -a RandomCrossOver -s random_seeds_100.txt -i sine_datapoints1.txt -o results.txt
where nn_weights_1.py is the name of the code file,
-r tells it how many runs to perform,
-n tells it how many individuals are in the population,
-g tells it how many generations in each run,
-t tells it a maximum time to run (in seconds),
python
算法
神经网络
训练
遗传
权值
Related Source Codes
Word memorization software
0
0
no vote
CWT Python example
0
0
no vote
STFT example
0
0
no vote
Chinese Speech Recognition
0
0
no vote
python-ela
0
0
no vote
No comment