This repository, is the code base for the Neural Network course. The documentation for the code can be found in the html folder, you can visualized with a broswer , Therefore it's necessary to clone the git and then open the index.html file which can be found in html/project/index.html a simple snippet of code to run the network is :
import project.test.music_test as tmusic
import numpy as np
import project.esn.core as c
def test_generated():
train_len = test_len = 6200
init_len = 3200
music = (tgen.test_patterns[2] * 300)
data = c.Data(np.array(list(~(tmusic.all * 10))), (tmusic.all * 10).tempo,
init_len,train_len,test_len)
with c.Run(
**
{
"data": data,
"reservoir": 5000,
"in_out": 9,
"leaking_rate": 0.3,
"reg": 1e-8,
"transformer": ta.Transformers.pow_prob,
"t_param": 1,
"density": 0.04
"spectral_radius": 1.3
"t_squeeze": np.tanh,
"noise" : 0
}) as gen:
return gen()