saving

checking out how to save current weight matrices and load them back in.



In [1]:

    
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

first just checking that the flattening and reshaping works as expected



In [13]:

    
test = np.random.randn(11,11,4,100)
test.shape









    Out[13]:





(11, 11, 4, 100)



In [14]:

    
test_flat = test.flatten()
test_flat.shape









    Out[14]:





(48400,)



In [17]:

    
np.savetxt('test.txt', test_flat)
test_back = np.loadtxt('test.txt').reshape((11,11,4,100))
test_back.shape









    Out[17]:





(11, 11, 4, 100)



In [18]:

    
np.mean(test - test_back)









    Out[18]:





0.0

looks good.