Currently I am trying to plot a simple Relu function using matplotlib. My problem is that when plotting, matplotlib ignores the zeroes and does not plot them.
Heres my functions
def relu(t):
t[t <= 0] = 0
return t
def relu_plot():
t = np.arange(-2, 2, 0.1)
k = np.zeros(40)
plt.plot(t, relu(t))
plt.show()
print(len(relu(t)))
relu_plot()
And here is what matplotlib returns
The plot works if I instead use an array of size 40 with only zeroes, and the relu function is indeed returning an array of size 40. Any help would be appreciated
Update: Instead of just plotting relu(t), I tried plotting a concatenated array as follows
k = np.concatenate((np.zeros(20), j))
plt.plot(t, k)
And it works as expected. I think then that it is a problem with how I changed the values to 0 in the relu function but I am not sure yet.
>Solution :
Your relu function modifies t, so using t for your x variable produces the unexpected results (no longer spanning -2 to 2, but 0 to 2).
def relu(t):
return t.clip(min=0)
Output:
