How to use Tensorflow for signal processing?

Recently, I began to study CNN and tensor flow (beginner). I am trying to train a simple convolutional neuron to identify my randomly generated exponential signals from their tau value (time constant).

My question is how to assign them to classes and find out how the tensor process will be learned from this data?

The code I used to generate the signals below is

import matplotlib.pyplot as plt
import random
import numpy as np

lorange= 1
hirange= 10
amplitude= random.uniform(-10,10)
t= 10
random.seed()
tau=random.uniform(lorange,hirange)
x=np.arange(t)

plt.xlabel('t=time')
plt.ylabel('x(t)')
plt.plot(x, amplitude*np.exp(-x/tau))
plt.show()
+4
source share
1 answer

@zerogravty

It's great that you do signal processing with Tensorflow or deep learning in Tensorflow.

github Tensorflow. . .

+4

All Articles