init_1 = np.random.randn(5, 5)
with pm.Model() as neural_network:
weights_in_1 = pm.Normal(‘w_in_1’, 0, sd=1,
shape=(5, 5), testval=init_1)
return neural_network
As shown in the above code, I used the pm.Normal() function to define aweights_in_1. Now I want to do something with weights_in_1(such as using it as a convolution filter)。 This function returns a FreeRV variable to me. I can’t directly manipulate it. How can I read the matrix data of FreeRV?