tfp.experimental.nn.util.make_kernel_bias_posterior_mvn_diag

View source on GitHub

Create learnable posterior for Variational layers with kernel and bias.

kernel_shape ...
bias_shape ...
kernel_initializer ... Default value: None (i.e., tf.initializers.glorot_uniform()).
bias_initializer ... Default value: None (i.e., tf.initializers.zeros()).
kernel_batch_ndims ... Default value: 0.
bias_batch_ndims ... Default value: 0.
dtype ... Default value: tf.float32.
kernel_name ... Default value: "posterior_kernel".
bias_name ... Default value: "posterior_bias".

kernel_and_bias_distribution ...