Probabilistic-Programming-and-Bayesian-Methods-for-Hackers

Probabilistic-Programming-and-Bayesian-Methods-for-Hackers copied to clipboard

The default of `return_inferencedata` was `False` but in the future (v4.0) it will be `True`. However, the rest of the code is written for `False` case so I changed all...

toppare

Ch2, tensorflow import issue

1

Hi there, I was tryting to just run all cells on the tensorflow chapter 2 book, and it hits this error right at the start. ``` AttributeError Traceback (most recent...

worthy7

I think we want to find the posterior distr. of `cheaters`, not `liars`

toppare

Chapter 6 issue: ValueError: num must be 1 <= num <= 4, not 0

4

Hi, found this issue in Chapter 6, changing plt.subplot(2, 2, i) to plt.subplot(2, 2, i+1) fixes the issue with the indexes, but you might want to adopt a different approach...

attibalazs

AttributeError: module 'tensorflow' has no attribute 'contrib'

8

tensorflow.contrib.eager is no longer in TensorFlow, but it's used in the jupyter notebooks and in google colab. ```python import tensorflow as tf tfe = tf.contrib.eager # Eager Execution #@markdown Check...

natefrisch01

Change trace code due to update in PyMC library

1

nhkhoi91

Chapter 2 link to GreenhillWardSacks paper is broken.

1

This link (in the challenger example)- http://mdwardlab.com/sites/default/files/GreenhillWardSacks.pdf seems to be dead. I did find a version on github but just fyi.

EffiFeld

Chap 1, Section: Introducing our first hammer: PyMC3, example error, PyMC3,

2

PyMC3, python 3.84, win10, Jupyter-lab code from Chap1 Sec: Introducing our first hammer: PyMC3, import pymc3 as pm import theano.tensor as tt with pm.Model() as model: alpha = 1.0/count_data.mean() #...

lrlgogo

Section: A Proposed Solution The correction one should be: the posterior is a Beta(a=1+X, b=1+N-X)

lc82111

So far I like this book. It is very well written. I don't want to sound pedantic here, but I don't like the description of the continuous distribution in chap...

HVoltBb