fixing images

bayes-nets/approximate.md

@@ -80,4 +80,4 @@ We will not prove this, but if we repeat this process enough times, our later sa
 
 The pseudocode for Gibbs Sampling is provided below.
 
-![Gibbs Sampling](../assets/images/Gibbs.png)
+<img src="{{ site.baseurl }}/assets/images/Gibbs.png" alt="Gibbs Sampling" />

bayes-nets/d-separation.md

@@ -17,10 +17,10 @@ We will present all three canonical cases of connected three-node two-edge Bayes
 
 ## 6.4.1 Causal Chains
 
-![Causal Chain with no observations](../assets/images/chain_free.PNG)
+<img src="{{ site.baseurl }}/assets/images/chain_free.PNG" alt="Causal Chain with no observations" />
 *Figure 1: Causal Chain with no observations.*
 
-![Causal Chain with Y observed](../assets/images/chain_observed.PNG)
+<img src="{{ site.baseurl }}/assets/images/chain_observed.PNG" alt="Causal Chain with Y observed" />
 *Figure 2: Causal Chain with Y observed.*
 
 Figure 1 is a configuration of three nodes known as a **causal chain**. It expresses the following representation of the joint distribution over $$X$$, $$Y$$, and $$Z$$:
@@ -178,7 +178,7 @@ Any path in a graph from $$X$$ to $$Y$$ can be decomposed into a set of 3 consec
 
 Here are some examples of applying the $$d$$-separation algorithm:
 
-![Example 1](../assets/images/rbtt.png)
+<img src="{{ site.baseurl }}/assets/images/rbtt.PNG" alt="Example 1" />
 
 This graph contains the common effect and causal chain canonical graphs. 
 

bayes-nets/representation.md

@@ -35,7 +35,7 @@ As an example of a Bayes Net, consider a model where we have five binary random
 
 Assume the alarm can go off if either a burglary or an earthquake occurs, and that Mary and John will call if they hear the alarm. We can represent these dependencies with the graph shown below.
 
-![Basic Bayes Nets Example](../assets/images/basic_bayes_nets.png)
+<img src="{{ site.baseurl }}/assets/images/basic_bayes_nets.png" alt="Basic Bayes Nets Example" />
 
 <p></p>
 In this Bayes Net, we would store probability tables $$P(B)$$, $$P(E)$$, $$P(A | B, E)$$, $$P(J | A)$$ and $$P(M | A)$$.