notebooks

ComputationalPsychiatry · Nov 7, 2023 · 916870d · 916870d
1 parent dddbc32
commit 916870d
Show file tree

Hide file tree

Showing 3 changed files with 124 additions and 89 deletions.
diff --git a/docs/source/notebooks/1.1-Binary_HGF.ipynb b/docs/source/notebooks/1.1-Binary_HGF.ipynb
diff --git a/docs/source/notebooks/1.1-Binary_HGF.md b/docs/source/notebooks/1.1-Binary_HGF.md
@@ -179,6 +179,12 @@ three_levels_hgf = three_levels_hgf.input_data(input_data=u)
 three_levels_hgf.plot_trajectories();
 ```
 
+#### Surprise
+
+```{code-cell} ipython3
+three_levels_hgf.surprise()
+```
+
 ## Learning parameters with MCMC sampling
 In the previous section, we assumed we knew the parameters of the HGF models that were used to filter the input data. This can give us information on how an agent using these values would behave when presented with these inputs. We can also adopt a different perspective and consider that we want to learn these parameters from the data. Here, we are going to set some of the parameters free and use Hamiltonian Monte Carlo methods (NUTS) to sample their probability density.
 

diff --git a/docs/source/notebooks/1.3-Continuous_HGF.ipynb b/docs/source/notebooks/1.3-Continuous_HGF.ipynb