167
167
"source": [
168
168
"#### Code 2.6\n",
169
169
"\n",
170
- "Computing the posterior using the quadratic aproximation (quad)."
170
+ "Computing the posterior using the quadratic approximation (quad)."
171
171
]
172
172
},
173
173
{
213
213
],
214
214
"source": [
215
215
"data = np.repeat((0, 1), (3, 6))\n",
216
- "with pm.Model() as normal_aproximation :\n",
216
+ "with pm.Model() as normal_approximation :\n",
217
217
" p = pm.Uniform(\"p\", 0, 1) # uniform priors\n",
218
218
" w = pm.Binomial(\"w\", n=len(data), p=p, observed=data.sum()) # binomial likelihood\n",
219
219
" mean_q = pm.find_MAP()\n",
347
347
"\n",
348
348
"for idx, ps in enumerate(zip(w, n)):\n",
349
349
" data = np.repeat((0, 1), (ps[1] - ps[0], ps[0]))\n",
350
- " with pm.Model() as normal_aproximation :\n",
350
+ " with pm.Model() as normal_approximation :\n",
351
351
" p = pm.Uniform(\"p\", 0, 1) # uniform priors\n",
352
352
" w = pm.Binomial(\"w\", n=len(data), p=p, observed=data.sum()) # binomial likelihood\n",
353
353
" mean_q = pm.find_MAP()\n",
472
472
},
473
473
"nbformat": 4,
474
474
"nbformat_minor": 1
475
- }
475
+ }
0 commit comments