|
209 | 209 | "\n", |
210 | 210 | "# For the already prepared, I'm using Binomial's conj. prior.\n", |
211 | 211 | "for k, N in enumerate(n_trials):\n", |
212 | | - " sx = plt.subplot(len(n_trials)/2, 2, k+1)\n", |
| 212 | + " sx = plt.subplot(len(n_trials)//2, 2, k+1)\n", |
213 | 213 | " plt.xlabel(\"$p,ドル probability of heads\") \\\n", |
214 | 214 | " if k in [0, len(n_trials)-1] else None\n", |
215 | 215 | " plt.setp(sx.get_yticklabels(), visible=False)\n", |
|
650 | 650 | ], |
651 | 651 | "source": [ |
652 | 652 | "import pymc3 as pm\n", |
653 | | - "import theano.tensor as tt\n", |
654 | 653 | "\n", |
655 | 654 | "with pm.Model() as model:\n", |
656 | 655 | " alpha = 1.0/count_data.mean() # Recall count_data is the\n", |
|
730 | 729 | "### Mysterious code to be explained in Chapter 3.\n", |
731 | 730 | "with model:\n", |
732 | 731 | " step = pm.Metropolis()\n", |
733 | | - " trace = pm.sample(10000, tune=5000,step=step)" |
| 732 | + " trace = pm.sample(10000, tune=5000,step=step, return_inferencedata=False)" |
734 | 733 | ] |
735 | 734 | }, |
736 | 735 | { |
|
0 commit comments