I have data with potentially 3 switch points: [data] The model will automatically choose Metropolis for the sampling of the tau_1 and tau_2, and NUTS for the lambdas. N = np.arange(0, all_points) with pm.Model() as model: alpha = 1.0/data.mean() lambda_1 = pm.Exponential(“lambda_1”, alpha) l…

Yes, more details is in the notebook, link copied below: junpenglao/Planet_Sakaar_Data_Science/blob/master/PyM3QnA/stackoverflow_49144144.ipynb { "cells": [ { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stderr", "output_ty…

Switch point Metropolis tuning

Questions

junpenglao April 13, 2018, 5:58am 6

That’s pretty strange… what is your pymc3 version?
Also, could you please paste the full error trace?

Topic		Replies	Views
Sampling time is very long (minutes per sample) Questions	13	4561	November 2, 2017
Geometric variable not being properly Sampled Questions from_github	22	3328	June 22, 2017
Converting Model from Stan Questions	15	3176	June 27, 2017
Correctly applying switchpoint model? Questions	2	2352	April 23, 2019
Sampling the Banana-shaped distribution Questions	14	3872	July 24, 2018

Switch point Metropolis tuning

Related topics