18 questions
1
vote
1
answer
295
views
How to use PyMC, PyMC-Marketing, and NumPyro together when they require conflicting JAX versions?
I’m trying to build a probabilistic model in Python that requires using PyMC, PyMC-Marketing, and NumPyro in the same script.
My environment: Python 3.12, Windows 64-bit
Installed packages:
pymc (...
- 27
0
votes
0
answers
37
views
Numpyro producing the exact same value for parameters for every sample
I have written the following numpyro model which models a gaussian process and then performs binomial regression. The issue i am having is that I get the same values for the parameters for every ...
- 3,878
0
votes
0
answers
34
views
Hierarchical Time Series Modeling for Data with Different Lengths in Numpyro
Before we get started, this is what I am importing:
import numpy as np
import os
import pandas as pd
import matplotlib.pyplot as plt
idx = pd.IndexSlice
# import jax
from jax import random
import jax....
0
votes
0
answers
68
views
Making predictions using numpyro and MCMC
I am following the numpyro example in celerite2 here, which has a numpyro interface.
Further up using emceee they make predictions here
I have tried to make predictions using the numpyro interface, ...
- 849
1
vote
0
answers
239
views
Jupyter kernel in vscode crashes after installing numpyro
I have created an python environment using conda (python 3.12) which includes geopandas and numpyro. The minute I install numpyro using either pip or conda/mamba (I tried both), My jupyter kernel ...
- 3,878
1
vote
1
answer
358
views
Numpyro AR(1) mean switching model sampling incongrouencies
I'm trying to estimate an AR(1) process y with a switching mean according to a latent state S =0,1 that evolves as a markov process with fixed transition probabilities (as in here). In short, it takes ...
1
vote
0
answers
50
views
Estimating the posterior of a parameter vector which is "partially observed" in Numpyro
I have a parameter space that is M dimensional. My prior on the parameters is independent N(0,1) in that space. My observation space is a full rank linear transformation of that parameter space. If ...
- 11
1
vote
0
answers
74
views
Discepency in results fitting a bambi model with pymc backend locally, compared to remote repository - with same package versions in both environments
I’m working on a Bayesian model using the bambi modeling interface, Pymc backend and numpyro as the inference library.
Problem
When I fit my model locally using the following line of code, the ...
- 11
2
votes
1
answer
6k
views
Google Colab can't change Python version
Recently, all my notebooks have been updated to python 3.10, and I can't run my old code. I was using jax 0.2.17 and numpyro 0.7.1, but these no longer work with my version, so I tried changing to ...
1
vote
0
answers
292
views
Numpyro: Error when using MCMC with a model that uses scan
I am trying to get the following model to work:
def model_dynamic(self, hemp_size_t, values_t):
# Unpack the values at time t
t, actions_performed = values_t
# Check if harvesting are ...
2
votes
2
answers
2k
views
JAX with JIT and custom differentiation
I am working with JAX through numpyro. Specially, I want to use a B-spline function (e.g. implemented in scipy.interpolate.BSpline) to transform different points into a spline where the input depends ...
- 4,841
1
vote
1
answer
484
views
GARCH models in Numpyro
I was curious how best to implement GARCH models in numpyro. I tried reading https://num.pyro.ai/en/stable/tutorials/time_series_forecasting.html but found it generally unclear (the model notation ...
- 691
-1
votes
1
answer
146
views
How to predict time series with limited data
I have a dataset with four columns: date, category, product, rate(%). I would like to be able to forecast the rate for every product in my data. The major issue I'm having is that because products ...
- 25
1
vote
0
answers
88
views
Pyro for differential equations
Is it possible to infer differential equation parameters using pyro? I found an example with numpyro I was wondering if this is possible with pyro as well?
- 45
0
votes
0
answers
78
views
Gaussian Process Kernels for time series : one kernel for t<0 and one for t>0, is it possible?
I have small time series, (t_i, Y_i)_{i<N} (N ~ 20 and varying from one to another). Now, I know that for some t_i <0 the behaviour is different than for t_i>0, and from one series to the ...
- 402
3
votes
1
answer
228
views
numpyro.render_model cannot be found
I am working with the Bayesian models in NumPyro. It is a relatively new library. I tried to visualize my model by following the Numpyro manual: http://num.pyro.ai/en/latest/tutorials/model_rendering....
- 31
1
vote
1
answer
593
views
A JAX custom VJP function for multiple input variable does not work for NumPyro/HMC-NUTS
I am trying to use a custom VJP (vector-Jacobian product) function as a model for a HMC-NUTS in numpyro. I was able to make a single variable function that works for HMC-NUTS as follows:
import jax....
13
votes
1
answer
9k
views
NumPyro vs Pyro: Why is former 100x faster and when should I use the latter?
From Pytorch-Pyro's website:
We’re excited to announce the release of NumPyro, a NumPy-backed Pyro using JAX for automatic differentiation and JIT compilation, with over 100x speedup for HMC and ...
- 1,541