# BrokenProcessPool with sample\_smc example

**URL:** <https://discourse.pymc.io/t/brokenprocesspool-with-sample-smc-example/16701>\
**Category:** General\
**Created:** [March 15, 2025, 4:25pm UTC](https://discourse.pymc.io/t/brokenprocesspool-with-sample-smc-example/16701 "2025-03-15T16:25:05Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Marina\_Thalassini](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/marina_thalassini/32/9055_2.png) [@Marina\_Thalassini](https://discourse.pymc.io/u/Marina_Thalassini)\
**Post date:** [March 15, 2025, 4:25pm UTC](https://discourse.pymc.io/t/brokenprocesspool-with-sample-smc-example/16701/1 "2025-03-15T16:25:05Z")

</div>

Hello,  
I am running the example provided in pymc [website](https://www.pymc.io/projects/examples/en/latest/samplers/SMC-ABC_Lotka-Volterra_example.html) for Aproximate Bayesian Computation. This is the code I run:

```auto
from scipy.integrate import odeint

# Definition of parameters
# a = 1.0 # These values are now passed as arguments to competition_model
# b = 0.1
# c = 1.5
# d = 0.75

# Lotka - Volterra equation
def dX_dt(X, t, a, b, c, d):
    """Return the growth rate of fox and rabbit populations."""
    return np.array([a * X[0] - b * X[0] * X[1], -c * X[1] + d * b * X[0] * X[1]])

# simulator function
def competition_model(rng, a, b, size=100, X0=[10.0, 5.0], time=15, c=1.5, d=0.75):
    """
    Competition model with parameters a, b, c, and d.

    Includes c and d, X0, size, and time as arguments with default values
    to avoid referencing external variables.
    This ensures that the correct values are used within the multiprocessing context.
    """
    t = np.linspace(0, time, size)
    return odeint(dX_dt, y0=X0, t=t, rtol=0.01, args=(a, b, c, d))

# function for generating noisy data to be used as observed data.
def add_noise(a, b, size=100, X0=[10.0, 5.0], time=15, c=1.5, d=0.75):
    """
    Add noise to the competition model output.

    Includes c and d, X0, size, and time as arguments with default values
    to ensure consistency with competition_model.
    """
    noise = np.random.normal(size=(size, 2))
    simulated = competition_model(None, a, b, size=size, X0=X0, time=time, c=c, d=d) + noise
    return simulated

with pm.Model() as model_lv:
    a = pm.HalfNormal("a", 1.0)
    b = pm.HalfNormal("b", 1.0)

    sim = pm.Simulator("sim", competition_model, params=(a, b), epsilon=10, observed=observed)

    idata_lv = pm.sample_smc()

```

Running on PyMC v5.20.1

But i get the followin error from the simulator:

BrokenProcessPool: A process in the process pool was terminated abruptly while the future was running or pending.

---

<div class="post-metadata">

**Author:** ![iavicenna](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/iavicenna/32/7923_2.png) [@iavicenna](https://discourse.pymc.io/u/iavicenna)\
**Post date:** [March 25, 2025, 4:16pm UTC](https://discourse.pymc.io/t/brokenprocesspool-with-sample-smc-example/16701/3 "2025-03-25T16:16:27Z")

</div>

Hello,

Well the example in the page also does not work for me but it seems to be a sizing issue. The simulator class seems to resize everything to two dimensional arrays so for instance constants a,b become 1x1 arrays which breaks the dimension of odeint returning 1 x 1 x N sized arrays and an error. If I slightly modify it, it works for me and produces roughly the same results as in the page. I wonder if this resizing is an intended feature because it seems to make life generally difficult.

 ![Figure 2025-03-25 162135](https://canada1.discourse-cdn.com/flex036/uploads/pymc3/original/2X/3/34a6b474680cd946865296f6f21f6254cb38fc52.png)

The page needs to be updated though.

```auto
import numpy as np
import pymc as pm
import arviz as az
from scipy.integrate import odeint

# Definition of parameters
a = 1.0
b = 0.1
c = 1.5
d = 0.75

# initial population of rabbits and foxes
X0 = [10.0, 5.0]
# size of data
size = 100
# time lapse
time = 15
t = np.linspace(0, time, size)

# Lotka - Volterra equation
def dX_dt(X, t, a, b, c, d):
    """Return the growth rate of fox and rabbit populations."""

    return np.array([a * X[0] - b * X[0] * X[1], -c * X[1] + d * b * X[0] * X[1]])

# simulator function
def competition_model(rng, a, b, size=None):
  
    if isinstance(a, np.ndarray) and a.ndim==2:
      a = a[0,0]
    if isinstance(b, np.ndarray) and b.ndim==2:
      b = b[0,0]
  
    return odeint(dX_dt, y0=X0, t=t, rtol=0.01, args=(a, b, c, d))
  
# function for generating noisy data to be used as observed data.
def add_noise(a, b):
    noise = np.random.normal(size=(size, 2))
    simulated = competition_model(None, a, b) + noise
    return simulated

observed = add_noise(a, b)
with pm.Model() as model_lv:
    a = pm.HalfNormal("a", 1.0)
    b = pm.HalfNormal("b", 1.0)

    sim = pm.Simulator("sim", competition_model, params=(a, b), epsilon=10, observed=observed)
    sim.eval()
    idata_lv = pm.sample_smc()

az.plot_trace(idata_lv, kind="rank_vlines")

```