使用pymc.Matplot绘制PyMC模型时出错

Question

成功拟合后，使用pymc.Matplot绘制PyMC模型时出现错误

#!/usr/bin/env python2
import pylab
import pymc as pm
import numpy as np

import seir as mod
reload(mod) 

mc = pm.MCMC(mod)
mc.use_step_method(pm.AdaptiveMetropolis, [mod.beta, mod.gamma, mod.s0, mod.eF, mod.i0])
mc.sample(iter = 1000000, burn = 200000, thin = 500, verbose = 1)

Plotting eF


Traceback (most recent call last):

  File "<ipython-input-14-31405478c9cb>", line 1, in <module>
    pm.Matplot.plot(mc)

  File "C:\Users\tele\Anaconda\lib\site-packages\pymc\Matplot.py", line 341, in wrapper
    f(data, name, *args, **kwargs)

  File "C:\Users\tele\Anaconda\lib\site-packages\pymc\Matplot.py", line 455, in plot
    fontmap=fontmap)

  File "C:\Users\tele\Anaconda\lib\site-packages\pymc\Matplot.py", line 377, in wrapper
    f(pymc_obj, *args, **kwargs)

  File "C:\Users\tele\Anaconda\lib\site-packages\pymc\Matplot.py", line 578, in histogram
    hist(data.tolist(), bins, histtype='stepfilled')

  File "C:\Users\tele\Anaconda\lib\site-packages\matplotlib\pyplot.py", line 2958, in hist
    stacked=stacked, data=data, **kwargs)

  File "C:\Users\tele\Anaconda\lib\site-packages\matplotlib\__init__.py", line 1810, in inner
    return func(ax, *args, **kwargs)

  File "C:\Users\tele\Anaconda\lib\site-packages\matplotlib\axes\_axes.py", line 5995, in hist
    m, bins = np.histogram(x[i], bins, weights=w[i], **hist_kwargs)

  File "C:\Users\tele\Anaconda\lib\site-packages\numpy\lib\function_base.py", line 249, in histogram
    n += np.bincount(indices, weights=tmp_w, minlength=bins).astype(ntype)

ValueError: operands could not be broadcast together with shapes (11,) (12,) (11,) 

在某些情况下会出现错误，但并不总是如此。由于这是pyMC的一个内部函数，而且fit工作正常(实际上非常好)，我不确定该怎么做(顺便说一句，我在使用相同模型调用scores = pm.geweke(mc, intervals=20)时通常会发现另一个错误(不确定是否相关))。

Answer 1

刚刚发现了抽取numpy数组的错误，愚蠢的问题: pyMC崩溃，因为代表可能性的随机变量的平均值(模拟)和观测值的大小不相同。顺便说一句，pyMC提供的错误并不是很有表现力。