How can I pass the AttributeError: 'DataFrame' object has no attribute 'flatten in Python?

Question

我已经使用 python 创建了空间马尔可夫矩阵但是我得到了这个错误（AttributeError: 'DataFrame' object has no attribute 'flatten' ）。我对python不熟悉所以，希望能帮我解决这个问题？代码是

import numpy as np
import pysal`enter code here`
import pandas as pd
import pysal as ps

f = pd.read_csv("C:\Users\Yousif\Desktop\Spatial.MC\updated_testdata1-36-Copy.csv")
f

pci = f[list(map(str,range(1, 1096)))]

#f = ps.open(ps.examples.get_path("usjoin36.csv"))
#f
#pci = np.array([f.by_col[str(y)] for y in range(1,1096)])

pci = pci.transpose()
rpci = pci/(pci.mean(axis=0))

#w = ps.open(ps.examples.get_path("states36.gal")).read()

w = ps.open("C:/Users/Yousif/Desktop/Spatial.MC/states-36-Copy.gal").read( )

w.transform = 'r'

sm = ps.Spatial_Markov(rpci, w, fixed=True, k=5, variable_name='rpci')
for p in sm.P:
    print(p)

Answer 1

似乎 ps.Spatial_Markov 想要一个 numpy 数组，而您传递的是 pandas 数据框。尝试通过修改此行

从数据框中提取数组

pci = f[list(map(str,range(1, 1096)))].as_matrix()

更新

的确，docs 是关于这一点的：

pysal.spatial_dynamics.markov.Spatial_Markov(y, w, k=4, permutations=0, fixed=False, variable_name=None)

Parameters:
- y (array) – (n,t), one row per observation, one column per state of each observation, with as many columns as time periods. ...

How can I pass the AttributeError: 'DataFrame' object has no attribute 'flatten in Python?

How can I pass the AttributeError: 'DataFrame' object has no attribute 'flatten in Python?

python

spatial

markov-chains