Python index() doesn't work

Question

I have a list of dataframes like this one :

arr = [df1, df2, df3]

And I want to get a position of element in this list:

position_of_df2 = arr.index(df2)

But python gives error on this line:

ValueError: The truth value of a DataFrame is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

What Im doing wrong? PS what is the best way to get index of dataframe in array? Only iteration?

Not answering your question but you perhaps could solve your problem by storing your data frames in a dict with the key as the name, this would allow you to easily access them. — tmcnicol, May 15 '18 at 09:01

jpp · Accepted Answer · 2018-05-15T08:56:42.473

list.index works by checking equality of your input with elements from the list and looking for True. It then returns the index of the first match.

Testing equality of 2 dataframes returns a dataframe:

df1 = pd.DataFrame([[1, 2]])
df2 = pd.DataFrame([[1, 2]])

print(df1 == df2)

      0     1
0  True  True

The truthness of the result is ambiguous:

print(bool(df1 == df2))

# ValueError: The truth value of a DataFrame is ambiguous.
# Use a.empty, a.bool(), a.item(), a.any() or a.all().

Option 1

In my opinion, the best way to check if you have the correct dataframe is to use an ordered dictionary and define keys (preferably, use descriptive names as keys):

from collections import OrderedDict

o = OrderedDict([(1, df1), (2, df2), (3, df3)])

print(list(o.keys()).index(2))  # 1

Option 2

Use a generator expression with is, which returns True only if 2 variables point to the same object:

lst = [df1, df2, df3]

res = next(i for i, j in enumerate(lst) if j is df2)  # 1

Therefore, iterate manually and check identity (with `is`) instead. — Ignacio Vazquez-Abrams, May 15 '18 at 08:42

score 1 · Answer 2 · answered May 15 '18 at 08:57

1

Based on the answer of @jpp above, also quoting from the answer, https://stackoverflow.com/a/19918849/423725, improvised a solution.

df1 = pandas.DataFrame([1, 2])
df2 = pandas.DataFrame([3, 4])
df3 = pandas.DataFrame([5, 6])

arr = [df1, df2, df3]

def isEqual(df1, df2):
    from pandas.util.testing import assert_frame_equal

    try:
        assert_frame_equal(df1, df2)
        return True
    except:  # appeantly AssertionError doesn't catch all
        return False

def indexDF(df, arr):
    for index, dataframe in enumerate(arr):
        if isEqual(df, dataframe):
            return index

indexDF(df2, arr)
# 1

answered May 15 '18 at 08:57

BcK

2,548
1
13
27

The only issue is if `df1 = df2 = pd.DataFrame([[1, 2]])`, then both dataframes will be considered the same for indexing purposes. I'm not *sure* that's what OP wants, but it should be noted. – jpp May 15 '18 at 09:01
@jpp That's correct. It's confusing to have same dataframes with different id's. Question is, am I looking for the closest one or the one that exactly matches with specified df. – BcK May 15 '18 at 09:07
yeah, it's about the first occurence in array. thank you @BcK for this answer – Scacal Mardan May 15 '18 at 09:14

Python index() doesn't work

2 Answers2