CODEWITHSUNDEEP
pythonvaex
Ophir Yoktan
Ophir Yoktan

Reputation: 8449

how can a specific cell be accessed in a vaex data frame?

vaex is a library similar to pandas, that provides a dataframe class I'm looking for a way to access a specific cell by row and column

for example:

import vaex
df = vaex.from_dict({'a': [1,2,3], 'b': [4,5,6]})
df.a[0] # this works in pandas but not in vaex

Upvotes: 1

Views: 2626

Answers (2)

Ben Epstein
Ben Epstein

Reputation: 101

@Maarten Breddels is the owner of Vaex, so I would take his word. But it's possible he wrote that answer before Vaex added slicing, which in this case would be much less "clunky" as described.

import vaex

df = vaex.example()
df.x[:1].values  # Access row 0
df.x[1:3].values  # Access rows 1 and 2

Upvotes: 1

Maarten Breddels
Maarten Breddels

Reputation: 1484

In this specific case you could do df.a.values[0], but if this was a virtual column, it would lead to the whole column being evaluated. What would be faster to do (say in a case of > 1 billon rows, and a virtual column), is to do: 

df['r'] = df.a + df.b
df.evaluate('r', i1=2, i2=3)[0]

This will evaluate the virtual column/expression r, from row 2 to 3 (an array of length 1), and get the first element.

This is rather clunky, and there is an issue open on this: https://github.com/vaexio/vaex/issues/238

Maybe you are surprised that vaex does not have something as 'basic' as this, but vaex is often used for really large datasets, where you don't access individual rows that often, so we don't run into this a lot.

Upvotes: 1

Related Questions