11 Data import

11.1 Introduction

In this chapter, you’ll learn how to read plain-text rectangular files into Python. Here, we’ll only scratch the surface of data import, but many of the principles will translate to other forms of data. We’ll finish with a few pointers to packages that are useful for other types of data.

11.1.1 Prerequisites

In this chapter, you’ll learn how to load flat files in R with the pandas package.

import pandas as pd
import altair as alt
import numpy as np

11.2 Getting started

The pandas input/output functions are concerned with turning varied files into pandas data frames for use in Python:

pd.read_csv() reads comma delimited files, pd.read_table() reads general delimited files with any delimiter.
pd.read_fwf() reads fixed width files. You can specify fields either by their widths with widths or their position with colspecs arguments.

These functions all have similar syntax: once you’ve mastered one, you can use the others with ease. For the rest of this chapter we’ll focus on pd.read_csv(). Not only are csv files one of the most common forms of data storage, but once you understand pd.read_csv(), you can easily apply your knowledge to all the other input/output functions in pandas. You can see the full set of functions by reading the pandas documentation.

The first argument to pd.read_csv() is the most important: it’s the path to the file to read.

heights = pd.read_csv("data/heights.csv")

You can also supply an inline csv file with the use of StringIO in the io package. StringIO creates a buffer from a string. This is useful for experimenting with pandas and for creating reproducible examples to share with others:

from io import StringIO

data = StringIO("""
a,b,c
1,2,3
4,5,6
""")

pd.read_csv(data)
#>    a  b  c
#> 0  1  2  3
#> 1  4  5  6

In both cases pd.read_csv() uses the first line of the data for the column names, which is a very common convention. There are two cases where you might want to tweak this behaviour:

Sometimes there are a few lines of metadata at the top of the file. You can use skiprows = n to skip the first n lines; or use comment = "#" to drop all lines that start with (e.g.) #. Both arguments have other methods that can be used - see the pandas documentation.


data_metada = StringIO("""The first line of metadata
  The second line of metadata
  x,y,z
  1,2,3""")

pd.read_csv(data_metada, skiprows = 2)
#>      x  y  z
#> 0    1  2  3
data_comment = StringIO("""# A comment I want to skip
  x,y,z
  1,2,3
""")

pd.read_csv(data_comment, comment = "#")
#>      x  y  z
#> 0    1  2  3

The data might not have column names. You can use header = None to tell read_csv() not to treat the first row as headings, and instead label them sequentially from 1 to n:
```
data_nonames = StringIO("""1,2,3\n4,5,6""")
pd.read_csv(data_nonames, header = None)
#>    0  1  2
#> 0  1  2  3
#> 1  4  5  6
```
("\n" is a convenient shortcut for adding a new line. You’ll learn more about it and other types of string escape in string basics.)

Alternatively you can pass names an list of names which will be used as the column names:
```
data_nonames = StringIO("""1,2,3\n4,5,6""")
pd.read_csv(data_nonames, names = ["x", "y", "z"], header = None)
#>    x  y  z
#> 0  1  2  3
#> 1  4  5  6
```

Another option that commonly needs tweaking is na: this specifies the value (or values) that are used to represent missing values in your file:

data_missing = StringIO("""a,b,c\n1,2,.""")
pd.read_csv(data_missing, na_values = ".")
#>    a  b   c
#> 0  1  2 NaN

This is all you need to know to read ~75% of CSV files that you’ll encounter in practice. You can also easily adapt what you’ve learned to read fixed width files with pd.read_fwf(). To read in more challenging files, you’ll need to learn more about how pandas parses each column, turning them into DataFrame objects.

11.2.1 Exercises

What function would you use to read a file where fields were separated with
“|”?
Apart from hte file path, skiprows, and comment, what other arguments does pd.read_csv() have?
What are the most important arguments to pd.read_fwf()?
Sometimes strings in a CSV file contain commas. To prevent them from causing problems they need to be surrounded by a quoting character, like " or '. By default, pd.read_csv() assumes that the quoting character will be ". What argument to pd.read_csv() do you need to specify to read the following text into a data frame?
```
"x,y\n1,'a,b'"
```

11.3 Parsing a vector

Before we get into the details of how pandas reads files from disk, we need to take a little detour to talk about the astype() function. This function casts a pandas object to a more specialised data type like a logical or integer. The pandas package has a special .to_datetime() function to convert dates:

pd.Series(["TRUE", "FALSE", np.nan]).astype('bool')
#> 0    True
#> 1    True
#> 2    True
#> dtype: bool
pd.Series(["TRUE", "FALSE", "FALSE"]).astype('bool')
#> 0    True
#> 1    True
#> 2    True
#> dtype: bool
pd.Series(["1", "2", "3"]).astype("int")
#> 0    1
#> 1    2
#> 2    3
#> dtype: int64
pd.to_datetime(pd.Series(['2010-01-01', '1970-10-14']), infer_datetime_format=True)
#> 0   2010-01-01
#> 1   1970-10-14
#> dtype: datetime64[ns]

These functions are useful in their own right, but are also an important building block for pandas. Once you’ve learned how the individual parsers work in this section, we’ll circle back and see how they fit together to parse a complete file in the next section.

Using arsers is mostly a matter of understanding what’s available and how they deal with different types of input. There are eight particularly important dtypes that you can use within astype():

bool, int, and float parse logicals integers, and floating point numbers respectively. The bool dtype has trouble with nan values and may not perform as expected.
object handles string and mixed values.
category create factors, the data structure that pandas uses to represent categorical variables with fixed and known values.
to_datetime() allows you to parse various date & time specifications. These are the most complicated because there are so many different ways of writing dates.

to_numeric() allows you to coerce mixed value columns to numeric using errors = coerce

pd.to_numeric(pd.Series(["bb", "1", "True", "9", np.nan]), errors = 'coerce')
#> 0    NaN
#> 1    1.0
#> 2    NaN
#> 3    9.0
#> 4    NaN
#> dtype: float64

The following sections describe these parsers in more detail.

11.3.1 Numbers

It seems like it should be straightforward to parse a number, but three problems make it tricky:

People write numbers differently in different parts of the world. For example, some countries use . in between the integer and fractional parts of a real number, while others use ,.
Numbers are often surrounded by other characters that provide some context, like “$1000” or “10%”.
Numbers often contain “grouping” characters to make them easier to read, like “1,000,000”, and these grouping characters vary around the world.

You will need to use regex expressions to remove the unwanted characters before converting the value to a float.

11.3.2 Strings

It seems like strings should be really simple. Unfortunately life isn’t so simple, as there are multiple ways to represent the same string. To understand what’s going on, we need to dive into the details of how computers represent strings. In Python, we can get at the underlying representation of a string using hex() on a binary object:

b'Hathaway'.hex()
# and convert it back
#> '4861746861776179'
bytes.fromhex('4861746861776179').decode()
#> 'Hathaway'

Each hexadecimal number represents a byte of information: 48 is H, 61 is a, and so on. The mapping from hexadecimal number to character is called the encoding, and in this case the encoding is called ASCII. ASCII does a great job of representing English characters, because it’s the American Standard Code for Information Interchange.

Things get more complicated for languages other than English. In the early days of computing there were many competing standards for encoding non-English characters, and to correctly interpret a string you needed to know both the values and the encoding. For example, two common encodings are Latin1 (aka ISO-8859-1, used for Western European languages) and Latin2 (aka ISO-8859-2, used for Eastern European languages). In Latin1, the byte b1 is “±”, but in Latin2, it’s “ą”! Fortunately, today there is one standard that is supported almost everywhere: UTF-8. UTF-8 can encode just about every character used by humans today, as well as many extra symbols (like emoji!).

Python often defaults to UTF-8: pandas assumes your data is UTF-8 encoded when you read it, and always uses it when writing. This is a good default, but will fail for data produced by older systems that don’t understand UTF-8. If this happens to you, your strings will look weird when you print them. Sometimes just one or two characters might be messed up; other times you’ll get complete gibberish. For example:

Encodings are a rich and complex topic, and I’ve only scratched the surface here. If you’d like to learn more I’d recommend reading the detailed explanation at http://kunststube.net/encoding/.

11.3.3 Category (Factors)

Pandas uses category to represent categorical variables that have a known set of possible values. Give pd.Categorical() a Series of known levels to convert other unexpected value that are present to nan:

fruit =  ["apple", "banana"]
pd.Categorical(["apple", "banana", "bananana"], categories = fruit)
#> ['apple', 'banana', NaN]
#> Categories (2, object): ['apple', 'banana']

But if you have many problematic entries, it’s often easier to leave as character vectors and then use the tools you’ll learn about in strings and factors to clean them up.

11.3.4 Dates and times

Pandas provides an extensive set of capabilities for working with time using NumPy’s datetime64 and timedelta64 dtypes. You primarily use pd.to_datetime() whether you want a date or a date and time. When called without any additional arguments:

pd.to_datetime() expects an ISO8601 date-time. ISO8601 is an international standard in which the components of a date are organised from biggest to smallest: year, month, day, hour, minute, second.
```
pd.to_datetime(["2010-10-01T2010"])
# If time is omitted, it will be set to midnight and only print date.
#> DatetimeIndex(['2010-10-01 20:10:00'], dtype='datetime64[ns]', freq=None)
pd.to_datetime(["2010-10-01"])
#> DatetimeIndex(['2010-10-01'], dtype='datetime64[ns]', freq=None)
```
This is the most important date/time standard, and if you work with dates and times frequently, I recommend reading https://en.wikipedia.org/wiki/ISO_8601

If these defaults don’t work for your data you can supply your own date-time format, built up of the following pieces:

Year: %Y (4 digits).; %y (2 digits); 00-69 -> 2000-2069, 70-99 -> 1970-1999.
Month: %m (2 digits).; %b (abbreviated name, like “Jan”).; %B (full name, “January”).
Day: %d (2 digits).; %e (optional leading space).
Time: %H 0-23 hour.; %I 0-12, must be used with %p.; %p AM/PM indicator.; %M minutes.; %S integer seconds.; %OS real seconds.; %Z Time zone (as name, e.g. America/Chicago). Beware of abbreviations: if you’re American, note that “EST” is a Canadian time zone that does not have daylight savings time. It is not Eastern Standard Time! We’ll come back to this in time zones.; %z (as offset from UTC, e.g. +0800).
Non-digits: %. skips one non-digit character.; %* skips any number of non-digits.

The best way to figure out the correct format is to create a few examples in a character vector, and test with one of the parsing functions. For example:

pd.to_datetime(["01/02/15"], format = "%m/%d/%y")
#> DatetimeIndex(['2015-01-02'], dtype='datetime64[ns]', freq=None)
pd.to_datetime(["01/02/15"], format = "%d/%m/%y")
#> DatetimeIndex(['2015-02-01'], dtype='datetime64[ns]', freq=None)
pd.to_datetime(["01/02/15"], format = "%y/%m/%d")
#> DatetimeIndex(['2001-02-15'], dtype='datetime64[ns]', freq=None)

The pd.to_datetime() function has a handy argument infer_datetime_format which can often allow you to avoid entring format strings.

pd.to_datetime(["01/02/15"], infer_datetime_format = True)
#> DatetimeIndex(['2015-01-02'], dtype='datetime64[ns]', freq=None)

11.3.5 Exercises

What are the most common encodings used in Europe? What are the most common encodings used in Asia? Do some googling to find out.

Generate the correct format string to parse each of the following dates and times:

d1 = ["January 1, 2010"]
d2 = ["2015-Mar-07"]
d3 = ["06-Jun-2017"]
d4 = ["August 19 (2015)", "July 1 (2015)"]
d5 = ["12/30/14"] # Dec 30, 2014

11.4 Parsing a file

Now that you’ve learned how to parse an Series, it’s time to return to the beginning and explore how pandas parses a file. There are two new things that you’ll learn about in this section:

How pandas automatically guesses the type of each column.
How to override the default specification.

11.4.1 Strategy

Pandas uses a heuristic to figure out the type of each column: at first it tries to convert all values to an integer. If there is an error then it moves to the next data type. The last data type is an object which a Series of strings. The heuristic tries each of the following types in the listed order, stopping when it finds a match:

integer: contains only numeric characters (and -).
float: contains only valid doubles (including numbers like 4.5e-5).
bool: contains only “F”, “T”, “FALSE”, “TRUE”, “False” or “True”.

In addition, you can use the parse_dates argument to to parse datetime columns during parsing. See the read_csv documentation for guidance.

If none of these rules apply, then the column will stay as a Series of strings.

When using pd.read_csv(), I highly recommend always supplying dtypes. This ensures that you have a consistent, reproducible, and fast data import script. If you rely on the default guesses and your data changes, pandas will continue to read it in.

If you’re having major parsing problems, sometimes it’s easier to just read into a character vector of lines with readlines().Then you can use the string parsing skills you’ll learn later to parse more exotic formats. You can read more about Python’s Input/Output understand readlines().

11.5 Writing to a file

pandas also comes with a useful function for writing data back to disk: to_csv(). to_csv() increase the chances of the output file being read back in correctly by:

Always encoding strings in UTF-8.
Saving dates and date-times in ISO8601 format so they are easily parsed elsewhere.

If you want to export a csv file to Excel, use to_excel().

The most important argument is path (the location to save it). You can also specify how missing values are written with na_rep and if the index is included in the export index = False.

df = pd.DataFrame({'name': ['Raphael', 'Donatello'],
                   'mask': ['red', 'purple'],
                   'weapon': ['sai', 'bo staff']})
df.to_csv("my_file.csv", index=False)

Note that the type information is lost when you save to csv:

This makes CSVs a little unreliable for caching interim results—you need to recreate the column specification every time you load in. We recommend the feather format:

The feather format implements a fast binary file format that can be shared across programming languages:

df = pd.DataFrame({'name': ['Raphael', 'Donatello'],
               'mask': ['red', 'purple'],
               'weapon': ['sai', 'bo staff']})
df.to_feather("my_file.ftr")

There is a feather package in R to share files quickly between the two tools.

library(feather)
write_feather(challenge, "challenge.feather")
read_feather("challenge.feather")

11.6 Other types of data

To get other types of data into Python, we recommend you review pandas IO tools.