Pandas unique() Function: A Comprehensive Guide
Pandas is a widely used library for data manipulation in Python. It provides fast, efficient, and user-friendly data structures that seamlessly integrate with various file formats such as CSV, Excel, and SQL.
One of the most useful functions in Pandas is unique(). This function returns an array of unique values present in the input data.
Importance of the unique() function and its advantages over NumPy.unique():
The unique() function in Pandas is faster than NumPy’s unique() function. This is because it is optimized for Pandas DataFrames and Series. Pandas can handle data structures such as categorical data, string data, and mixed data types efficiently. In contrast, NumPy’s unique() function can only handle numerical data. Additionally, in terms of performance, Pandas’ unique() function is superior to NumPy’s unique() function when dealing with larger datasets.
Syntax of the unique() function and its input parameter:
The syntax of the Pandas unique() function is straightforward. We can call it on any Pandas object, such as a DataFrame, Series, or Index.
The syntax is as follows:
pandas.unique(values, sort=False)Here, ‘values’ represent the input data, and “sort” is an optional boolean parameter that specifies whether to return a sorted result or not. By default, sort is set to False.
Examples of using the unique() function with Index, Array, Series, Categorical, and DataFrame input:
1. Index:
To get unique values from a Pandas Index object, we can use the unique() method.
import pandas as pd
import numpy as np
idx = pd.Index([1, 1, 2, 2, 3, 3, 3, np.nan])
print(idx.unique())
Output:
Float64Index([1.0, 2.0, 3.0, nan], dtype='float64')2. Array:
We can use the unique() method with NumPy arrays.
arr = np.array([1, 1, 2, 2, 3, 3, 3, np.nan])
print(pd.unique(arr))
Output:
array([ 1., 2., 3., nan])3. Series:
We can use the unique() method with Series objects.
s = pd.Series([1, 2, 3, 1, 2, 3, 4, 5], name="my_series")
print(s.unique())
Output:
array([1, 2, 3, 4, 5])4. Categorical:
We can use the unique() method with categorical data.
cat_data = pd.Categorical(['b', 'a', 'c', 'a', 'b', 'd'])
print(pd.unique(cat_data))
Output:
['b', 'a', 'c', 'd']
Categories (4, object): ['b', 'a', 'c', 'd']
5. DataFrame:
We can use the unique() method with Pandas DataFrames.
df = pd.DataFrame({'A': [1, 2, 3, 1, 2, 3], 'B': ['a', 'b', 'c', 'd', 'e', 'f']})
print(df.B.unique())
Output:
array(['a', 'b', 'c', 'd', 'e', 'f'], dtype=object)Conclusion:
In this article, we explored the syntax and implementation of the Pandas unique() function. We also compared it with NumPy’s unique() function and explained its importance.
The unique() function is extremely valuable when we need to extract unique or distinct values from our dataset. It is efficient, fast, and user-friendly.
We hope that this article has helped you understand the unique() function in depth and apply it to your datasets.
Examples of Using Pandas unique() function:
In the previous section, we introduced the use of the Pandas unique() function and how it differs from NumPy’s unique().
In this section, we will go over some examples in detail, showing how to use the function with different input types.
Example 1: Using Index as input
In this example, we will retrieve unique values from a Pandas Index object.
import pandas as pd
import numpy as np
idx = pd.Index([1, 2, 3, 1, 2, 3])
unique_values = pd.unique(idx)
print(unique_values)
Output:
[1 2 3]
Here, we have created a Pandas Index object with duplicate values and then called the unique() method, which will return only the unique values.
Example 2: Using Array and Series as input
In this example, we will retrieve unique values from a NumPy array and a Pandas Series object.
import pandas as pd
import numpy as np
# array input
arr = np.array([1, 2, 2, 3, 3, 3])
unique_values_arr = pd.unique(arr)
# series input
s = pd.Series([1, 2, 2, 3, 3, 3])
unique_values_series = s.unique()
print(unique_values_arr)
print(unique_values_series)
Output:
[1 2 3]
[1 2 3]
In this example, we have used two different input types, an array and a Series object, to retrieve unique values. Both cases return the same output, and the function works seamlessly with both types.
Example 3: Using Categorical input
In this example, we will retrieve unique values from a Pandas Categorical data type.
import pandas as pd
# Categorical input
cat_data = pd.Categorical(['a', 'a', 'b', 'b', 'c'])
unique_values = pd.unique(cat_data)
print(unique_values)
Output:
['a', 'b', 'c']
Categories (3, object): ['a', 'b', 'c']
Here, we have created a Pandas Categorical object with multiple duplicates and then called the unique() method, which will return only the unique categories.
Example 4: Using DataFrame input
In this example, we will retrieve unique values from a Pandas DataFrame.
import pandas as pd
# DataFrame input
data = {'name': ['Nick', 'Mike', 'Sarah', 'Mike'], 'age': [30, 32, 25, 32]}
df = pd.DataFrame(data)
unique_values = df['name'].unique()
print(unique_values)
Output:
['Nick' 'Mike' 'Sarah']
Here, we have created a Pandas DataFrame with duplicate values in one of the columns, ‘name,’ and then called the unique() method on that column. The method returns only the unique values from that column.
Conclusion and Summary:
The Pandas unique() function is a helpful tool for data manipulation in Python. It allows us to retrieve only the unique values from different types of data structures, such as Index, arrays, Series, Categorical, and DataFrames.
Unique values provide an essential aspect of data analysis, and the function makes this process more manageable. In the examples above, we have demonstrated how to use the function with different input types.
However, this is by no means an exhaustive list, and there are many other use cases for the Pandas unique() function. Overall, the function is easy to use and efficient, making it a valuable tool for any data analyst or scientist.
In summary, the Pandas unique() function is a powerful tool that allows us to retrieve only the unique values from different types of data structures, such as Index, arrays, Series, Categorical, and DataFrames. It is faster and more efficient than NumPy’s unique() function and is optimized for Pandas data structures.
The function is essential for data analysis and manipulation and is easy to use. The examples provided in this article show how to use the function with different input types.
As data analysis becomes increasingly critical, the Pandas unique() function proves to be a valuable tool for data scientists and analysts.