Picture by Editor
# Introduction
In line with CrowdFlower’s survey, knowledge scientists spend 60% of their time organizing and cleansing the information.
On this article, we’ll stroll via constructing a knowledge cleansing pipeline utilizing a real-life dataset from DoorDash. It incorporates almost 200,000 meals supply information, every of which incorporates dozens of options comparable to supply time, whole gadgets, and retailer class (e.g., Mexican, Thai, or American delicacies).
# Predicting Meals Supply Occasions with DoorDash Knowledge
DoorDash goals to estimate the time it takes to ship meals precisely, from the second a buyer locations an order to the time it arrives at their door. In this knowledge challenge, we’re tasked with growing a mannequin that predicts the full supply length primarily based on historic supply knowledge.
Nevertheless, we gained’t do the entire challenge—i.e., we gained’t construct a predictive mannequin. As an alternative, we’ll use the dataset supplied within the challenge and create a knowledge cleansing pipeline.
Our workflow consists of two main steps.
# Knowledge Exploration
Let’s begin by loading and viewing the primary few rows of the dataset.
// Load and Preview the Dataset
import pandas as pd
df = pd.read_csv("historical_data.csv")
df.head()
Right here is the output.
This dataset consists of datetime columns that seize the order creation time and precise supply time, which can be utilized to calculate supply length. It additionally incorporates different options comparable to retailer class, whole merchandise depend, subtotal, and minimal merchandise value, making it appropriate for varied forms of knowledge evaluation. We are able to already see that there are some NaN values, which we’ll discover extra intently within the following step.
// Discover The Columns With data()
Let’s examine all column names with the data() methodology. We are going to use this methodology all through the article to see the modifications in column worth counts; it’s a very good indicator of lacking knowledge and total knowledge well being.
Right here is the output.
As you possibly can see, we now have 15 columns, however the variety of non-null values differs throughout them. This implies some columns comprise lacking values, which might have an effect on our evaluation if not dealt with correctly. One last item: the created_at and actual_delivery_time knowledge varieties are objects; these needs to be datetime.
# Constructing Knowledge Cleansing Pipeline
On this step, we construct a structured knowledge cleansing pipeline to organize the dataset for modeling. Every stage addresses frequent points comparable to timestamp formatting, lacking values, and irrelevant options.
// Fixing the Date and Time Columns Knowledge Sorts
Earlier than doing knowledge evaluation, we have to repair the columns that present the time. In any other case, the calculation that we talked about (actual_delivery_time – created_at) will go flawed.
What we’re fixing:
- created_at: when the order was positioned
- actual_delivery_time: when the meals arrived
These two columns are saved as objects, so to have the ability to do calculations appropriately, we now have to transform them to the datetime format. To try this, we will use datetime capabilities in pandas. Right here is the code.
import pandas as pd
df = pd.read_csv("historical_data.csv")
# Convert timestamp strings to datetime objects
df["created_at"] = pd.to_datetime(df["created_at"], errors="coerce")
df["actual_delivery_time"] = pd.to_datetime(df["actual_delivery_time"], errors="coerce")
df.data()
Right here is the output.
As you possibly can see from the screenshot above, the created_at and actual_delivery_time are datetime objects now.
Among the many key columns, store_primary_category has the fewest non-null values (192,668), which suggests it has essentially the most lacking knowledge. That’s why we’ll deal with cleansing it first.
// Knowledge Imputation With mode()
One of many messiest columns within the dataset, evident from its excessive variety of lacking values, is store_primary_category. It tells us what sort of meals shops can be found, like Mexican, American, and Thai. Nevertheless, many rows are lacking this data, which is an issue. For example, it may possibly restrict how we will group or analyze the information. So how can we repair it?
We are going to fill these rows as an alternative of dropping them. To try this, we are going to use smarter imputation.
We write a dictionary that maps every store_id to its most frequent class, after which use that mapping to fill in lacking values. Let’s see the dataset earlier than doing that.
Right here is the code.
import numpy as np
# International most-frequent class as a fallback
global_mode = df["store_primary_category"].mode().iloc[0]
# Construct store-level mapping to essentially the most frequent class (quick and sturdy)
store_mode = (
df.groupby("store_id")["store_primary_category"]
.agg(lambda s: s.mode().iloc[0] if not s.mode().empty else np.nan)
)
# Fill lacking classes utilizing the store-level mode, then fall again to world mode
df["store_primary_category"] = (
df["store_primary_category"]
.fillna(df["store_id"].map(store_mode))
.fillna(global_mode)
)
df.data()
Right here is the output.
As you possibly can see from the screenshot above, the store_primary_category column now has the next non-null depend. However let’s double-check with this code.
df["store_primary_category"].isna().sum()
Right here is the output exhibiting the variety of NaN values. It’s zero; we removed all of them.
And let’s see the dataset after the imputation.
// Dropping Remaining NaNs
Within the earlier step, we corrected the store_primary_category, however did you discover one thing? The non-null counts throughout the columns nonetheless don’t match!
This can be a clear signal that we’re nonetheless coping with lacking values in some a part of the dataset. Now, relating to knowledge cleansing, we now have two choices:
- Fill these lacking values
- Drop them
On condition that this dataset incorporates almost 200,000 rows, we will afford to lose some. With smaller datasets, you’d have to be extra cautious. In that case, it’s advisable to research every column, set up requirements (resolve how lacking values can be stuffed—utilizing the imply, median, most frequent worth, or domain-specific defaults), after which fill them.
To take away the NaNs, we are going to use the dropna() methodology from the pandas library. We’re setting inplace=True to use the modifications on to the DataFrame while not having to assign it once more. Let’s see the dataset at this level.
Right here is the code.
df.dropna(inplace=True)
df.data()
Right here is the output.
As you possibly can see from the screenshot above, every column now has the identical variety of non-null values.
Let’s see the dataset after all of the modifications.
// What Can You Do Subsequent?
Now that we now have a clear dataset, right here are some things you are able to do subsequent:
- Carry out EDA to grasp supply patterns.
- Engineer new options like supply hours or busy dashers ratio so as to add extra which means to your evaluation.
- Analyze correlations between variables to extend your mannequin’s efficiency.
- Construct completely different regression fashions and discover the best-performing mannequin.
- Predict the supply length with the best-performing mannequin.
# Remaining Ideas
On this article, we now have cleaned the real-life dataset from DoorDash by addressing frequent knowledge high quality points, comparable to fixing incorrect knowledge varieties and dealing with lacking values. We constructed a easy knowledge cleansing pipeline tailor-made to this knowledge challenge and explored potential subsequent steps.
Actual-world datasets may be messier than you assume, however there are additionally many strategies and methods to unravel these points. Thanks for studying!
Nate Rosidi is a knowledge scientist and in product technique. He is additionally an adjunct professor educating analytics, and is the founding father of StrataScratch, a platform serving to knowledge scientists put together for his or her interviews with actual interview questions from high firms. Nate writes on the newest traits within the profession market, offers interview recommendation, shares knowledge science initiatives, and covers every thing SQL.
