Project: Customer Analytics: Preparing Data for Modeling

A common problem when creating models to generate business value from data is that the datasets can be so large that it can take days for the model to generate predictions. Ensuring that your dataset is stored as efficiently as possible is crucial for allowing these models to run on a more reasonable timescale without having to reduce the size of the dataset.

You've been hired by a major online data science training provider called Training Data Ltd. to clean up one of their largest customer datasets. This dataset will eventually be used to predict whether their students are looking for a new job or not, information that they will then use to direct them to prospective recruiters.

You've been given access to customer_train.csv, which is a subset of their entire customer dataset, so you can create a proof-of-concept of a much more efficient storage solution. The dataset contains anonymized student information, and whether they were looking for a new job or not during training:

Column	Description
`student_id`	A unique ID for each student.
`city`	A code for the city the student lives in.
`city_development_index`	A scaled development index for the city.
`gender`	The student's gender.
`relevant_experience`	An indicator of the student's work relevant experience.
`enrolled_university`	The type of university course enrolled in (if any).
`education_level`	The student's education level.
`major_discipline`	The educational discipline of the student.
`experience`	The student's total work experience (in years).
`company_size`	The number of employees at the student's current employer.
`company_type`	The type of company employing the student.
`last_new_job`	The number of years between the student's current and previous jobs.
`training_hours`	The number of hours of training completed.
`job_change`	An indicator of whether the student is looking for a new job (`1`) or not (`0`).

# Import necessary libraries
import pandas as pd

# Load the dataset
ds_jobs = pd.read_csv("customer_train.csv")

# View the dataset
ds_jobs.head()

# Create a copy of ds_jobs for transforming
ds_jobs_transformed = ds_jobs.copy()

# Start coding here. Use as many cells as you like!
import pandas as pd

# Load dataset
ds_jobs = pd.read_csv("customer_train.csv")

# Record original memory usage & show info BEFORE transformations
orig_mem = ds_jobs.memory_usage(deep=True).sum()
print("=== BEFORE TRANSFORMATION ===")
ds_jobs.info(memory_usage="deep")
print("Memory usage (MB):", orig_mem / 1024**2, "\n")

# ----------------------------
# TYPE TRANSFORMATIONS
# ----------------------------

# 1. Booleans (binary columns)
ds_jobs["relevant_experience"] = ds_jobs["relevant_experience"].map({
    "Has relevant experience": True,
    "No relevant experience": False
}).astype("bool")

ds_jobs["job_change"] = ds_jobs["job_change"].astype("bool")

# 2. Integers
ds_jobs["student_id"] = ds_jobs["student_id"].astype("int32")
ds_jobs["training_hours"] = ds_jobs["training_hours"].astype("int32")

# 3. Floats
ds_jobs["city_development_index"] = ds_jobs["city_development_index"].astype("float16")

# 4. Nominal categoricals (gender kept as category)
nominal_cols = ["city", "major_discipline", "company_type", "gender"]
for col in nominal_cols:
    ds_jobs[col] = ds_jobs[col].astype("category")

# 5. Ordinal categoricals (ordered, not numeric)
edu_order = ["Primary School", "High School", "Graduate", "Masters", "Phd"]
exp_order = ["<1"] + [str(i) for i in range(1, 21)] + [">20"]
size_order = ["<10", "10-49", "50-99", "100-499", "500-999",
              "1000-4999", "5000-9999", "10000+"]
last_job_order = ["never", "0", "1", "2", "3", "4", ">4"]

ds_jobs["education_level"] = pd.Categorical(
    ds_jobs["education_level"],
    categories=edu_order,
    ordered=True
)

ds_jobs["experience"] = pd.Categorical(
    ds_jobs["experience"],
    categories=exp_order,
    ordered=True
)

ds_jobs["company_size"] = pd.Categorical(
    ds_jobs["company_size"],
    categories=size_order,
    ordered=True
)

ds_jobs["last_new_job"] = pd.Categorical(
    ds_jobs["last_new_job"],
    categories=last_job_order,
    ordered=True
)

# ----------------------------
# enrolled_university: normalize + ordered categorical
# ----------------------------

def normalize_enrolled(x):
    """Normalize common variants to canonical keys, preserve NaN."""
    if pd.isna(x):
        return x
    s = str(x).strip().lower()
    s = s.replace('-', ' ').replace('/', ' ').replace('__', '_')
    s = s.replace('  ', ' ').replace(' ', '_')
    # common variants -> canonical
    if s in ("no_enrollment", "no_enrolment", "no_enroll", "none", "not_enrolled"):
        return "no_enrollment"
    if "part" in s:
        return "part_time_course"
    if "full" in s:
        return "full_time_course"
    # fallback: return the cleaned token
    return s

# apply normalization
ds_jobs["enrolled_university"] = ds_jobs["enrolled_university"].map(normalize_enrolled)

# define natural order for enrollment commitment
enrolled_order = ["no_enrollment", "part_time_course", "full_time_course"]
enrolled_dtype = pd.CategoricalDtype(categories=enrolled_order, ordered=True)

# cast to ordered categorical (unknown / unexpected values -> NaN)
ds_jobs["enrolled_university"] = ds_jobs["enrolled_university"].astype(enrolled_dtype)

# ----------------------------
# FILTERING STEP
# ----------------------------

# Helper functions for filtering
def exp_to_num(x):
    if pd.isna(x):
        return None
    if x == "<1":
        return 0
    if x == ">20":
        return 21
    try:
        return int(x)
    except:
        return None

def size_to_min(x):
    if pd.isna(x):
        return None
    if x == "<10": return 0
    if x == "10-49": return 10
    if x == "50-99": return 50
    if x == "100-499": return 100
    if x == "500-999": return 500
    if x == "1000-4999": return 1000
    if x == "5000-9999": return 5000
    if x == "10000+": return 10000
    return None

exp_numeric = ds_jobs["experience"].map(exp_to_num)
size_numeric = ds_jobs["company_size"].map(size_to_min)

# Final filtered DataFrame (copy to avoid SettingWithCopy warnings)
ds_jobs_transformed = ds_jobs[(exp_numeric >= 10) & (size_numeric >= 1000)].copy()

# ----------------------------
# RESULTS / VERIFICATION
# ----------------------------

final_mem = ds_jobs_transformed.memory_usage(deep=True).sum()
print("=== AFTER TRANSFORMATION (FILTERED) ===")
ds_jobs_transformed.info(memory_usage="deep")
print("Memory usage (MB):", final_mem / 1024**2)

reduction_mb = (orig_mem - final_mem) / 1024**2
reduction_pct = (orig_mem - final_mem) / orig_mem * 100 if orig_mem else 0
print(f"\nMemory reduction: {reduction_mb:.2f} MB ({reduction_pct:.1f}%)")