Course tools and Python foundations

Open in Colab

This notebook introduces the small set of Python tools used throughout the course. It is not a comprehensive Python tutorial: the goal is to be comfortable loading, inspecting, selecting, filtering, and plotting engineering data.

Learning objectives

1. Import the Python libraries

The next cell demonstrates how to import a Python library.

import numpy as np makes the NumPy library available in this notebook. The as keyword assigns the short alias np to NumPy’s namespace, so we can write np.array(...) instead of the longer numpy.array(...). Using np is a common convention in scientific Python.

# Import NumPy first: it is the numerical-array tool used in this section.
import numpy as np

2. Numerical arrays

A NumPy array stores a collection of numerical values. Array operations act element-by-element, which makes calculations compact and readable.

# Store repeated temperature measurements in a one-dimensional numerical array.
temperatures_c = np.array([15.2, 18.7, 21.4, 24.1])
# NumPy applies this unit conversion to every element of the array at once.
temperatures_k = temperatures_c + 273.15

print("Temperatures in K:", temperatures_k)
print(f"Mean temperature: {temperatures_c.mean():.1f} °C")
print(f"Range: {temperatures_c.min():.1f}–{temperatures_c.max():.1f} °C")
Temperatures in K: [288.35 291.85 294.55 297.25]
Mean temperature: 19.9 °C
Range: 15.2–24.1 °C

The array is useful for a single quantity observed repeatedly. A table with many named variables is better represented by a Pandas DataFrame.

Compared with a native Python list, a NumPy array is designed for numerical work. It supports vectorized calculations—such as adding 273.15 to every temperature—in one clear operation, stores values efficiently, and is much faster for the large arrays used in data analysis and machine learning.

3. Load the course dataset

The Gas Turbine CO and NOx Emissions dataset records the operation of a gas turbine between 2011 and 2015. It combines ambient conditions, turbine sensor measurements, energy yield, and two emissions outcomes: carbon monoxide (CO) and nitrogen oxides (NOx).

Each row represents one hourly observation of the system: the measured operating and environmental conditions during that hour, together with the emissions observed at that time. This course returns to the same dataset repeatedly so that the modeling choices remain connected to a physical system.

The original data and documentation are available from the UCI Machine Learning Repository.

# Pandas represents a table with named columns as a DataFrame.
import pandas as pd
from pathlib import Path
from urllib.request import urlretrieve

# Use the committed course version so everyone analyzes the same data.
DATA_URL = (
    "https://raw.githubusercontent.com/changyaochen/MECE4520/master/"
    "site/data/gas-turbine-course.csv"
)
# Prefer a local copy; otherwise download it when this notebook runs independently.
data_path = next(
    (path for path in [
        Path("../data/gas-turbine-course.csv"),
        Path("site/data/gas-turbine-course.csv"),
        Path("gas-turbine-course.csv")]
        if path.exists()
    ),
    Path("gas-turbine-course.csv"),
)
if not data_path.exists():
    urlretrieve(DATA_URL, data_path)
data = pd.read_csv(data_path)

# Inspect the dataset size before selecting variables or fitting a model.
print(f"Rows: {len(data):,}")
print(f"Columns: {data.shape[1]}")
data.head()
Rows: 36,733
Columns: 12
campaign_year AT AP AH AFDP GTEP TIT TAT TEY CDP CO NOX
0 2011 4.5878 1018.7 83.675 3.5758 23.979 1086.2 549.83 134.67 11.898 0.32663 81.952
1 2011 4.2932 1018.3 84.235 3.5709 23.951 1086.1 550.05 134.67 11.892 0.44784 82.377
2 2011 3.9045 1018.4 84.858 3.5828 23.990 1086.5 550.19 135.10 12.042 0.45144 83.776
3 2011 3.7436 1018.3 85.434 3.5808 23.911 1086.5 550.17 135.03 11.990 0.23107 82.505
4 2011 3.7516 1017.8 85.182 3.5781 23.917 1085.9 550.00 134.67 11.910 0.26747 82.028

A good first habit is to inspect the column names and units before modeling. This data dictionary identifies every column in the course dataset.

Variable Meaning
AT Ambient temperature (°C)
AP Ambient pressure (mbar)
AH Ambient humidity (%)
AFDP Air filter differential pressure (mbar)
GTEP Gas turbine exhaust pressure (mbar)
TIT Turbine inlet temperature (°C)
TAT Turbine after temperature (°C)
TEY Turbine energy yield (MWh)
CDP Compressor discharge pressure (mbar)
CO Carbon monoxide emissions (mg/m³)
NOX Nitrogen oxides emissions (mg/m³)

4. Select and filter data

Data analysis often begins with two simple operations: selecting the columns (variables) relevant to a question and filtering the rows (observations) that meet a condition. The next cell demonstrates both operations on the gas-turbine data.

# Select a small set of ambient, turbine, and emissions variables for inspection.
selected = data[["AT", "TIT", "TEY", "CO", "NOX"]]
# Use the median to divide the observations into relatively warmer and cooler conditions.
warm_conditions = selected[selected["AT"] > selected["AT"].median()]

print("Selected variables:")
display(selected.head())
print(f"Rows above the median ambient temperature: {len(warm_conditions):,}")
warm_conditions.describe().loc[["mean", "std", "min", "max"]]
Selected variables:
AT TIT TEY CO NOX
0 4.5878 1086.2 134.67 0.32663 81.952
1 4.2932 1086.1 134.67 0.44784 82.377
2 3.9045 1086.5 135.10 0.45144 83.776
3 3.7436 1086.5 135.03 0.23107 82.505
4 3.7516 1085.9 134.67 0.26747 82.028 
Rows above the median ambient temperature: 18,365
AT TIT TEY CO NOX
mean 24.005367 1083.897964 132.012501 2.057256 59.22925
std 3.905948 15.765051 13.316517 1.602460 7.97456
min 17.802000 1024.800000 100.040000 0.001594 25.90500
max 37.103000 1100.900000 160.370000 37.746000 119.91000

5. Make a first plot

A scatter plot helps us inspect whether two variables move together. Each point below is one hourly observation. We draw only a random subset so the plot remains readable.

# Import Matplotlib only when we are ready to make a figure.
import matplotlib.pyplot as plt

# Use a readable plotting style for the exploratory figure below.
plt.style.use("seaborn-v0_8-whitegrid")

# Plot a reproducible subset so overlapping points do not hide the overall pattern.
# This is useful for large datasets.
plot_data = data.sample(n=2_000, random_state=4520)

# Each point represents one hourly turbine observation.
fig, ax = plt.subplots(figsize=(7, 4.5))
ax.scatter(plot_data["TIT"], plot_data["NOX"], alpha=0.25, s=12)
ax.set(
    xlabel="Turbine inlet temperature, TIT (°C)",
    ylabel="NOx emissions (mg/m³)",
    title="A first view of turbine inlet temperature and NOx emissions",
)
plt.show()

Check-in

  1. Change the plotting code to use AT or TEY on the horizontal axis. What pattern do you see?
  2. What is one question you would ask before interpreting this plot as a causal relationship?
  3. In one sentence, explain the difference between a NumPy array and a Pandas DataFrame.

Keep your answers in your own notes. We will build on this dataset in the next notebook.

Return to the Foundations page