Project 4 Workbook

Course DS 250

Author

Brother Clark

Project 4 WorkBook

Tutoring Lab Info

The data science lab is a resource you can use in person, online, and in Slack.

Data Import

1. Importing Data

Importing data is the first step in any data analysis task. It involves bringing datasets into your Python environment for further processing and analysis.

Code Snippet

url = "https://raw.githubusercontent.com/byuidatascience/data4dwellings/master/data-raw/dwellings_ml/dwellings_ml.csv"
dwellings_ml = pd.read_csv(url)

Install Correct Packages

1. pip3/pip install -m

  1. NumPy: Essential for numerical computations in Python, NumPy provides powerful array manipulation capabilities, enabling efficient handling of large datasets commonly encountered in machine learning tasks.

  2. Seaborn: Built on top of Matplotlib, Seaborn offers a high-level interface for creating visually appealing statistical plots, facilitating quick exploration and visualization of relationships within datasets.

  3. scikit-learn: As a versatile machine learning library, scikit-learn simplifies the implementation of various algorithms, streamlining the development and deployment of predictive models through its user-friendly API and extensive functionality.

Code Snippet
python/path/ pip -m  install pandas plotly numpy seaborn scikit-learn

Metric Libraries Used

2. Metrics Machine Learning

  1. train_test_split: This function is used for splitting datasets into training and testing sets, essential for evaluating machine learning models’ performance.

  2. GaussianNB: Gaussian Naive Bayes is a classification algorithm based on Bayes’ theorem, commonly used for simple classification tasks with continuous input variables.

  3. RandomForestClassifier: RandomForestClassifier is an ensemble learning method that constructs multiple decision trees during training and outputs the mode of the classes as the prediction for classification tasks.

  4. GradientBoostingClassifier: Gradient Boosting is a boosting ensemble learning technique that builds weak learners sequentially, focusing on the errors made by previous learners, thus improving predictive performance.

  5. DecisionTreeClassifier: Decision trees are a non-parametric supervised learning method used for classification and regression tasks, breaking down a dataset into smaller subsets based on different feature values.

  6. metrics: The metrics module in scikit-learn provides various evaluation metrics for assessing the performance of machine learning models, such as accuracy, precision, recall, F1 score, etc.

  7. tree: The tree module in scikit-learn provides tools for constructing, visualizing, and interpreting decision trees and other tree-based models.

Code Snippet
from sklearn.tree import DecisionTreeClassifier


classifier_DT = DecisionTreeClassifier(max_depth=11)


classifier_DT.fit(X_train, y_train)


y_pred = classifier_DT.predict(X_test)
Back to top