CMPIF2100: Cliff Notes & Sandbox

These are my auto-generated cliff notes, produced in part with my CMPIF2100 Lab Transcriber. As per the permission given during the 5/28 study hall, I'm sharing them with the rest of the cohort. I'll do my best to keep this repo updated.

A letter from the note taker

Cliff Notes

The notes are organized by module:

• Canyon Notes/ (Interconnected, illustrated lessons; which are deeper and layered, enriched with worked examples, callouts, sidebars, and real graphs, but never padded that cover entire halves of a semester and projects)
• Cliff Jumper Notes/ (each module's notes woven into one connected lesson)
• Cliff Notes Module 01/ (course orientation, what data science is, the data-science pipeline, and list comprehensions (part 1))
• Cliff Notes Module 02/ (list comprehensions (part 2))
• Cliff Notes Module 03/ (the programming process and mindset: getting unstuck, problem-solving, and practical Python tips)
• Cliff Notes Module 04/ (statistics foundations: variance, standard deviation, standard error, confidence intervals, and random sampling)
• Cliff Notes Module 05/ (NumPy: 1D and 2D arrays, reshaping and transposing, conditional filtering, summary statistics, and random number generation)
• Cliff Notes Module 06/ (pandas basics: Series, DataFrames, selecting columns, and filtering rows)
• Cliff Notes Module 07/ (pandas wrangling: grouping and aggregation, concatenation, summarizing Series/DataFrames, and missing-value EDA)
• Cliff Notes Module 08/ (visualization foundations: matplotlib and seaborn, bar charts, histograms, boxplots, plotting from pandas, and choosing a plot type)
• Cliff Notes Module 09/ (relationship plots: scatterplots and point density, categorical point plots, faceted histograms, conditional KDE, and violin plots)
• Cliff Notes Module 10/ (many continuous variables at once: pair plots, correlation coefficients and heat maps, seaborn styles and color palettes, and standardization with scikit-learn)
• Cliff Notes Module 11/ (predictive analytics and linear regression: the goal of predictive modeling, models and regression vs classification, the linear regression formula and its coefficients, a river-depth code example, and the model's four assumptions)
• Cliff Notes Module 12/ (linear regression with statsmodels: fitting OLS models with the formula API, reading the summary table, p-values and confidence intervals on coefficient estimates, predicting over an input grid, and confidence vs prediction intervals on those predictions)
• Cliff Notes Module 13/ (richer inputs for linear regression: additive multi-predictor models, standardizing variables so coefficients compare fairly, categorical inputs via dummy coding and reference categories, interaction features and their prediction visualizations, and nonlinear feature transformations, all still linear in the coefficients)
• Cliff Notes Module 14/ (honest model evaluation: residuals and R-squared, root mean squared error, comparing models on training scores and why that rewards complexity, k-fold cross-validation, the patsy design-matrix bridge to scikit-learn, choosing the simplest model within one standard error, and pipelines that keep preprocessing inside the folds)

Trainers

Per-subject documents that focus on a single topic in depth, rather than following the course module by module. See the folder's README for details.

• Trainers/Module-2_Python_Loops_Trainer.pdf (a focused trainer on Python loops)
• Trainers/Module-4_Statistics_Simulation_Trainer.pdf (a focused trainer on statistics and simulation)
• Trainers/Module-8-9_Data_Visualization_Toolkit_Trainer_Part_1.pdf (the matplotlib, pandas, and seaborn functions behind the Module 8-9 plot types; Part 1 of 2)

Guides

Standalone walkthroughs and setup guides. Each linked folder is its own repository, gathered here for convenience:

• Guides/Goodbye AI - On the Road with VS Code.pdf (going AI-free in VS Code, the follow-up to the Coursera-to-VS-Code guide)
• Guides/UCFCECS_ALL_PYTHON_LOOPS_INTRO/ (an intro to every kind of Python loop, from the UCF Assisted Learning Center)
• Guides/upitt-anaconda-setup-guide/ (setting up Anaconda for CMPINF-2100 at the University of Pittsburgh)
• Guides/CourseraToVSCode/ (a guide to using VS Code with Coursera, the prequel to the Goodbye AI guide above)

Tools

Small utilities I built for the cohort. Each linked folder is its own repository, gathered here for convenience:

• Tools/CMPIF2100_LAB_VER_CHECK/ (two lines of script that save your sanity checking package versions)
• Tools/CMPIF2100-Lab-Transcriber-2.0/ (records Windows system audio and transcribes lab sessions live)
• Tools/VEShell/ (Very Easy Shell: an Electron + xterm.js + node-pty terminal wrapper)

A note on these folders: the items under Guides/ and Tools/ are Git submodules, they point to their own repositories so nothing is duplicated here. Browsing on GitHub just works: click a folder and it takes you to that repo. If you git clone this repo and want those folders filled in on your machine, add --recursive (or run git submodule update --init --recursive afterward).

Sandbox

Also included is Sandbox.ipynb, an entirely blank-slate Jupyter notebook I use as "scratch paper" while working through assignments. Feel free to use it the same way.

Contributors

• DatJavaClass (Victor S), author and director. Conceived these notes, established their format and structure, directed their creation, and fact-checked, edited, and quality-controlled every one, with assistance by Claude. Some material may have been derived from assigned material, but has not been copied verbatim. For source materials please contact CMPINF-2100 Faculty and Assistants.

As Always! You're awesome, Stay awesome! and I wish everyone the best of grades!

Can you find the passive aggressive semi-colon?