
Already know some of this? A 2-minute test skips the lessons you know.
Take the placement testA free 12-lesson course built from Keith Galli's 1h 34m of YouTube video. 1 day at 90 minutes a day, with study notes and a quiz in every lesson. A 2-minute placement test skips what you already know.
Taught by Keith Galli on YouTube. Videos play in YouTube's own player, so the creator keeps every view.
Create DataFrames manually, inspect structure using .head(), .info(), .describe(), .columns, .index, .shape, and .dtypes.
Load data using pd.read_csv, pd.read_parquet, and pd.read_excel; compare file formats for size and speed; load from local paths or raw GitHub URLs.
Filter rows using boolean indexing with .loc[] and bracket syntax; combine multiple numeric conditions with parentheses and & operator; select specific columns after filtering.
Filter string columns using .str.contains(), .str.startswith(), .str.endswith() with case sensitivity control; apply complex regex patterns for name/character matching; handle NaN with na=False.
Add scalar and conditional columns using np.where(); drop columns with .drop(columns=...); rename columns with .rename(); understand in-place vs. copy behavior.
Apply vectorized string methods via .str accessor; convert columns to datetime with pd.to_datetime() and extract components (.dt.year, .dt.month) for grouping.
Apply custom row-wise or element-wise functions using .apply() with lambda or named functions for complex column derivations.
Combine DataFrames with pd.merge() (inner, left, right, outer joins) on keys; stack DataFrames vertically with pd.concat(); manage suffixes and indexes.
Detect NaNs with .isna()/.notna(); fill with scalars, column statistics (.mean()), or .interpolate() for linear trends; drop rows with .dropna(subset=...); assign back or use inplace=True.
Group by datetime components (.dt.year, .dt.month) after conversion; sort and reset_index for clean tabular output; analyze trends like athlete birth distributions.
Compare periods with .shift(); rank values with .rank(ascending=False); compute running totals with .cumsum(); apply rolling window calculations with .rolling(window).sum().
Leverage PyArrow (Pandas 2.0) for faster string ops and memory efficiency; use GitHub Copilot and ChatGPT to generate Pandas code for ranking, pivoting, regex, and plotting.