# pandas Quick Reference

*DataFrames, selection, aggregation, merging, and more*

> Source: pandas Documentation (pandas.pydata.org) · MIT

## DataFrames

### Creating DataFrames

```
import pandas as pd
df = pd.DataFrame({
    "name": ["Alice", "Bob", "Carol"],
    "age": [25, 30, 35],
    "score": [88, 92, 79]
})
```

### Inspection

| Command | Description |
|---------|-------------|
| `df.head(n)` | First n rows (default 5) |
| `df.tail(n)` | Last n rows |
| `df.shape` | Tuple of (rows, columns) |
| `df.dtypes` | Data type of each column |
| `df.info()` | Column types, non-null counts |
| `df.describe()` | Statistics for numeric columns |
| `df.columns` | Column names as Index |
| `df.index` | Row labels |

## Reading Data

### Common Readers

```
df = pd.read_csv("data.csv")
df = pd.read_excel("data.xlsx")
df = pd.read_json("data.json")
df = pd.read_sql(query, connection)
```

### Writing Data

```
df.to_csv("out.csv", index=False)
df.to_excel("out.xlsx", index=False)
df.to_json("out.json", orient="records")
```

### Read Options

| Command | Description |
|---------|-------------|
| `sep=";"` | Custom delimiter |
| `header=None` | No header row in file |
| `usecols=[0,2]` | Read specific columns only |
| `nrows=100` | Read first 100 rows |
| `na_values=["N/A"]` | Treat as NaN |

## Selection

### Columns

```
df["name"]              # single column (Series)
df[["name", "age"]]     # multiple columns (DataFrame)
df.name                 # attribute access (simple names)
```

### Rows with loc / iloc

```
df.loc[0]              # row by label
df.loc[0:2, "name"]    # rows 0-2, column "name"
df.iloc[0]             # row by position
df.iloc[0:2, 0:2]     # first 2 rows, 2 cols
```

### loc vs iloc

| Command | Description |
|---------|-------------|
| `df.loc[row, col]` | Select by **label** (inclusive end) |
| `df.iloc[row, col]` | Select by **position** (exclusive end) |
| `df.at[row, col]` | Fast scalar access by label |
| `df.iat[row, col]` | Fast scalar access by position |

## Filtering

### Boolean Filtering

```
df[df["age"] > 25]
df[df["name"].str.contains("li")]
df[(df["age"] > 25) & (df["score"] > 80)]
df[df["name"].isin(["Alice", "Bob"])]
```

### Handling Missing Data

```
df.isna().sum()          # NaN count per column
df.dropna()              # drop rows with any NaN
df.fillna(0)             # fill NaN with 0
df["col"].fillna(df["col"].mean())
```

### Sorting

```
df.sort_values("age")                # ascending
df.sort_values("age", ascending=False)
df.sort_values(["age", "score"])      # multi
```

## Aggregation

### Common Aggregations

| Command | Description |
|---------|-------------|
| `df["col"].sum()` | Sum of column |
| `df["col"].mean()` | Mean |
| `df["col"].median()` | Median |
| `df["col"].std()` | Standard deviation |
| `df["col"].min() / .max()` | Min / max |
| `df["col"].count()` | Non-null count |
| `df["col"].nunique()` | Number of unique values |
| `df["col"].value_counts()` | Frequency of each value |

### Multiple Aggregations

```
df.agg({"age": "mean", "score": ["min", "max"]})
df.describe()  # summary stats for all numeric
```

## GroupBy

### Basic Grouping

```
df.groupby("dept")["salary"].mean()
df.groupby("dept").agg(
    avg_sal=("salary", "mean"),
    count=("salary", "count")
)
```

### Multiple Groups

```
df.groupby(["dept", "year"])["sales"].sum()
df.groupby("dept").size()  # rows per group
```

### Transform & Apply

```
df["z_score"] = df.groupby("dept")["salary"] \
    .transform(lambda x: (x - x.mean()) / x.std())
df.groupby("dept").apply(lambda g: g.nlargest(3, "salary"))
```

## Merging

### Merge (SQL-style Join)

```
pd.merge(df1, df2, on="id")          # inner
pd.merge(df1, df2, on="id", how="left")
pd.merge(df1, df2, left_on="uid",
         right_on="user_id")
```

### Join Types

| Command | Description |
|---------|-------------|
| `how="inner"` | Keep only matching rows (default) |
| `how="left"` | Keep all left rows, NaN for no match |
| `how="right"` | Keep all right rows |
| `how="outer"` | Keep all rows from both sides |

### Concatenation

```
pd.concat([df1, df2])             # stack rows
pd.concat([df1, df2], axis=1)     # side by side
pd.concat([df1, df2], ignore_index=True)
```

## Pivot Tables

### Pivot Table

```
df.pivot_table(
    values="sales", index="region",
    columns="quarter", aggfunc="sum"
)
```

### Reshaping

```
df.melt(id_vars=["name"],
        value_vars=["q1", "q2"],
        var_name="quarter", value_name="sales")
```

### Cross Tabulation

```
pd.crosstab(df["dept"], df["gender"])
pd.crosstab(df["dept"], df["gender"],
            normalize="index")  # row percentages
```

## Time Series

### DateTime Basics

```
df["date"] = pd.to_datetime(df["date"])
df["year"] = df["date"].dt.year
df["month"] = df["date"].dt.month
df["weekday"] = df["date"].dt.day_name()
```

### Date Ranges & Resampling

```
pd.date_range("2025-01-01", periods=12, freq="ME")
df.set_index("date").resample("ME")["sales"].sum()
```

*freq="ME" requires pandas ≥ 2.2. Use "M" on older versions.*

### Accessor Attributes

| Command | Description |
|---------|-------------|
| `.dt.year / .dt.month / .dt.day` | Extract date components |
| `.dt.hour / .dt.minute` | Extract time components |
| `.dt.day_name()` | Weekday name (Monday, etc.) |
| `.dt.days_in_month` | Days in that month |

## Common Patterns

### Rename Columns

```
df.rename(columns={"old": "new"})
df.columns = ["a", "b", "c"]  # replace all
```

### Add / Modify Columns

```
df["total"] = df["q1"] + df["q2"]
df["grade"] = df["score"].apply(
    lambda x: "A" if x >= 90 else "B"
)
```

### Drop Columns / Rows

```
df.drop(columns=["temp"])
df.drop_duplicates(subset=["name"])
df.reset_index(drop=True)
```

### String Operations

```
df["name"].str.lower()
df["name"].str.contains("ali", case=False)
df["name"].str.split(" ").str[0]  # first name
```