Skip to content

PIPE_OPERATOR_QUICKREF.md

Latest commit

 

History

History
112 lines (88 loc) · 2.37 KB

PIPE_OPERATOR_QUICKREF.md

File metadata and controls

112 lines (88 loc) · 2.37 KB

Pipe Operator Quick Reference

✅ What Works

1. DataFrame to Plot

df >> Plot() >> plot_points('x', 'y') >> show()

2. Plot to Verb Functions

Plot(df) >> plot_points('x', 'y') >> add_color('category') >> show()

3. Pandas Methods Then Plot

df.filter(...).sort_values(...) >> Plot() >> plot_points('x', 'y') >> show()

4. Method Chaining (Always Works)

Plot(df).plot_points('x', 'y').add_color('category').show()

❌ What Doesn't Work

DataFrame with Lambda Functions

# WRONG - pandas doesn't support this
df >> (lambda d: d[d['x'] > 0]) >> Plot()  # TypeError!

Why: Pandas DataFrames don't natively support the >> operator with functions.

🔧 Solutions

Option 1: Use Pandas Methods First (Recommended)

# CORRECT
filtered_df = df[df['x'] > 0].sort_values('y')
plot = filtered_df >> Plot() >> plot_points('x', 'y') >> show()

# Or in one chain:
plot = (df[df['x'] > 0]
    .sort_values('y')
    >> Plot()
    >> plot_points('x', 'y')
    >> show())

Option 2: Use PipeFrame Package

pip install pipeframe
from pipeframe import DataFrame as pf_df

plot = (pf_df(df)
    >> (lambda d: d[d._df['x'] > 0])  # Now lambdas work!
    >> (lambda d: d.sort_values('y'))
    >> Plot()
    >> plot_points('x', 'y')
    >> show())

Option 3: Separate Data & Viz

# Traditional approach
filtered = df[df['x'] > 0]
plot = Plot(filtered).plot_points('x', 'y').show()

📝 Key Imports

from pipeplotly import Plot
from pipeplotly.verbs import (
    plot_points, plot_lines, plot_bars,
    add_color, add_size, add_labels,
    set_theme, show, save
)

🎯 Best Practices

  1. For simple viz: Use method chaining (.)
  2. For data + viz: Use pandas methods then pipe to Plot
  3. For complex pipelines: Use pipeframe package
  4. Restart kernel: After code changes to pipeplotly source

Examples

Simple Plot

df >> Plot() >> plot_points('x', 'y') >> show()

With Styling

df >> Plot() >> plot_points('x', 'y') >> add_color('cat', palette='viridis') >> set_theme('dark') >> show()

Data Transformation

df[df['x'] > 10].sort_values('y') >> Plot() >> plot_lines('x', 'y') >> show()

Interactive

df >> Plot() >> plot_points('x', 'y') >> to_interactive() >> show()