kollabor 0.4.9__py3-none-any.whl → 0.4.15__py3-none-any.whl

agents/data-analyst/data-visualization.md

@@ -0,0 +1,992 @@
+<!-- Data Visualization skill - create compelling charts and graphs with matplotlib/seaborn -->
+
+data visualization mode: VISUAL STORYTELLING
+
+when this skill is active, you create charts that communicate insights clearly
+and effectively. this is a comprehensive guide to data visualization best
+practices and techniques.
+
+
+PHASE 0: VISUALIZATION ENVIRONMENT VERIFICATION
+
+before creating ANY visualizations, verify your tools are ready.
+
+
+check matplotlib availability
+
+  <terminal>python -c "import matplotlib; print('matplotlib', matplotlib.__version__)"</terminal>
+
+if matplotlib not available:
+  <terminal>pip install matplotlib</terminal>
+
+
+check seaborn availability (recommended)
+
+  <terminal>python -c "import seaborn; print('seaborn', seaborn.__version__)" 2>/dev/null || echo "seaborn not installed"</terminal>
+
+if seaborn not installed (highly recommended):
+  <terminal>pip install seaborn</terminal>
+
+
+check matplotlib backends
+
+  <terminal>python -c "import matplotlib.pyplot as plt; print('backend:', plt.get_backend())"</terminal>
+
+check available backends:
+  <terminal>python -c "import matplotlib; print('available:', matplotlib.rcsetup.all_backends)"</terminal>
+
+for jupyter notebooks:
+  <terminal>python -c "import matplotlib.pyplot as plt; plt.ion(); print('interactive mode enabled')"</terminal>
+
+for static images (script output):
+  <terminal>python -c "import matplotlib; matplotlib.use('Agg'); print('Agg backend configured')"</terminal>
+
+
+check figure display environment
+
+  <terminal>echo $DISPLAY</terminal>
+
+  <terminal>python -c "import os; print('JUPYTER_NOTEBOOK:', 'notebook' in os.environ.get('IPythonKernel', ''))"</terminal>
+
+  <terminal>python -c "import matplotlib.pyplot as plt; plt.figure(); print('figure creation works')"</terminal>
+
+
+verify output directory
+
+  <terminal>ls -la plots/ 2>/dev/null || mkdir -p plots && echo "created plots/ directory"</terminal>
+
+  <terminal>ls -la figures/ 2>/dev/null || mkdir -p figures && echo "created figures/ directory"</terminal>
+
+
+check common data libraries
+
+  <terminal>python -c "import pandas; print('pandas', pandas.__version__)" 2>/dev/null || echo "pandas not installed"</terminal>
+  <terminal>python -c "import numpy; print('numpy', numpy.__version__)" 2>/dev/null || echo "numpy not installed"</terminal>
+
+if missing:
+  <terminal>pip install pandas numpy</terminal>
+
+
+PHASE 1: VISUALIZATION FUNDAMENTALS
+
+
+understand your audience and purpose
+
+before creating any chart, answer these questions:
+
+  who will see this visualization?
+    - technical audience (developers, data scientists)
+    - business stakeholders (managers, executives)
+    - general audience (customers, public)
+    - mixed audience
+
+  what is the purpose?
+    - exploration (discovering patterns)
+    - explanation (communicating findings)
+    - persuasion (convincing action)
+    - monitoring (tracking metrics)
+
+  what action should the viewer take?
+    - make a decision
+    - understand a trend
+    - compare options
+    - spot anomalies
+
+  what data complexity can they handle?
+    - simple metrics and comparisons
+    - multivariate relationships
+    - statistical distributions
+    - time series with seasonality
+
+
+choose the right chart type
+
+based on your analysis goal:
+
+comparing values:
+  [ok] bar chart          - compare categories
+  [ok] column chart       - compare categories (vertical)
+  [ok] grouped bars       - compare multiple series
+  [ok] stacked bars       - show part-to-whole
+
+showing distribution:
+  [ok] histogram          - frequency distribution
+  [ok] density plot       - smooth distribution
+  [ok] box plot           - quartiles and outliers
+  [ok] violin plot        - distribution + box plot
+  [ok] ridgeline plot     - multiple distributions
+
+showing relationships:
+  [ok] scatter plot       - two variables
+  [ok] line chart         - trends over time
+  [ok] bubble chart       - three variables
+  [ok] heat map           - correlation matrix
+
+showing composition:
+  [ok] pie chart          - part-to-whole (avoid if >5 categories)
+  [ok] stacked bar        - composition over time
+  [ok] area chart         - composition over time
+  [ok] treemap            - hierarchical composition
+
+showing geospatial:
+  [ok] choropleth map     - values by region
+  [ok] bubble map         - locations with magnitude
+  [ok] flow map           - movement/connections
+
+
+chart selection decision tree
+
+  is it time series data?
+    yes -> line chart (if few series) or small multiples (if many)
+    no  -> continue
+
+  are you comparing categories?
+    yes -> bar chart (if horizontal labels long) or column chart
+    no  -> continue
+
+  are you showing distribution?
+    yes -> histogram (simple) or box plot (with outliers)
+    no  -> continue
+
+  are you showing correlation?
+    yes -> scatter plot (2 variables) or heat map (many variables)
+    no  -> continue
+
+  are you showing part-to-whole?
+    yes -> pie chart (if <=5 categories) or stacked bar (if time series)
+    no  -> reconsider data and goals
+
+
+PHASE 2: PLOT SETUP AND CONFIGURATION
+
+
+basic plot template
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+  import pandas as pd
+  import numpy as np
+
+  # set style
+  sns.set_style("whitegrid")
+  plt.figure(figsize=(12, 6))
+
+  # create your plot
+  # ... plotting code ...
+
+  # add labels and title
+  plt.xlabel("X-axis label")
+  plt.ylabel("Y-axis label")
+  plt.title("Descriptive Title Here")
+
+  # add grid
+  plt.grid(True, alpha=0.3)
+
+  # adjust layout
+  plt.tight_layout()
+
+  # save or show
+  plt.savefig("plots/output.png", dpi=300, bbox_inches="tight")
+  # plt.show()
+
+
+configure matplotlib defaults
+
+  import matplotlib.pyplot as plt
+
+  # set global style
+  plt.style.use('seaborn-v0_8-whitegrid')
+
+  # configure defaults
+  plt.rcParams.update({
+      'figure.figsize': (12, 6),
+      'font.size': 11,
+      'axes.labelsize': 12,
+      'axes.titlesize': 14,
+      'xtick.labelsize': 10,
+      'ytick.labelsize': 10,
+      'legend.fontsize': 10,
+      'figure.dpi': 100,
+      'savefig.dpi': 300,
+      'axes.grid': True,
+      'grid.alpha': 0.3,
+  })
+
+
+seaborn style options
+
+  import seaborn as sns
+
+  # available styles
+  # 'darkgrid', 'whitegrid', 'dark', 'white', 'ticks'
+
+  # set style
+  sns.set_style("whitegrid")
+
+  # set color palette
+  sns.set_palette("husl")
+
+  # available palettes:
+  # 'deep', 'muted', 'pastel', 'bright', 'dark', 'colorblind'
+  # 'husl', 'hls', 'Set1', 'Set2', 'Set3'
+  # 'Blues', 'Reds', 'Greens', etc. (sequential)
+  # 'RdBu', 'RdYlBu', etc. (diverging)
+
+
+color palette best practices
+
+  categorical data (5-7 categories):
+    sns.set_palette("husl", n_colors=len(categories))
+
+  sequential data (low to high):
+    sns.set_palette("Blues")
+
+  diverging data (neutral + extremes):
+    sns.set_palette("RdBu_r")
+
+  accessible colors:
+    sns.set_palette("colorblind")
+
+
+PHASE 3: BASIC CHART TYPES
+
+
+bar charts (categorical comparisons)
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+  import pandas as pd
+
+  # simple bar chart
+  plt.figure(figsize=(10, 6))
+  sns.barplot(data=df, x='category', y='value')
+  plt.title("Sales by Category")
+  plt.xlabel("Category")
+  plt.ylabel("Sales ($)")
+  plt.xticks(rotation=45)
+  plt.tight_layout()
+  plt.savefig("plots/bar_chart.png", dpi=300)
+
+
+  # horizontal bar chart (better for long labels)
+  plt.figure(figsize=(10, 6))
+  sns.barplot(data=df, y='category', x='value')
+  plt.title("Sales by Category")
+  plt.xlabel("Sales ($)")
+  plt.ylabel("Category")
+  plt.tight_layout()
+  plt.savefig("plots/horizontal_bar_chart.png", dpi=300)
+
+
+  # grouped bar chart (multiple series)
+  plt.figure(figsize=(12, 6))
+  sns.barplot(data=df, x='category', y='value', hue='year')
+  plt.title("Sales by Category and Year")
+  plt.xlabel("Category")
+  plt.ylabel("Sales ($)")
+  plt.legend(title="Year")
+  plt.tight_layout()
+  plt.savefig("plots/grouped_bar_chart.png", dpi=300)
+
+
+histograms (distributions)
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+
+  # basic histogram
+  plt.figure(figsize=(10, 6))
+  sns.histplot(data=df, x='value', bins=30)
+  plt.title("Distribution of Values")
+  plt.xlabel("Value")
+  plt.ylabel("Frequency")
+  plt.tight_layout()
+  plt.savefig("plots/histogram.png", dpi=300)
+
+
+  # histogram with kde curve
+  plt.figure(figsize=(10, 6))
+  sns.histplot(data=df, x='value', bins=30, kde=True)
+  plt.title("Distribution of Values with KDE")
+  plt.xlabel("Value")
+  plt.ylabel("Frequency")
+  plt.tight_layout()
+  plt.savefig("plots/histogram_kde.png", dpi=300)
+
+
+  # multiple histograms
+  plt.figure(figsize=(10, 6))
+  sns.histplot(data=df, x='value', hue='category', bins=30, alpha=0.5)
+  plt.title("Distribution by Category")
+  plt.xlabel("Value")
+  plt.ylabel("Frequency")
+  plt.tight_layout()
+  plt.savefig("plots/histogram_multiple.png", dpi=300)
+
+
+box plots (distributions with outliers)
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+
+  # single box plot
+  plt.figure(figsize=(10, 6))
+  sns.boxplot(data=df, y='value')
+  plt.title("Distribution of Values")
+  plt.ylabel("Value")
+  plt.tight_layout()
+  plt.savefig("plots/boxplot.png", dpi=300)
+
+
+  # box plot by category
+  plt.figure(figsize=(12, 6))
+  sns.boxplot(data=df, x='category', y='value')
+  plt.title("Distribution by Category")
+  plt.xlabel("Category")
+  plt.ylabel("Value")
+  plt.xticks(rotation=45)
+  plt.tight_layout()
+  plt.savefig("plots/boxplot_category.png", dpi=300)
+
+
+  # box plot with outliers highlighted
+  plt.figure(figsize=(12, 6))
+  sns.boxplot(data=df, x='category', y='value', showfliers=True)
+  plt.title("Distribution by Category with Outliers")
+  plt.xlabel("Category")
+  plt.ylabel("Value")
+  plt.xticks(rotation=45)
+  plt.tight_layout()
+  plt.savefig("plots/boxplot_outliers.png", dpi=300)
+
+
+scatter plots (relationships)
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+
+  # basic scatter plot
+  plt.figure(figsize=(10, 6))
+  sns.scatterplot(data=df, x='variable_x', y='variable_y')
+  plt.title("Relationship between X and Y")
+  plt.xlabel("Variable X")
+  plt.ylabel("Variable Y")
+  plt.tight_layout()
+  plt.savefig("plots/scatter.png", dpi=300)
+
+
+  # scatter with hue (color by category)
+  plt.figure(figsize=(10, 6))
+  sns.scatterplot(data=df, x='variable_x', y='variable_y', hue='category')
+  plt.title("Relationship by Category")
+  plt.xlabel("Variable X")
+  plt.ylabel("Variable Y")
+  plt.legend(title="Category")
+  plt.tight_layout()
+  plt.savefig("plots/scatter_hue.png", dpi=300)
+
+
+  # scatter with size (third variable)
+  plt.figure(figsize=(10, 6))
+  sns.scatterplot(data=df, x='variable_x', y='variable_y',
+                  hue='category', size='variable_z')
+  plt.title("Multi-variable Relationship")
+  plt.xlabel("Variable X")
+  plt.ylabel("Variable Y")
+  plt.legend(title="Category")
+  plt.tight_layout()
+  plt.savefig("plots/scatter_size.png", dpi=300)
+
+
+line charts (time series)
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+
+  # single line chart
+  plt.figure(figsize=(14, 6))
+  sns.lineplot(data=df, x='date', y='value')
+  plt.title("Value Over Time")
+  plt.xlabel("Date")
+  plt.ylabel("Value")
+  plt.tight_layout()
+  plt.savefig("plots/line_chart.png", dpi=300)
+
+
+  # multiple lines
+  plt.figure(figsize=(14, 6))
+  sns.lineplot(data=df, x='date', y='value', hue='category')
+  plt.title("Values Over Time by Category")
+  plt.xlabel("Date")
+  plt.ylabel("Value")
+  plt.legend(title="Category")
+  plt.tight_layout()
+  plt.savefig("plots/line_chart_multiple.png", dpi=300)
+
+
+  # line chart with confidence interval
+  plt.figure(figsize=(14, 6))
+  sns.lineplot(data=df, x='date', y='value', ci=95)
+  plt.title("Value Over Time with 95% CI")
+  plt.xlabel("Date")
+  plt.ylabel("Value")
+  plt.tight_layout()
+  plt.savefig("plots/line_chart_ci.png", dpi=300)
+
+
+PHASE 4: ADVANCED VISUALIZATION TECHNIQUES
+
+
+small multiples (facet grids)
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+
+  # facet grid by category
+  g = sns.FacetGrid(df, col='category', col_wrap=3,
+                    height=4, aspect=1.2)
+  g.map(sns.histplot, 'value', bins=20)
+  g.fig.suptitle("Distribution by Category", y=1.02)
+  plt.tight_layout()
+  plt.savefig("plots/facet_grid.png", dpi=300)
+
+
+  # facet grid with multiple variables
+  g = sns.FacetGrid(df, row='category1', col='category2',
+                    height=4, aspect=1.2)
+  g.map(sns.scatterplot, 'x', 'y')
+  g.fig.suptitle("Scatter Plots by Category", y=1.02)
+  plt.tight_layout()
+  plt.savefig("plots/facet_grid_2d.png", dpi=300)
+
+
+  # pair plot (all pairwise relationships)
+  sns.pairplot(df, hue='category', diag_kind='hist')
+  plt.suptitle("Pairwise Relationships", y=1.02)
+  plt.tight_layout()
+  plt.savefig("plots/pair_plot.png", dpi=300)
+
+
+heat maps (correlation matrices)
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+  import pandas as pd
+
+  # correlation heatmap
+  plt.figure(figsize=(10, 8))
+  correlation_matrix = df.corr()
+  sns.heatmap(correlation_matrix, annot=True, fmt='.2f',
+              cmap='coolwarm', center=0,
+              square=True, linewidths=0.5)
+  plt.title("Correlation Matrix")
+  plt.tight_layout()
+  plt.savefig("plots/correlation_heatmap.png", dpi=300)
+
+
+  # heatmap with custom palette
+  plt.figure(figsize=(12, 8))
+  sns.heatmap(correlation_matrix, annot=True, fmt='.2f',
+              cmap='RdBu_r', center=0, vmin=-1, vmax=1,
+              cbar_kws={'label': 'Correlation'},
+              square=True, linewidths=0.5)
+  plt.title("Correlation Matrix")
+  plt.tight_layout()
+  plt.savefig("plots/correlation_heatmap_custom.png", dpi=300)
+
+
+violin plots (detailed distributions)
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+
+  # single violin plot
+  plt.figure(figsize=(8, 6))
+  sns.violinplot(data=df, y='value')
+  plt.title("Distribution of Values")
+  plt.ylabel("Value")
+  plt.tight_layout()
+  plt.savefig("plots/violinplot.png", dpi=300)
+
+
+  # violin plot by category
+  plt.figure(figsize=(12, 6))
+  sns.violinplot(data=df, x='category', y='value')
+  plt.title("Distribution by Category")
+  plt.xlabel("Category")
+  plt.ylabel("Value")
+  plt.xticks(rotation=45)
+  plt.tight_layout()
+  plt.savefig("plots/violinplot_category.png", dpi=300)
+
+
+  # split violin plot
+  plt.figure(figsize=(12, 6))
+  sns.violinplot(data=df, x='category', y='value',
+                 hue='subcategory', split=True)
+  plt.title("Distribution by Category and Subcategory")
+  plt.xlabel("Category")
+  plt.ylabel("Value")
+  plt.legend(title="Subcategory")
+  plt.xticks(rotation=45)
+  plt.tight_layout()
+  plt.savefig("plots/violinplot_split.png", dpi=300)
+
+
+time series visualization
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+
+  # time series with trend line
+  plt.figure(figsize=(14, 6))
+  sns.lineplot(data=df, x='date', y='value', label='Value')
+  sns.regplot(data=df, x='date', y='value', scatter=False,
+              label='Trend', color='red')
+  plt.title("Value Over Time with Trend")
+  plt.xlabel("Date")
+  plt.ylabel("Value")
+  plt.legend()
+  plt.tight_layout()
+  plt.savefig("plots/time_series_trend.png", dpi=300)
+
+
+  # time series with moving average
+  df['moving_avg'] = df['value'].rolling(window=7).mean()
+
+  plt.figure(figsize=(14, 6))
+  sns.lineplot(data=df, x='date', y='value',
+                label='Value', alpha=0.6)
+  sns.lineplot(data=df, x='date', y='moving_avg',
+                label='7-day Moving Avg', linewidth=2)
+  plt.title("Value Over Time with Moving Average")
+  plt.xlabel("Date")
+  plt.ylabel("Value")
+  plt.legend()
+  plt.tight_layout()
+  plt.savefig("plots/time_series_ma.png", dpi=300)
+
+
+  # time series with seasonality decomposition
+  from statsmodels.tsa.seasonal import seasonal_decompose
+
+  df.set_index('date', inplace=True)
+  decomposition = seasonal_decompose(df['value'], model='additive',
+                                     period=365)
+
+  fig, axes = plt.subplots(4, 1, figsize=(14, 10))
+  decomposition.observed.plot(ax=axes[0], title='Observed')
+  decomposition.trend.plot(ax=axes[1], title='Trend')
+  decomposition.seasonal.plot(ax=axes[2], title='Seasonal')
+  decomposition.resid.plot(ax=axes[3], title='Residual')
+  plt.tight_layout()
+  plt.savefig("plots/time_series_decomposition.png", dpi=300)
+
+
+PHASE 5: CUSTOMIZATION AND STYLING
+
+
+annotations and text
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+
+  fig, ax = plt.subplots(figsize=(10, 6))
+
+  sns.barplot(data=df, x='category', y='value', ax=ax)
+
+  # add value labels on bars
+  for i, v in enumerate(df['value']):
+      ax.text(i, v, f'{v:.1f}', ha='center', va='bottom')
+
+  # add annotation for max value
+  max_idx = df['value'].idxmax()
+  max_cat = df.loc[max_idx, 'category']
+  max_val = df.loc[max_idx, 'value']
+  ax.annotate(f'Max: {max_val}',
+               xy=(max_idx, max_val),
+               xytext=(max_idx, max_val * 1.1),
+               arrowprops=dict(arrowstyle='->', color='red'))
+
+  plt.title("Sales by Category with Annotations")
+  plt.xlabel("Category")
+  plt.ylabel("Sales ($)")
+  plt.xticks(rotation=45)
+  plt.tight_layout()
+  plt.savefig("plots/annotated_bars.png", dpi=300)
+
+
+custom legends
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+
+  fig, ax = plt.subplots(figsize=(10, 6))
+
+  sns.lineplot(data=df, x='date', y='value',
+                hue='category', ax=ax)
+
+  # customize legend
+  ax.legend(title='Category',
+            bbox_to_anchor=(1.05, 1),
+            loc='upper left',
+            ncol=1,
+            frameon=True,
+            shadow=True)
+
+  plt.title("Values Over Time")
+  plt.xlabel("Date")
+  plt.ylabel("Value")
+  plt.tight_layout()
+  plt.savefig("plots/custom_legend.png", dpi=300)
+
+
+custom axes and limits
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+
+  fig, ax = plt.subplots(figsize=(10, 6))
+
+  sns.scatterplot(data=df, x='x', y='y', hue='category', ax=ax)
+
+  # set custom limits
+  ax.set_xlim(0, 100)
+  ax.set_ylim(0, 100)
+
+  # log scale
+  ax.set_xscale('log')
+  ax.set_yscale('log')
+
+  # custom ticks
+  ax.set_xticks([1, 10, 100])
+  ax.set_yticks([1, 10, 100])
+
+  plt.title("Log-scale Scatter Plot")
+  plt.xlabel("X (log scale)")
+  plt.ylabel("Y (log scale)")
+  plt.tight_layout()
+  plt.savefig("plots/custom_axes.png", dpi=300)
+
+
+subplots and figure layout
+
+  import matplotlib.pyplot as plt
+  import seaborn as sns
+
+  fig, axes = plt.subplots(2, 2, figsize=(14, 10))
+
+  # subplot 1: histogram
+  sns.histplot(data=df, x='value', bins=30, ax=axes[0, 0])
+  axes[0, 0].set_title('Distribution')
+
+  # subplot 2: box plot
+  sns.boxplot(data=df, x='category', y='value', ax=axes[0, 1])
+  axes[0, 1].set_title('By Category')
+  axes[0, 1].tick_params(axis='x', rotation=45)
+
+  # subplot 3: scatter plot
+  sns.scatterplot(data=df, x='x', y='y', ax=axes[1, 0])
+  axes[1, 0].set_title('Relationship')
+
+  # subplot 4: time series
+  sns.lineplot(data=df, x='date', y='value', ax=axes[1, 1])
+  axes[1, 1].set_title('Over Time')
+
+  plt.tight_layout()
+  plt.savefig("plots/subplots.png", dpi=300)
+
+
+PHASE 6: EXPORTING AND FORMATTING
+
+
+high-quality output formats
+
+  import matplotlib.pyplot as plt
+
+  # PNG (lossless, transparency support)
+  plt.savefig("plots/output.png",
+              dpi=300,
+              bbox_inches='tight',
+              transparent=False,
+              facecolor='white')
+
+  # PDF (vector, publication quality)
+  plt.savefig("plots/output.pdf",
+              bbox_inches='tight',
+              transparent=False,
+              facecolor='white')
+
+  # SVG (vector, web-friendly)
+  plt.savefig("plots/output.svg",
+              bbox_inches='tight',
+              transparent=False,
+              facecolor='white')
+
+  # EPS (vector, LaTeX friendly)
+  plt.savefig("plots/output.eps",
+              bbox_inches='tight',
+              transparent=False,
+              facecolor='white')
+
+
+resolution and size guidelines
+
+  for web display:
+    dpi: 72-100
+    figsize: (10, 6) or (12, 6)
+    format: PNG
+
+  for presentations:
+    dpi: 150-200
+    figsize: (12, 7) or (14, 8)
+    format: PNG or PDF
+
+  for publications:
+    dpi: 300-600
+    figsize: (8, 5) or (10, 6)
+    format: PDF, EPS, or SVG
+
+  for posters:
+    dpi: 300+
+    figsize: (20, 15) or larger
+    format: PDF or PNG
+
+
+PHASE 7: INTERACTIVE VISUALIZATION
+
+
+plotly interactive charts
+
+  import plotly.express as px
+  import plotly.graph_objects as go
+
+  # install if needed
+  # pip install plotly
+
+  # interactive scatter plot
+  fig = px.scatter(df, x='x', y='y', color='category',
+                   hover_data=['value'], title="Interactive Scatter")
+  fig.write_html("plots/interactive_scatter.html")
+  fig.show()
+
+
+  # interactive line chart
+  fig = px.line(df, x='date', y='value', color='category',
+                title="Interactive Time Series")
+  fig.write_html("plots/interactive_line.html")
+  fig.show()
+
+
+  # interactive histogram
+  fig = px.histogram(df, x='value', color='category',
+                     nbins=30, title="Interactive Distribution")
+  fig.write_html("plots/interactive_histogram.html")
+  fig.show()
+
+
+PHASE 8: DATA VISUALIZATION CHECKLIST
+
+
+pre-visualization checklist
+
+  [ ] understand audience and purpose
+  [ ] define the message you want to communicate
+  [ ] choose appropriate chart type for data and goal
+  [ ] verify data quality and completeness
+  [ ] handle missing values appropriately
+  [ ] ensure data types are correct
+
+
+design checklist
+
+  [ ] axis labels are clear and descriptive
+  [ ] title communicates main insight
+  [ ] color palette is accessible and appropriate
+  [ ] legend is positioned to not obscure data
+  [ ] font size is readable at output size
+  [ ] aspect ratio preserves data proportions
+  [ ] grid lines aid reading without distraction
+  [ ] annotations add value, not clutter
+
+
+accuracy checklist
+
+  [ ] y-axis starts at zero (unless intentionally broken)
+  [ ] scales are appropriate for data range
+  [ ] error bars/CI shown where appropriate
+  [ ] sample size indicated for small datasets
+  [ ] outliers are not arbitrarily removed
+  [ ] time series uses consistent intervals
+  [ ] geographic projections are accurate
+
+
+accessibility checklist
+
+  [ ] color-blind friendly palette
+  [ ] high contrast (minimum 4.5:1 for text)
+  [ ] patterns in addition to colors
+  [ ] alt text provided for web use
+  [ ] scalable vector format for large displays
+  [ ] annotations are readable
+
+
+PHASE 9: COMMON VISUALIZATION MISTAKES
+
+
+mistake: misleading y-axis
+
+wrong:
+  y-axis doesn't start at zero, exaggerating differences
+  different scales used for comparison
+
+correct:
+  start y-axis at zero for bar/column charts
+  use consistent scales for comparison
+  if breaking axis is necessary, clearly mark it
+
+
+mistake: too much data in one chart
+
+wrong:
+  20+ categories in a single pie chart
+  50+ time series in one line chart
+  all variables in one scatter plot matrix
+
+correct:
+  aggregate small categories into "other"
+  use small multiples for many series
+  filter to most important variables
+
+
+mistake: 3d charts for 2d data
+
+wrong:
+  3d pie charts
+  3d bar charts
+  3d line charts
+
+correct:
+  use 2d charts - they're more accurate and readable
+  3d only adds value for actual 3d data
+
+
+mistake: rainbow color palettes
+
+wrong:
+  using all colors of the rainbow
+  colors that don't have perceptual ordering
+
+correct:
+  use sequential palettes for ordered data
+  use diverging palettes for data with meaningful midpoint
+  use categorical palettes for nominal data
+
+
+mistake: decorative over functional
+
+wrong:
+  excessive styling that obscures data
+  fancy fonts that are hard to read
+  effects that don't add insight
+
+correct:
+  form follows function
+  every visual element should convey information
+  keep it simple and clear
+
+
+PHASE 10: VISUALIZATION RULES (MANDATORY)
+
+
+while this skill is active, these rules are MANDATORY:
+
+  [1] ALWAYS START WITH DATA EXPLORATION
+      never visualize without understanding the data
+      check distributions, missing values, outliers first
+
+  [2] CHOOSE THE RIGHT CHART TYPE
+      match chart to data type and analysis goal
+      don't force data into inappropriate visualizations
+
+  [3] LABEL EVERYTHING CLEARLY
+      axis labels, title, legend - all must be descriptive
+      assume reader knows nothing about the context
+
+  [4] USE ACCESSIBLE COLORS
+      color-blind friendly palettes
+      sufficient contrast
+      multiple cues (color + pattern) when possible
+
+  [5] PRESERVE DATA PROPORTIONS
+      y-axis starts at zero for comparison charts
+      don't distort with broken axes
+      maintain aspect ratio for spatial data
+
+  [6] HANDLE OUTLIERS APPROPRIATELY
+      show outliers in distributions
+      don't hide them
+      explain if they're excluded
+
+  [7] PROVIDE CONTEXT
+      sample size, time period, data source
+      confidence intervals where appropriate
+      methodology notes
+
+  [8] AVOID CHART JUNK
+      remove 3d effects, unnecessary borders
+      simplify grid lines
+      every element should serve a purpose
+
+  [9] TEST AT OUTPUT SIZE
+      verify readability at intended display size
+      adjust font sizes and figure dimensions
+      ensure legends don't overlap
+
+  [10] TELL A STORY
+      highlight key insights
+      guide viewer to important patterns
+      use annotations strategically
+
+
+FINAL REMINDERS
+
+
+clarity over cleverness
+
+the goal is communication, not decoration.
+if a chart doesn't communicate, it's failed.
+
+
+less is more
+
+remove everything that doesn't add information.
+simplify, simplify, simplify.
+
+
+data first, visualization second
+
+understand the data before visualizing.
+visualization should reveal insights,
+not create them.
+
+
+test with your audience
+
+show drafts to intended audience.
+get feedback on clarity and effectiveness.
+iterate based on feedback.
+
+
+document your choices
+
+why this chart type?
+  why these colors?
+  why this aggregation?
+  why this filtering?
+
+future viewers will thank you.
+
+
+now create visualizations that inform,
+insight, and inspire action.