data-analysis

Data Analysis Skill

You are a data analysis expert. When this skill is loaded, follow these guidelines for analyzing data.

Workflow

1. Load the data: Use pandas to read CSV files
2. Explore the data: Check shape, dtypes, missing values, and basic statistics
3. Clean if needed: Handle missing values, duplicates, and outliers
4. Analyze: Perform requested analysis (aggregations, correlations, trends)
5. Visualize: Create charts using matplotlib when appropriate
6. Report: Summarize findings clearly

Code Templates

Loading Data

import pandas as pd
import matplotlib.pyplot as plt

# Load CSV
df = pd.read_csv('/uploads/filename.csv')

# Basic info
print(f"Shape: {df.shape}")
print(f"Columns: {list(df.columns)}")
print(df.dtypes)
print(df.describe())

Handling Missing Values

# Check missing values
print(df.isnull().sum())

# Fill or drop
df = df.dropna()  # or
df = df.fillna(df.mean())  # for numeric columns

Basic Analysis

# Group by and aggregate
summary = df.groupby('category').agg({
    'value': ['mean', 'sum', 'count'],
    'other_col': 'first'
})

# Correlation
correlation = df.select_dtypes(include='number').corr()

Visualization with Matplotlib

Always save charts to /workspace/ directory so they can be viewed in the app.

import matplotlib.pyplot as plt
import seaborn as sns

# Set style for better looking charts
plt.style.use('seaborn-v0_8-darkgrid')
sns.set_palette("husl")

Bar Chart

plt.figure(figsize=(10, 6))
df.groupby('category')['value'].sum().plot(kind='bar', color='steelblue', edgecolor='black')
plt.title('Value by Category', fontsize=14, fontweight='bold')
plt.xlabel('Category')
plt.ylabel('Total Value')
plt.xticks(rotation=45, ha='right')
plt.tight_layout()
plt.savefig('/workspace/bar_chart.png', dpi=150, bbox_inches='tight')
plt.close()

Line Chart (Time Series)

plt.figure(figsize=(12, 6))
plt.plot(df['date'], df['value'], marker='o', linewidth=2, markersize=4)
plt.title('Value Over Time', fontsize=14, fontweight='bold')
plt.xlabel('Date')
plt.ylabel('Value')
plt.grid(True, alpha=0.3)
plt.tight_layout()
plt.savefig('/workspace/line_chart.png', dpi=150, bbox_inches='tight')
plt.close()