DataAnalyzer¶

from augini import DataAnalyzer
import pandas as pd

# Initialize with configuration
config = {
    'api_key': 'your-api-key',
    'model': 'gpt-4-turbo-preview'
}

analyzer = DataAnalyzer(config=config)

# Load your data and prepare analyzer
data = pd.read_csv('your_data.csv')
analyzer.fit(data)

# Ask questions about your data
insights = analyzer.chat("What are the main trends in this dataset?")
print(insights)

# Ask about statistical patterns
stats = analyzer.chat(
    "What are the key statistical patterns in the data? "
    "Include mean, median, and correlations in your analysis."
)

# Ask about patterns in time series
patterns = analyzer.chat(
    "What patterns do you see in the data over time? "
    "Focus on the 'date' column."
)

# Ask about trends by category
trends = analyzer.chat(
    "How do metrics vary across different categories? "
    "Group the analysis by 'category' column."
)

# Enable conversation memory for context-aware analysis
analyzer = DataAnalyzer(
    api_key='your-api-key',
    enable_memory=True
)
analyzer.fit(data)

# First question
response1 = analyzer.chat(
    "What's the average age in the dataset?",
    use_memory=True
)

# Follow-up question (uses context from previous question)
response2 = analyzer.chat(
    "How does it correlate with income?",
    use_memory=True
)

# Ask specific analytical questions
analysis = analyzer.chat(
    "Create a cohort analysis based on signup date. "
    "Show retention rates over time and identify key patterns."
)

config = {
    # Model settings
    'model': 'gpt-4-turbo-preview',
    'temperature': 0.7,

    # Memory settings
    'enable_memory': True,
    'context_window_tokens': 1000,

    # Debug settings
    'debug': True,
    'log_level': 'INFO'
}