Get query results

API reference for the DataAnalyzr ask method

DataAnalyzr.ask(
    user_input: str = None,
    outputs: list = None,
    plot_path: str = None,
    recommendations_params: dict = None,
    counts: dict = None,
    context: dict = None,
    **kwargs,
) -> dict

This method orchestrates the analysis process and generates visualization, insights, recommendations, and tasks based on the specified inputs and parameters. It is the recommended method to use for generating visualization, insights, recommendations, and tasks.

1. The method coordinates the analysis process and generates visualisation, insights, recommendations, and tasks based on the specified parameters.
2. Uses the analysis, visualisation, insights, recommendations, and tasks methods internally.
3. It allows for customization of outputs, analysis rerun, insights usage, recommendations format, output type, and counts.
4. The generated outputs are returned as a dictionary.

Parameters

user_input · string · required

The input string provided by the user for generating outputs.

user_input = "What are the insights for the dataset?"

outputs · list

List of outputs to generate. Possible values are visualisation, insights, recommendations, and tasks. Defaults to ["visualisation", "insights", "recommendations", "tasks"].

outputs = ["insights", "recommendations"]

plot_path · string

Path to save the generated plot. Only relevant when outputs includes visualisation. Defaults to generated_plots/<random-plot-name>.png.

plot_path = "path/to/save/plot.png"

recommendations_params · dictionary

Parameters for generating recommendations. Includes from_insights, output_type and json_format.

recommendations_params = {
    "from_insights": True,
    "output_type": "json",
    "json_format": {
        "Recommendation": "Recommendation text",
        "Reason": "Reason for the recommendation",
        "Impact": "Impact of the recommendation",
    },
}

Show details

from_insights · boolean

Whether to generate recommendations from insights. Default is True.

output_type · Literal['json', 'text']

Type of output for recommendations. Default is json.

json_format · dictionary

Format for the JSON output of recommendations.

counts · dictionary

Dictionary containing the counts of insights, recommendations, and tasks to generate.

counts = {
    "insights": 3,
    "recommendations": 3,
    "tasks": 5,
}

Show details

insights · integer

Number of insights to generate. Default is 3.

recommendations · integer

Number of recommendations to generate. Default is 3.

tasks · integer

Number of tasks to generate. Default is 5.

context · dictionary

Context dictionary for the analysis. Includes analysis, visualisation, insights, recommendations, and tasks.

context = {
    "analysis": "Analyze sales data",
    "visualisation": "Context for visualisation",
    "insights": "Context for insights",
    "recommendations": "Improve sales performance",
    "tasks": "Analyze sales data",
}

Show details

analysis · string

Context for the analysis method.

visualisation · string

Context for the visualisation method.

insights · string

Context for the insights method.

recommendations · string

Context for the recommendations method.

tasks · string

Context for the tasks method.

rerun_analysis · boolean

Whether to rerun the analysis before generating outputs. Default is True.

rerun_analysis = True

max_retries · integer

Maximum number of retries for generating outputs. Default is 10.

max_retries = 5

time_limit · integer

Time limit in seconds for generating outputs. Default is 45 for analysis and 60 for visualization.

time_limit = 60

auto_train · boolean

Whether to automatically add questions with their SQL query or Python code to the vector store. Default is True.

auto_train = True

Returns

response · dictionary

Dictionary containing the generated outputs. Includes visualisation, insights, recommendations, and tasks.

response = {
    "visualisation": "path/to/plot.png",
    "insights": "-Insight 1\n-Insight 2\n-Insight 3",
    "recommendations": "Recommendation 1\nRecommendation 2\nRecommendation 3",
    "tasks": "Task 1\nTask 2\nTask 3\nTask 4\nTask 5",
}

Show details

visualisation · string

Path to the generated plot.

insights · string

Insights generated based on the analysis output.

recommendations · string

Recommendations generated based on the analysis output.

tasks · string

Tasks generated based on the analysis output.

Example usage

# Ask for visualisation, insights, recommendations, and tasks
response = data_analyzr.ask(
    user_input="What are the insights for the dataset?",
    outputs=["visualisation", "insights", "recommendations", "tasks"],
    plot_path="path/to/plot.png",
    recommendations_params={
        "from_insights": True,
        "output_type": "json",
        "json_format": {
            "Recommendation": "Recommendation text",
            "Reason": "Reason for the recommendation",
            "Impact": "Impact of the recommendation",
        },
    },
    counts={
        "insights": 3,
        "recommendations": 3,
        "tasks": 5,
    },
    context={
        "analysis": "Analyze sales data",
        "visualisation": "Context for visualisation",
        "insights": "Context for insights",
        "recommendations": "Improve sales performance",
        "tasks": "Analyze sales data",
    },
    rerun_analysis=True,
    max_retries=5,
    time_limit=60,
    auto_train=True,
)

# Display the visualisation
from PIL import Image
Image.open(response["visualisation"]).show()

# Print insights
print(response["insights"])

# Print recommendations
print(response["recommendations"])

# Print tasks
print(response["tasks"])