Explore What Changed
PriorityX helps answer a practical monitoring question: which entities need attention, and how have their priorities changed over time?
Fit entity priorities in Python, then open the saved results in AutoViz. The local app uses Plotly charts to link a quadrant matrix, a YTD transition timeline, and trajectories. Select an entity across charts, zoom or reset the view, and inspect exact values in tables.
Use your own entity labels with Monthly or Quarterly results. Quarterly opens when available. Adjust the automatic layout and save portable view settings. Export an interactive HTML view that reopens offline, or PNG and SVG figures through Matplotlib.
AutoViz requires Python 3.12 or later and runs locally in your browser. Opening results and changing layouts do not refit models or change analytical values.
Read the AutoViz guide or the saved-results guide.
Try the Saved IT Incidents Example
Save the public example bundle as incidents-quarterly.json. From that folder, install PriorityX in a Python 3.12+ environment and launch it:
python -m pip install priorityx
python -m priorityx app --bundle incidents-quarterly.json
The sample contains five services with retained saved history from 2023 to 2025. It opens Quarterly results through 2025-Q4, with a 2025 YTD timeline and three automatically selected trajectories. This demonstrates existing saved results, rather than a new fit or a benchmark.
Follow the example walkthrough or the installation guide.
Analyze Your Event Data
The Python API estimates two relative scores with Generalized Linear Mixed Models: an entity’s baseline level compared with its peers, and its trend. Those scores place each entity in a quadrant for prioritization.
import pandas as pd
import priorityx as px
events = pd.read_csv("events.csv", parse_dates=["opened_at"])
matrix = px.fit_priority_matrix(
events,
entity_col="service",
timestamp_col="opened_at",
temporal_granularity="quarterly",
)
px.plot_priority_matrix(matrix, entity_name="Service", plot_path="matrix.png")
PriorityX also provides cumulative movement tracking, transition detection, driver analysis, custom score axes, and weighted composite indices. The Python quickstart covers analysis; the saved-results guide explains how to open its outputs in AutoViz.