ryanontheinside-ComfyUI_RealtimeNodes/tests/README.md

ComfyUI RealTimeNodes Profiling Suite

This directory contains comprehensive profiling tests for the ComfyUI RealTimeNodes.

Features

• CPU and memory profiling for all detector nodes
• Tests with various image sizes (256x256, 512x512, 1024x1024)
• Different ROI configurations
• Sequence processing tests
• Multiple image types (static, noisy, gradient)
• JSON output of timing results
• Detailed memory usage tracking

Setup

pip install -r requirements.txt

Running Profiling Tests

python test_profiling.py

This will:

1. Create a test dataset in test_data/
2. Run comprehensive tests on all detectors
3. Generate two output files:
   • detector_profile.stats: Detailed CPU profiling data
   • profiling_results.json: Structured results including timing and memory usage

Visualizing Results

CPU Profile Visualization

snakeviz detector_profile.stats

This will open an interactive visualization showing:

• Call graphs
• Time spent in each function
• Number of calls
• Cumulative time

Memory Profile Analysis

Memory usage data is printed during test execution and saved in the JSON results file.

Test Coverage

The suite tests:

1. Individual detector nodes
   • Motion detector
   • Brightness detector
2. ROI configurations
   • Full frame
   • Partial frame
   • Different positions
3. Sequence processing
   • Multiple frame sequences
   • Different detector types
4. Image variations
   • Different sizes
   • Different content types
   • Different complexity levels

Interpreting Results

The profiling_results.json file contains:

• Timing data for each test configuration
• Memory usage statistics
• Node execution statistics

Use this data to:

• Identify performance bottlenecks
• Optimize memory usage
• Improve real-time processing capabilities