Test Plan
JLR AI Benchmarking Dashboard
Version 1.0 – July 2025
1. Purpose
This document describes the testing strategy for the JLR-AI-Benchmark-Prediction project
to ensure software quality, correctness, performance, and stability. It includes unit, integra-
tion, regression, and UI testing.
2. Scope
• Verify functionality of all ETL scripts and ML model pipelines
• Ensure predictions are consistent across platforms
• Validate Streamlit dashboard components and user interactions
• Ensure data correctness, bias correction accuracy, and UI responsiveness
3. Test Strategy
• Unit Tests: Test Python functions/modules independently
• Integration Tests: Test end-to-end workflows (data ingestion → model inference)
• Regression Tests: Run periodically to verify no functionality has broken after up-
dates
• UI Tests: Test user interface behavior manually or via screenshot comparison
4. Test Environment
• Python Version: 3.10+
• Libraries: pytest, pandas, xgboost, streamlit, joblib
• Database: PostgreSQL 14+
• Tools: VS Code, Streamlit, pgAdmin, GitHub Actions (optional)
5. Unit Test Coverage
• scripts/cleaning.py – test data formatting functions
• feature engineering.py – test metric generation and normalization
• bias correction.py – test bias application logic
• train xgboost.py – test model training with synthetic inputs
• predictor engine.py – test response structure
6. Sample Unit Test (Pytest)
def test_normalize_scores():
from scripts.bias_correction import normalize_scores
test_data = {’vendor’: [’NVIDIA’], ’score’: [100]}
df = pd.DataFrame(test_data)
result = normalize_scores(df)
assert ’normalized_score’ in result.columns
7. Integration Test Plan
• Input: ‘gpu.json‘
• Process:
1. Clean and normalize raw CSVs
2. Generate feature set and apply bias correction
3. Train model and test predictions from UI
• Expected Output:
– Valid latency, throughput, and efficiency predictions
– Confidence score > 90%
8. UI Testing Checklist
• Input panel loads and accepts JSON/gpu specs
• Predict button generates valid metrics
• Graphs render properly (bar chart, radar chart)
• Model selector updates outputs dynamically
• No component crashes under malformed input
9. Bug Reporting Protocol
• Reproduce the error and note environment
• Attach screenshot/log output
• File a GitHub Issue with:
– Reproduction steps
– Observed vs expected behavior
– Affected files/modules
10. Test Automation (Optional)
• Use pytest with CI tools like GitHub Actions
• Trigger tests on pull request or push to dev
• Sample .github/workflows/test.yml:
name: Test Pipeline
on: [push, pull_request]
jobs:
test:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- name: Set up Python
uses: actions/setup-python@v2
with:
python-version: 3.10
- name: Install dependencies
run: pip install -r requirements.txt
- name: Run tests
run: pytest
11. Test Ownership
• Unit Tests: Each developer for their module
• Integration Tests: Lead developer(s)
• Dashboard/UX Tests: Frontend or Streamlit specialist
• Review: Code reviewer must verify test coverage before merging
12. Approval and Review
• Test plan must be reviewed by supervisor or technical lead
• Coverage reports should be generated if using CI
