ai-critic 3.5.0

Graph-based evaluation engine for machine learning models

🚀 AI Critic 3.5.0 (Production Readiness Engine)

pip install ai-critic

AI Critic is a graph-based evaluation engine for machine learning models, designed to go beyond isolated metrics.

It runs a structured evaluation pipeline that analyzes multiple dimensions — performance, robustness, explainability, data quality, and structure — delivering a unified, interpretable, and actionable report.

---

🔥 WHAT’S NEW IN 3.5.0

🧠 Production-First Design

• One-line evaluation: evaluate()
• Simplified API for fast adoption
• Built for real-world deployment decisions

---

⚡ Standard Usage (NEW)

AI Critic is now designed to be used right after training:

import ai_critic

report = ai_critic.evaluate(model, X, y)

---

🚫 Quality Gate (NEW — CRITICAL)

Turn evaluation into a deployment decision:

from ai_critic import evaluate
from ai_critic.gate import enforce

report = evaluate(model, X, y)

enforce(report, threshold=75)

If the model is not good enough → deployment is blocked.

---

📦 Standardized Report (JSON-first)

All results follow the same schema:

report = {
    "scores": {},        # technical scores (0–1)
    "details": {},       # raw evaluator outputs
    "risk": {},          # interpretable score (0–100)
    "summary": {},       # human-readable insights
    "suggestions": []    # recommended actions
}

👉 This makes AI Critic:

• API-ready
• Easy to log and persist
• Production-ready

---

⚡ Improved Graph Engine

• Dependency-aware execution (topological sort)
• Parallel execution support
• Deterministic evaluation order

---

🎯 Multi-layer Scoring System

• Technical score (0–1) → aggregation layer
• Risk score (0–100) → decision layer

---

💡 Integrated Suggestion Engine

• Automatically generates recommendations based on model behavior

---

🧩 Plugin System

• Clean evaluator interface
• Dependency-aware plugins
• Easily extensible evaluation pipeline

---

⚡ QUICK START

🧠 One-liner (recommended)

import ai_critic

report = ai_critic.evaluate(model, X, y)

print(report["risk"])
print(report["summary"])

---

🔐 Production usage (recommended)

from ai_critic import evaluate
from ai_critic.gate import enforce

report = evaluate(model, X, y)

# 🚫 blocks bad models
enforce(report, threshold=75)

---

🧪 Full control (advanced)

from api.client import AICritic
from sklearn.ensemble import RandomForestClassifier
from sklearn.datasets import load_iris

data = load_iris()
X, y = data.data, data.target

model = RandomForestClassifier().fit(X, y)

critic = AICritic(weights={
    "performance": 1.0,
    "robustness": 1.5
})

report = critic.evaluate(model, X, y, parallel=True)

---

🧩 INTERNAL PIPELINE

evaluate()
   ↓
EvaluationGraph (nodes)
   ↓
raw_results
   ↓
ScoreAggregator (0–1)
   ↓
build_report()
   ↓
scoring.py (risk 0–100)
   ↓
summary.py (human-readable)
   ↓
SuggestionEngine

---

🧱 CORE COMPONENTS

1. Evaluation Graph

A DAG-based execution system:

• Automatically resolves dependencies
• Executes nodes in correct order
• Enables parallel execution

Example:

performance → robustness → explainability

---

2. Score Aggregator

Combines evaluator outputs:

critic = AICritic(weights={
    "performance": 1.0,
    "robustness": 2.0
})

---

3. Evaluator Plugins

Fully extensible via plugins:

from ai_critic.plugins.base import EvaluatorPlugin
from ai_critic.plugins.registry import EvaluatorRegistry

class FairnessEvaluator(EvaluatorPlugin):
    name = "fairness"
    dependencies = ["performance"]
    weight = 1.0

    def evaluate(self, model, dataset, context=None):
        return {
            "score": 0.92,
            "verdict": "stable",
            "message": "Fairness is acceptable"
        }

EvaluatorRegistry.register(FairnessEvaluator())

---

4. Risk Scoring (0–100)

Transforms technical signals into decision-ready output:

report["risk"] = {
    "global_score": 78.5,
    "verdict": "usable_with_caution",
    "component_scores": {...},
    "penalties": [...]
}

---

5. Human Summary

High-level interpretation:

report["summary"] = {
    "executive_summary": {
        "verdict": "⚠️ Risky",
        "deploy_recommended": False
    }
}

---

6. Suggestion Engine

Actionable insights:

[
    "Check for data leakage",
    "Improve robustness with regularization"
]

---

🖥️ CLI

Run directly from terminal:

ai-critic --model model.pkl --data dataset.csv --target label

🔥 CI/CD Mode (recommended)

ai-critic --model model.pkl --data dataset.csv --target label --fail-on-risk

👉 Fails automatically if model risk is too high.

---

🧠 DESIGN PHILOSOPHY

1. Single Source of Truth

One unified data format → no inconsistencies

---

2. Graph-first Thinking

Evaluation is a dependency-driven pipeline, not isolated functions

---

3. JSON-native

Everything is ready for:

• APIs
• dashboards
• logging
• SaaS platforms

---

4. Actionable AI

Not just metrics — decisions:

• Should you deploy?
• Where is the risk?
• What should be improved?

---

🔥 POSITIONING

AI Critic is not just a metrics library.

It is a:

🧠 Production gatekeeper for machine learning models

---

🚀 ROADMAP

• REST API (/evaluate)
• Visual dashboard
• Model monitoring (post-deployment)
• Continuous evaluation (CI/CD)
• Global benchmarking between models

---

📄 LICENSE

MIT License