Prompt

# 7-Step Machine Learning Model Development Chain

## Step 1: Problem Definition & Data Strategy
**Input:** Business problem requiring ML solution
**Task:** Define ML problem and data requirements
**Output:** ML problem definition and data strategy

---
**Instructions for Step 1:**
For ML problem: {ml_problem}

Define problem:
- Business objective translation
- ML problem formulation
- Success metrics definition
- Data requirements specification
- Feasibility assessment

**Pass to Step 2:** Problem definition

---

## Step 2: Data Collection & Exploration
**Input:** Problem definition from Step 1
**Task:** Collect data and perform exploratory analysis
**Output:** Analyzed dataset with insights

---
**Instructions for Step 2:**
Explore data:
- Data source identification
- Data collection and integration
- Exploratory data analysis
- Feature discovery
- Data quality assessment

**Pass to Step 3:** Explored dataset

---

## Step 3: Feature Engineering & Data Preprocessing
**Input:** Explored dataset from Step 2
**Task:** Engineer features and preprocess data
**Output:** Feature-engineered dataset

---
**Instructions for Step 3:**
Engineer features:
- Feature creation and selection
- Data transformation
- Missing value handling
- Outlier treatment
- Feature scaling and encoding

**Pass to Step 4:** Preprocessed data

---

## Step 4: Model Selection & Training
**Input:** Preprocessed data from Step 3
**Task:** Select and train ML models
**Output:** Trained ML models

---
**Instructions for Step 4:**
Train models:
- Algorithm selection
- Model architecture design
- Hyperparameter tuning
- Cross-validation
- Model training and optimization

**Pass to Step 5:** Trained models

---

## Step 5: Model Evaluation & Validation
**Input:** Trained models from Step 4
**Task:** Evaluate and validate model performance
**Output:** Validated ML model

---
**Instructions for Step 5:**
Validate model:
- Performance metric evaluation
- Bias and fairness testing
- Robustness assessment
- Business impact validation
- Model selection

**Pass to Step 6:** Validated model

---

## Step 6: Model Deployment & Integration
**Input:** Validated model from Step 5
**Task:** Deploy model to production environment
**Output:** Deployed ML model

---
**Instructions for Step 6:**
Deploy model:
- Deployment architecture design
- API development
- Integration with systems
- Performance optimization
- Security implementation

**Pass to Step 7:** Deployed model

---

## Step 7: Monitoring & Continuous Improvement
**Input:** Deployed model from Step 6
**Task:** Monitor model performance and improve continuously
**Output:** Production ML system with monitoring

---
**Instructions for Step 7:**
Monitor and improve:
- Performance monitoring setup
- Drift detection implementation
- Retraining pipeline creation
- A/B testing framework
- Continuous improvement cycle

**Final Output:** Production ML system with comprehensive monitoring and improvement framework