Your submission will be evaluated using a **custom composite metric** that balances **classification accuracy** and **regression error**:

### Stage 1: 
Prediction of Material & Equipment along with qty(**50 Pointers**)

1. **Classification (MasterItemNo)**  
   - Metric: **Accuracy** + **F1** (exact match between predicted and true values).  
   - Score range: `0 → 1` (higher is better).

2. **Regression (QtyShipped)**  
   - Metric: **Mean Absolute Error (MAE)**.  
   - To combine with accuracy, MAE is **normalized** by the range of true values:  
  `norm_mae = MAE/ (max(QtyShipped) - min(QtyShipped)`
   - Clipped into `[0,1]` and inverted to a score: 
      `reg_score = 1 - norm_mae`
    

3. **Final Score**  
   - Weighted equally:  
`final_score = 0.25×accuracy+0.25×F1+0.5×reg_scor`

   - Range: `0 → 1` (higher is better).

If all true `QtyShipped` values are identical, the regression component is automatically treated as perfect (`reg_score = 1`).

### **Stage 2: Host app (5 Pointers)**
- Host the application on a platform like Vercel or any other relevant technology.  
- Integrate a chatbot to provide answers to user queries based on the outputs from **Stage 1**.  
- Ensure smooth interaction between the hosted app and chatbot.  
- Provide a clean, user-friendly interface.  
- Verify deployment is publicly accessible.  


### **Stage 3: Identify Vendor from Web Scraping (15 Pointers)**
- Perform web scraping to identify relevant vendors.  
- Extract and present vendor details such as:  
  - Company name  
  - Services/products offered  
  - Location  
  - Contact information  
- Ensure compliance with ethical and legal scraping practices.  
- Provide structured output (JSON or UI integration).  


### **Stage 4: Integrate with Construction Project Schedule (10 Pointers)**
- Create a **synthetic construction project schedule** based on the given problem statement.  
- Integrate outputs from **Stage 2 & Stage 3** into the project schedule.  
- Represent it in a **Gantt chart / MS Project / tabular timeline** format.  
- Highlight dependencies, lead times, and milestones.  
- Show how vendor engagement and procurement timelines align with the project.  


### **Stage 5: Create Synthetic Procurement Plan (10 Pointers)**
- Develop a **procurement plan** tailored to the synthetic construction project.  
- Include:  
  - Materials & resources required  
  - Vendor mapping from Stage 3  
  - Lead times & procurement cycles  
  - Risk mitigation strategies  
- Present the plan in a clear, structured format.  


### **Stage 6: Procurement Management Platform (10 Pointers)**
- Design a **Procurement Management Platform** prototype.  
- Include:  
  - Vendor database integration  
  - Procurement request and approval workflow  
  - Timeline tracking (integration with Stage 4)  
  - Dashboard for monitoring procurement activities  
- Demonstrate how it streamlines procurement and vendor management for the synthetic construction project.  


## Example Evaluation  

**Problem Statement:**  
Create a predictive model for a **Data Center Project** with the following parameters:  

- **Power Capacity:** 25 MegaWatts  
- **Built-up Area:** 2 Lakhs Square Feet  
- **Location:** Maharashtra, India