mjpsm/checkin-detail-classifier

license: mit tags:

• —text-classification
• —nlp
• —productivity
• —transformers datasets:
• —custom metrics:
• —accuracy ---

Check-in Detail Classifier

🔍 Overview

The Check-in Detail Classifier evaluates whether a check-in is detailed or not detailed.

This model ensures that student check-ins provide meaningful insight into their work rather than vague or minimal responses.

🎯 Purpose

Not all check-ins are useful. This model distinguishes between:

• —High-quality, descriptive updates
• —Low-effort or vague check-ins

🧠 Model Details

• —Task: Binary Text Classification
• —Labels:
• —DETAILED → Clear, informative check-in
• —NOT_DETAILED → Vague or minimal check-in
• —Architecture: Transformer-based (fine-tuned)

📦 Use Cases

• —Quality control for daily check-ins
• —Automated feedback systems
• —Performance tracking dashboards
• —Educational analytics

🧪 Example

Input

Today I worked on debugging my FastAPI deployment and fixed a memory issue on Render

Ouput

{
  "label": "DETAILED",
  "confidence": 0.96
}

Input

I worked on my project

Output

{
  "label": "NOT_DETAILED",
  "confidence": 0.94
}

Pipeline Integration

This model runs after check-in validation:

1. 1.Check-in or Not
2. 2.Check-in Detail Classifier (this model)
3. 3.Roadblock Detection
4. 4.Generation (if needed)

Usage (Python)

from transformers import pipeline

classifier = pipeline("text-classification", model="mjpsm/checkin-detail-classifier")

result = classifier("Today I worked on debugging my API")
print(result)

Future Improvements

1. 1.Add scoring system (0–10 detail rating)
2. 2.Provide feedback suggestions
3. 3.Expand dataset with more nuanced examples

👤 Author

• —Mazamesso Meba
• —Information Systems Student | AI Developer