Dc-4nderson/checkin-model

Check-In Expansion Model (BART fine-tuned)

This is a fine-tuned version of facebook/bart-base for internship/tech daily check-in expansions.
It takes a short input (a concise task description) and outputs a more polished, detailed check-in note.

🧩 Model Details

• Base Model: facebook/bart-base
• Fine-tuned On: 1130 examples (910 train / 220 eval) of custom internship-style check-in data.
• Task: Expand a short input task into a more structured daily check-in.
• Language: English

🛠️ Intended Uses

• Writing structured internship progress updates.
• Generating more detailed check-in notes from short inputs.
• Automating team updates in workflows (e.g., via n8n + Hugging Face API).

🚀 Example

Input

This morning I’ll spend some time organizing scripts and later test a few cases I wrote yesterday.

Output

Today, I’ll begin by organizing the scripts in my repository to make the project structure more manageable. Afterward, I’ll rerun the test cases I wrote yesterday to confirm they pass consistently. My objective is to reduce clutter, improve reliability, and ensure the codebase is stable before I move forward.

📊 Evaluation

Trained with ROUGE evaluation metrics.

ROUGE-1: ~ 0.014810

⚖️ License

MIT License

📎 Citation

If you use this model, please cite:

@misc{checkin_bart_2025, title={Check-In Expansion Model (BART fine-tuned)}, author={Anderson, Dequan}, year={2025}, publisher={Hugging Face} }