Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
task_categories:
|
| 4 |
+
- multiple-choice
|
| 5 |
+
- question-answering
|
| 6 |
+
language:
|
| 7 |
+
- en
|
| 8 |
+
tags:
|
| 9 |
+
- stem
|
| 10 |
+
- mcq-solver
|
| 11 |
+
- flan-t5
|
| 12 |
+
- minilm
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
# Smart MCQ Solver Challenge
|
| 16 |
+
|
| 17 |
+
This repository contains the machine learning pipeline and model solution for the **Smart MCQ Solver Challenge**.
|
| 18 |
+
|
| 19 |
+
## Models Included
|
| 20 |
+
1. **Model 1 (Scratch)**: GBDT trained on TF-IDF lexical features
|
| 21 |
+
2. **Model 2 (Pretrained)**: `google/flan-t5-large` (770M) zero-shot token probability ranker
|
| 22 |
+
3. **Model 3 (Ensemble)**: Soft-voting blend of `google/flan-t5-large` and `sentence-transformers/all-MiniLM-L6-v2` (22M)
|
| 23 |
+
|
| 24 |
+
## Usage
|
| 25 |
+
Run `app.py` or `smart_mcq_solver_complete.ipynb` to execute predictions.
|