Update README.md
Browse files
README.md
CHANGED
|
@@ -1,61 +1,56 @@
|
|
| 1 |
---
|
| 2 |
-
|
| 3 |
license: mit
|
| 4 |
datasets:
|
| 5 |
- Lyon28/datasets-caca-3500
|
| 6 |
language:
|
| 7 |
- id
|
| 8 |
tags:
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
---
|
| 16 |
-
|
| 17 |
-
# Chatbot Caca - Retrieval-Based QA
|
| 18 |
-
|
| 19 |
-
Chatbot berbasis BM25 + TF-IDF untuk QA Bahasa Indonesia.
|
| 20 |
|
| 21 |
-
|
| 22 |
|
| 23 |
-
|
| 24 |
-
- **Size:** 2.69 MB
|
| 25 |
-
- **Algorithm:** Hybrid BM25 + TF-IDF + Fuzzy Matching
|
| 26 |
-
- **Dataset:** datasets-caca-3500 (3,500 QA pairs)
|
| 27 |
-
- **Language:** Indonesian
|
| 28 |
|
| 29 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
|
| 31 |
-
|
| 32 |
-
# Install dependencies
|
| 33 |
-
!pip install rank-bm25 scikit-learn huggingface-hub
|
| 34 |
|
| 35 |
-
|
| 36 |
-
|
|
|
|
| 37 |
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
filename="chatbot_caca.pkl"
|
| 41 |
-
)
|
| 42 |
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
|
| 48 |
-
|
| 49 |
-
|
|
|
|
|
|
|
| 50 |
|
| 51 |
-
|
|
|
|
| 52 |
|
| 53 |
-
|
| 54 |
-
- Accuracy: High for paraphrase matching
|
| 55 |
-
- Memory: ~3MB
|
| 56 |
|
| 57 |
-
|
|
|
|
|
|
|
| 58 |
|
| 59 |
-
|
| 60 |
|
| 61 |
-
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
license: mit
|
| 3 |
datasets:
|
| 4 |
- Lyon28/datasets-caca-3500
|
| 5 |
language:
|
| 6 |
- id
|
| 7 |
tags:
|
| 8 |
+
- retrieval
|
| 9 |
+
- qa
|
| 10 |
+
- indonesian
|
| 11 |
+
- bm25
|
| 12 |
+
- tfidf
|
|
|
|
| 13 |
---
|
| 14 |
+
# Chatbot Caca - Retrieval-Based QA
|
|
|
|
|
|
|
|
|
|
| 15 |
|
| 16 |
+
Chatbot berbasis BM25 + TF-IDF untuk QA Bahasa Indonesia.
|
| 17 |
|
| 18 |
+
## Model Details
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
|
| 20 |
+
- **Type:** Retrieval-based QA System
|
| 21 |
+
- **Size:** 2.69 MB
|
| 22 |
+
- **Algorithm:** Hybrid BM25 + TF-IDF + Fuzzy Matching
|
| 23 |
+
- **Dataset:** datasets-caca-3500 (3,500 QA pairs)
|
| 24 |
+
- **Language:** Indonesian
|
| 25 |
|
| 26 |
+
## Usage
|
|
|
|
|
|
|
| 27 |
|
| 28 |
+
```python
|
| 29 |
+
# Install dependencies
|
| 30 |
+
!pip install rank-bm25 scikit-learn huggingface-hub
|
| 31 |
|
| 32 |
+
# Download model
|
| 33 |
+
from huggingface_hub import hf_hub_download
|
|
|
|
|
|
|
| 34 |
|
| 35 |
+
model_path = hf_hub_download(
|
| 36 |
+
repo_id="Lyon28/caca-based-chatbot",
|
| 37 |
+
filename="chatbot_caca.pkl"
|
| 38 |
+
)
|
| 39 |
|
| 40 |
+
# Load model
|
| 41 |
+
import pickle
|
| 42 |
+
with open(model_path, 'rb') as f:
|
| 43 |
+
data = pickle.load(f)
|
| 44 |
|
| 45 |
+
print(f"Loaded {len(data['qa_pairs'])} QA pairs!")
|
| 46 |
+
```
|
| 47 |
|
| 48 |
+
## Performance
|
|
|
|
|
|
|
| 49 |
|
| 50 |
+
- Query speed: < 10ms
|
| 51 |
+
- Accuracy: High for paraphrase matching
|
| 52 |
+
- Memory: ~3MB
|
| 53 |
|
| 54 |
+
## Credits
|
| 55 |
|
| 56 |
+
Created by Lyon28
|