Rok Shimizu commited on
Add model card
Browse files
README.md
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- sklearn
|
| 5 |
+
- tabular-classification
|
| 6 |
+
- excel
|
| 7 |
+
- spreadsheet
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
# sheet-cell-classifier
|
| 11 |
+
|
| 12 |
+
RandomForest cell role classifier for Excel spreadsheets, used by
|
| 13 |
+
[sheet-call-tree](https://github.com/roksechs/sheet-call-tree).
|
| 14 |
+
|
| 15 |
+
## Model
|
| 16 |
+
|
| 17 |
+
Predicts whether a spreadsheet cell is a **header** (0) or **data** (1).
|
| 18 |
+
Trained on CTC (CIUS + SAUS) and ENTRANT datasets.
|
| 19 |
+
|
| 20 |
+
**Features (23):** gap-proximity (dist_above, dist_left), row/col numeric fractions,
|
| 21 |
+
format fields (bold, italic, colors, borders, alignment, data type), value type flags.
|
| 22 |
+
|
| 23 |
+
## Usage
|
| 24 |
+
|
| 25 |
+
```python
|
| 26 |
+
from huggingface_hub import hf_hub_download
|
| 27 |
+
import joblib
|
| 28 |
+
|
| 29 |
+
model_path = hf_hub_download(repo_id="RokRok/sheet-cell-classifier", filename="cell_classifier.joblib")
|
| 30 |
+
clf = joblib.load(model_path)
|
| 31 |
+
```
|
| 32 |
+
|
| 33 |
+
`sheet-call-tree >= 0.1.2` downloads and uses this model automatically.
|