Spaces:
Runtime error
Runtime error
Update download_data.py
Browse files- download_data.py +18 -0
download_data.py
CHANGED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from datasets import load_dataset
|
| 2 |
+
import os
|
| 3 |
+
|
| 4 |
+
# Load from HF
|
| 5 |
+
dataset = load_dataset("Azu/Handwritten-Mathematical-Expression-Convert-LaTeX")
|
| 6 |
+
|
| 7 |
+
os.makedirs("dataset/train", exist_ok=True)
|
| 8 |
+
os.makedirs("dataset/val", exist_ok=True)
|
| 9 |
+
|
| 10 |
+
# Split: 90% train, 10% val
|
| 11 |
+
for i, row in enumerate(dataset["train"]):
|
| 12 |
+
subset = "val" if i % 10 == 0 else "train"
|
| 13 |
+
image_path = f"dataset/{subset}/{i:05d}.png"
|
| 14 |
+
latex_path = f"dataset/{subset}/{i:05d}.txt"
|
| 15 |
+
|
| 16 |
+
row["image"].save(image_path)
|
| 17 |
+
with open(latex_path, "w", encoding="utf-8") as f:
|
| 18 |
+
f.write(row["label"])
|