Datasets:
create
Browse files- create_stargate_dataset.py +39 -0
create_stargate_dataset.py
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from datasets import Dataset
|
| 2 |
+
import json
|
| 3 |
+
|
| 4 |
+
# Load or define your metadata
|
| 5 |
+
sample_data = [
|
| 6 |
+
{
|
| 7 |
+
"filename": "CIA-RDP79-00999A000200010002-3.pdf",
|
| 8 |
+
"document_id": "CIA-RDP79-00999A000200010002-3",
|
| 9 |
+
"page_count": 10,
|
| 10 |
+
"image_count": 10,
|
| 11 |
+
"processed_at": "2025-03-22T22:04:19.172208",
|
| 12 |
+
"ocr_status": "pending",
|
| 13 |
+
"text_extracted": False,
|
| 14 |
+
"source": "CIA Stargate Archive",
|
| 15 |
+
"tags": [],
|
| 16 |
+
"pdf": None # Add actual file path or binary later
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"filename": "CIA-RDP96-00788R000700020002-2.pdf",
|
| 20 |
+
"document_id": "CIA-RDP96-00788R000700020002-2",
|
| 21 |
+
"page_count": 14,
|
| 22 |
+
"image_count": 14,
|
| 23 |
+
"processed_at": "2025-03-22T22:10:03.984102",
|
| 24 |
+
"ocr_status": "pending",
|
| 25 |
+
"text_extracted": False,
|
| 26 |
+
"source": "CIA Stargate Archive",
|
| 27 |
+
"tags": ["remote_viewing", "tasking_response"],
|
| 28 |
+
"pdf": None
|
| 29 |
+
}
|
| 30 |
+
]
|
| 31 |
+
|
| 32 |
+
# Create the dataset
|
| 33 |
+
dataset = Dataset.from_list(sample_data)
|
| 34 |
+
|
| 35 |
+
# Save it locally as a Hugging Face Arrow dataset
|
| 36 |
+
dataset.save_to_disk("stargate_raw_pdfs")
|
| 37 |
+
|
| 38 |
+
# (Optional) Preview
|
| 39 |
+
print(dataset)
|