Spaces:
Sleeping
Sleeping
Commit
·
4ce2e5d
1
Parent(s):
80fa8ee
feat(utils): Create utils, add get unique documents function
Browse files
utils.py
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
'''This module contains utility functions for the project'''
|
| 2 |
+
import mmh3
|
| 3 |
+
from haystack import Document
|
| 4 |
+
|
| 5 |
+
|
| 6 |
+
def get_unique_docs(dataset):
|
| 7 |
+
'''Get unique documents from dataset
|
| 8 |
+
|
| 9 |
+
Args:
|
| 10 |
+
dataset: list of dictionaries
|
| 11 |
+
|
| 12 |
+
Returns:
|
| 13 |
+
docs: list of haystack.Document
|
| 14 |
+
'''
|
| 15 |
+
unique_docs = set()
|
| 16 |
+
docs = list()
|
| 17 |
+
for doc in dataset:
|
| 18 |
+
if doc["context"] is not None and doc["context_id"] not in unique_docs:
|
| 19 |
+
unique_docs.add(doc["context_id"])
|
| 20 |
+
document = Document(content=doc["context"], meta={'title': doc["context_title"], 'context_id': doc["context_id"]})
|
| 21 |
+
docs.append(document)
|
| 22 |
+
return docs
|