citations

mms_benchmark/_modidx.py

@@ -5,4 +5,4 @@
                 'doc_host': 'https://Brand24-AI.github.io/mms_benchmark',
                 'git_url': 'https://github.com/Brand24-AI/mms_benchmark',
                 'lib_path': 'mms_benchmark'},
-  'syms': {}}
+  'syms': {'mms_benchmark.citations': {}}}

mms_benchmark/citations.py

@@ -0,0 +1,16 @@
+from typing import Dict, Union
+
+import bibtexparser
+from datasets import DatasetDict
+
+
+def get_citations(dataset: DatasetDict, citation_as_dict: bool = True) -> Dict[str, Union[Dict, str]]:
+    lines = dataset.citation.split("% Datasets: ")[1:]
+    original_dataset_to_bibtex = {}
+    for line in lines:
+        original_datasets, citation = line.split("\n", maxsplit=1)
+        dataset_list = original_datasets.split(", ")
+
+        for dataset in dataset_list:
+            original_dataset_to_bibtex[dataset] = bibtexparser.loads(citation).entries[0] if citation_as_dict else citation
+    return original_dataset_to_bibtex

nbs/00_dataset_card.ipynb

@@ -13,6 +13,16 @@
     "- bibliography: references.bib\n"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#| hide\n",
+    "from nbdev.showdoc import *"
+   ]
+  },
   {
    "attachments": {},
    "cell_type": "markdown",
@@ -518,17 +528,11 @@
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": null,
+   "attachments": {},
+   "cell_type": "markdown",
    "metadata": {},
-   "outputs": [],
    "source": [
-    "#| eval: false\n",
-    "from typing import Dict, Union\n",
-    "\n",
-    "import bibtexparser\n",
-    "\n",
-    "from datasets import DatasetDict"
+    "We can load citations as strings - easy adding to bibtex."
    ]
   },
   {
@@ -537,25 +541,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "#| eval: false\n",
-    "def get_citations(dataset: DatasetDict, citation_as_dict: bool = True) -> Dict[str, Union[Dict, str]]:\n",
-    "    lines = dataset.citation.split(\"% Datasets: \")[1:]\n",
-    "    original_dataset_to_bibtex = {}\n",
-    "    for line in lines:\n",
-    "        original_datasets, citation = line.split(\"\\n\", maxsplit=1)\n",
-    "        dataset_list = original_datasets.split(\", \")\n",
-    "\n",
-    "        for dataset in dataset_list:\n",
-    "            original_dataset_to_bibtex[dataset] = bibtexparser.loads(citation).entries[0] if citation_as_dict else citation\n",
-    "    return original_dataset_to_bibtex"
-   ]
-  },
-  {
-   "attachments": {},
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "We can load citations as strings - easy adding to bibtex."
+    "from mms_benchmark.citations import get_citations"
    ]
   },
   {