Asset cleanup

README.md

@@ -118,12 +118,12 @@ in `assets/scientific-lit-embeddings`.
 
 Outputs (annually):
 
-- FastText vectors: `assets/{en,zh}_merged_fasttext.bin`
-- tf-idf vectors and vocab: `assets/{en,zh}_merged_tfidf.bin` and TODO
+- FastText vectors: `assets/en_merged_fasttext.bin`
+- tf-idf vectors and vocab: `assets/en_merged_tfidf.bin` and TODO
 
 Outputs (~weekly):
 
-- Preprocessed corpus: `assets/corpus/{lang}_corpus-*.jsonl.gz`
+- Preprocessed corpus: `assets/corpus/en_corpus-*.jsonl.gz`
 
 ### 2. Field taxonomy
 
@@ -143,8 +143,8 @@ created embeddings for each field.
 
 Outputs (annually):
 
-- FastText field embeddings: `assets/{en,zh}_field_fasttext.bin`
-- tf-idf field embeddings: `assets/{en,zh}_field_tfidf.bin`
+- FastText field embeddings: `assets/en_field_fasttext.bin`
+- tf-idf field embeddings: `assets/en_field_tfidf.bin`
 
 ### 4. Entity embeddings
 
@@ -154,7 +154,7 @@ generate FastText _entity embeddings_. This is documented in the `wiki-field-tex
 
 Outputs (annually):
 
-- FastText entity embeddings: `assets/{en,zh}_field_mention_fasttext.bin`
+- FastText entity embeddings: `assets/en_field_mention_fasttext.bin`
 
 ### 5. Publication embedding
 

analysis/nslp-forc/forc.ipynb

@@ -144,7 +144,7 @@
     "levels_reversed = defaultdict(set)\n",
     "children = defaultdict(list)\n",
     "parents = {}\n",
-    "with open(\"../../assets/fields/all_fields_hierarchy.jsonl\", \"r\") as json_file:\n",
+    "with open(\"all_fields_hierarchy.jsonl\", \"r\") as json_file:\n",
     "    data = [json.loads(line) for line in json_file]\n",
     "for row in data:\n",
     "    levels[row[\"child_display_name\"].lower()] = int(row[\"child_level\"])\n",

analysis/venues/evaluate.py

@@ -15,7 +15,7 @@
 
 
 def main():
-    meta = pd.read_pickle(ASSETS_DIR / "fields/fos.pkl.gz")
+    meta = pd.read_pickle("fos.pkl.gz")
     meta.index = meta.index.astype(int)
     id_to_name = meta.query("level == 0")["display_name"].to_dict()
 

analysis/venues/evaluate_l2_venues.ipynb

@@ -24,7 +24,7 @@
     "levels_reversed = defaultdict(set)\n",
     "children = defaultdict(list)\n",
     "parents = {}\n",
-    "with open(\"../../assets/fields/all_fields_hierarchy.jsonl\", \"r\") as json_file:\n",
+    "with open(\"all_fields_hierarchy.jsonl\", \"r\") as json_file:\n",
     "    data = [json.loads(line) for line in json_file]\n",
     "for row in data:\n",
     "    levels[row[\"child_display_name\"].lower()] = int(row[\"child_level\"])\n",

analysis/venues/evaluate_venues.ipynb

@@ -13,7 +13,7 @@
     "\n",
     "from fos.settings import ASSETS_DIR\n",
     "\n",
-    "meta = pd.read_pickle(ASSETS_DIR / \"fields/fos.pkl.gz\")\n",
+    "meta = pd.read_pickle(\"fos.pkl.gz\")\n",
     "meta.index = meta.index.astype(int)\n",
     "id_to_name = meta.query(\"level == 0\")[\"display_name\"].to_dict()\n",
     "\n",
@@ -676,4 +676,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 0
-}
+}