Update tagged grants

tag.py

@@ -1,6 +1,7 @@
 import json
 
 from transformers import AutoModel, AutoTokenizer
+from tqdm import tqdm
 import srsly
 import typer
 
@@ -22,11 +23,14 @@ def tag(data_path, tagged_data_path, sample_size: int = 10):
     )
 
     texts = [grant["title_and_description"] for grant in data]
-    inputs = tokenizer(texts, padding="max_length")
-    labels = model(**inputs, return_labels=True)
+    for batch_index in tqdm(range(0, len(texts), 10)):
+        batch_texts = texts[batch_index:batch_index+10]
 
-    for i, tags in enumerate(labels):
-        data[i]["tags"] = tags
+        inputs = tokenizer(batch_texts, padding="max_length")
+        labels = model(**inputs, return_labels=True)
+
+        for i, tags in enumerate(labels):
+            data[i]["tags"] = tags
 
     srsly.write_jsonl(tagged_data_path, data)