Allows for caching of UD/German-HDT dataset; increase pct precision

zentrum-lexikographie · Nov 26, 2024 · 297dea2 · 297dea2
1 parent f57d2cc
commit 297dea2
Showing 1 changed file with 1 addition and 2 deletions.
diff --git a/dwdsmor/build/benchmark.py b/dwdsmor/build/benchmark.py
@@ -16,7 +16,6 @@ def ud_de_hdt_tokens():
         "universal_dependencies",
         "de_hdt",
         split="train",
-        streaming=True,
         trust_remote_code=True,
     )
     for s in sentences:
@@ -189,7 +188,7 @@ def compute_coverage(automata, limit=None, show_progress=False):
     args = arg_parser.parse_args()
 
     def pct(v):
-        return "{:.2%}".format(v)
+        return "{:.3%}".format(v)
 
     def count(v):
         return "{:,d}".format(v)