fix: clean, print num rows

WAcry · Mar 27, 2023 · 10db136 · 10db136
1 parent dfee696
commit 10db136
Showing 1 changed file with 2 additions and 2 deletions.
diff --git a/clean.py b/clean.py
@@ -8,7 +8,7 @@
 
 prompt_generation_dir = "raw_data_sanity_cleaned_delobotomized"
 for file in glob.glob(os.path.join(prompt_generation_dir, "*.jsonl")):
-    if "clean" in file:
+    if "clean.jsonl" in file:
         continue
     data = []
     print(file)
@@ -69,5 +69,5 @@
     print(f"Removed {prev_len - curr_len} rows")
 
     clean_name = file.split(".jsonl")[0] + "_clean.jsonl"
-    print(f"writing to {clean_name}")
+    print(f"writing to {curr_len} rows to {clean_name}")
     df.to_json(clean_name, orient="records", lines=True)