From 03c824c29c1311b69de5bca1bd444c72c96afa74 Mon Sep 17 00:00:00 2001
From: Daniel Han <danielhanchen@gmail.com>
Date: Tue, 29 Apr 2025 23:57:46 -0700
Subject: [PATCH] Update synthetic.py

---
 unsloth/dataprep/synthetic.py | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/unsloth/dataprep/synthetic.py b/unsloth/dataprep/synthetic.py
index f76e170c59..5eadacf448 100644
--- a/unsloth/dataprep/synthetic.py
+++ b/unsloth/dataprep/synthetic.py
@@ -238,6 +238,7 @@ prompts:
 
 def configure_synthetic_data_kit(
     model_name = "unsloth/Llama-3.1-8B-Instruct-unsloth-bnb-4bit",
+    output_folder = "synthetic_data_output",
     temperature = 0.7,
     top_p = 0.95,
     chunk_size = 4000,
@@ -248,6 +249,12 @@ def configure_synthetic_data_kit(
     cleanup_batch_size = 4,
     cleanup_temperature = 0.3,
 ):
+    locations = "pdf,html,youtube,docx,ppt,txt,output,generated,cleaned,final"
+    locations = locations.split(",")
+    for path in locations:
+        os.makedirs(os.path.join(output_folder, path), exist_ok = True)
+    pass
+    
     config = synthetic_config_string\
         .replace("{model_name}", str(model_name))\
         .replace("{temperature}", str(temperature))\