From 0e5b13f23eb9cf4818a219c2f2a20edc231065e8 Mon Sep 17 00:00:00 2001
From: t-yuxuanlei <t-yuxuanlei@microsoft.com>
Date: Wed, 18 Dec 2024 04:28:51 +0000
Subject: [PATCH] Fix RecExplainer bug: Automatically create directory if it
 does not exist

---
 RecExplainer/preprocess/amazon_generate_v3.py             | 2 ++
 RecExplainer/preprocess/data_preprocess_amazon.py         | 1 +
 RecExplainer/preprocess/eval_explan.py                    | 2 ++
 RecExplainer/preprocess/gpt_api.py                        | 1 +
 RecExplainer/preprocess/mf_amazon_video_games_generate.py | 2 ++
 RecExplainer/preprocess/unirec_utils/unirec_split.py      | 2 ++
 6 files changed, 10 insertions(+)

diff --git a/RecExplainer/preprocess/amazon_generate_v3.py b/RecExplainer/preprocess/amazon_generate_v3.py
index c625610..66d36df 100644
--- a/RecExplainer/preprocess/amazon_generate_v3.py
+++ b/RecExplainer/preprocess/amazon_generate_v3.py
@@ -189,8 +189,10 @@ def gen_uid2next_rank(user_items, args):
     tokenizer = AutoTokenizer.from_pretrained(args.model_name, model_max_length=args.model_max_length, use_fast=True)
 
     if args.gpt_query_file is not None:
+        os.makedirs(os.path.dirname(args.gpt_query_file), exist_ok=True)
         gen_uid2summary(user_items, meta_infos, args)
     else:
+        os.makedirs(os.path.dirname(args.save_intention_file), exist_ok=True)
         train_intention = []
         train_behavior = []
         train_both = []
diff --git a/RecExplainer/preprocess/data_preprocess_amazon.py b/RecExplainer/preprocess/data_preprocess_amazon.py
index 0016315..c932a24 100644
--- a/RecExplainer/preprocess/data_preprocess_amazon.py
+++ b/RecExplainer/preprocess/data_preprocess_amazon.py
@@ -383,4 +383,5 @@ def main_process(data_name, args, data_type='Amazon'):
 
 if __name__ == '__main__':
     args = parse_args()
+    os.makedirs(os.path.dirname(args.save_data_file), exist_ok=True)
     main_process(args.full_data_name, args=args, data_type='Amazon')
\ No newline at end of file
diff --git a/RecExplainer/preprocess/eval_explan.py b/RecExplainer/preprocess/eval_explan.py
index 65c62e1..baf47bb 100644
--- a/RecExplainer/preprocess/eval_explan.py
+++ b/RecExplainer/preprocess/eval_explan.py
@@ -3,6 +3,7 @@
 
 import pandas as pd
 import argparse
+import os
 
 def parse_args():
     parser = argparse.ArgumentParser(description="data process")
@@ -81,6 +82,7 @@ def eval_metric(args):
 if __name__ == '__main__':
     args = parse_args()
     if args.judge_query_file is not None:
+        os.makedirs(os.path.dirname(args.judge_query_file), exist_ok=True)
         eval_data_gen(args)
     else:
         eval_metric(args)
\ No newline at end of file
diff --git a/RecExplainer/preprocess/gpt_api.py b/RecExplainer/preprocess/gpt_api.py
index 7eef694..6699b13 100644
--- a/RecExplainer/preprocess/gpt_api.py
+++ b/RecExplainer/preprocess/gpt_api.py
@@ -165,4 +165,5 @@ def main(args):
     parser.add_argument("--output_file", type=str)
     parser.add_argument("--input_columns", type=str, default="question")
     args = parser.parse_args()
+    os.makedirs(os.path.dirname(args.output_file), exist_ok=True)
     main(args)
diff --git a/RecExplainer/preprocess/mf_amazon_video_games_generate.py b/RecExplainer/preprocess/mf_amazon_video_games_generate.py
index 7ba6d53..f3648bc 100644
--- a/RecExplainer/preprocess/mf_amazon_video_games_generate.py
+++ b/RecExplainer/preprocess/mf_amazon_video_games_generate.py
@@ -191,8 +191,10 @@ def gen_uid2next_rank(user_items, args):
     tokenizer = AutoTokenizer.from_pretrained(args.model_name, model_max_length=args.model_max_length, use_fast=True)
 
     if args.gpt_query_file is not None:
+        os.makedirs(os.path.dirname(args.gpt_query_file), exist_ok=True)
         gen_uid2summary(user_items, meta_infos, args)
     else:
+        os.makedirs(os.path.dirname(args.save_intention_file), exist_ok=True)
         train_intention = []
         train_behavior = []
         train_both = []
diff --git a/RecExplainer/preprocess/unirec_utils/unirec_split.py b/RecExplainer/preprocess/unirec_utils/unirec_split.py
index 98b7548..e6c0208 100644
--- a/RecExplainer/preprocess/unirec_utils/unirec_split.py
+++ b/RecExplainer/preprocess/unirec_utils/unirec_split.py
@@ -3,6 +3,7 @@
 
 import pandas as pd
 import argparse
+import os
 
 def parse_args():
     parser = argparse.ArgumentParser(description="data process")
@@ -64,5 +65,6 @@ def rank_split(args):
 
 if __name__ == '__main__':
     args = parse_args()
+    os.makedirs(os.path.dirname(args.train_file), exist_ok=True)
     unirec_split(args)
     rank_split(args)
\ No newline at end of file