Merge branch 'main' of github.com:ChenKy23/semeval2024-Task7

ChenKy23 · Feb 21, 2024 · de58d17 · de58d17
2 parents be51fc9 + 475a02e
commit de58d17
Show file tree

Hide file tree

Showing 3 changed files with 5 additions and 5 deletions.
diff --git a/README.md b/README.md
@@ -3,7 +3,7 @@
 The datasets can be found at <https://sites.google.com/view/numeval/data>. Please place them in the respective task directories, and refer to the input parameters for detailed configuration.
 
 ## Install Requirements
-- `Python version >= 3.8`
+- `Python version >= 3.9`
 - `PyTorch version >= 2.0.0`
 
 Install the package required:

diff --git a/subtask1/instruction_tuning_qqa.py b/subtask1/instruction_tuning_qqa.py
@@ -207,7 +207,7 @@ def preprocess_function(sample):
 if __name__ == '__main__':
     parser = argparse.ArgumentParser(description="training code")
     parser.add_argument("--data_train_pth", default='./Quantitative-101/QQA/QQA_train.json', help="dataset_train's path")
-    parser.add_argument("--data_dev_pth", default='./Quantitative-101/QQA/QQA_test.json', help="dataset_dev's path")
+    parser.add_argument("--data_dev_pth", default='./Quantitative-101/QQA/QQA_dev.json', help="dataset_dev's path")
     parser.add_argument("--data_test_pth", default='./Quantitative-101/QQA/QQA_test.json', help="dataset_test's path")
     parser.add_argument("--is_digit_base", default=False, help="whether to use digit")
     parser.add_argument("--has_dev", default=True, help="whether has dev dataset")
@@ -230,4 +230,4 @@ def preprocess_function(sample):
     parser.add_argument("--output_dir", default="save_res", help="output file's dir")
     args = parser.parse_args()
 
-    run(args)
+    run(args)
diff --git a/subtask2/instruction_tuning_nsqad.py b/subtask2/instruction_tuning_nsqad.py
@@ -297,7 +297,7 @@ def preprocess_function(sample):
     datasets = DatasetDict()
 
     if args.task == "train":
-        dataset_train = read_jsonl(data_train_pth)[0][:1000]
+        dataset_train = read_jsonl(data_train_pth)[0]
         data_split = int(len(dataset_train)*0.9)
         random.seed(args.seed)
         random.shuffle(dataset_train)
@@ -343,4 +343,4 @@ def preprocess_function(sample):
     parser.add_argument("--output_dir", default="save_res", help="output file's dir")
     args = parser.parse_args()
 
-    run(args)
+    run(args)