Exclude paths with node_modules

test.jsonl.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a60e9e12cc5c73a66b9c14b8fdd368b534dc7235bb12165043f658cccc7f5b1
-size 1074453
+oid sha256:fe7a4c72d97e1f21db58f9b80d0ff8c887663ffb8db9ec1adc485f18a4e17885
+size 5921647

train.jsonl.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4cf2c9605a9cbd1cb8ecfb1bdaafd52c337ed47489588256c45a487c9058aff7
-size 38163445
+oid sha256:1dfa446eba5f93b09f58efed81f9114ddaf973c1696d0ce75efb4f5e492dbefb
+size 33754693

train_split.py

@@ -16,7 +16,11 @@ import Levenshtein
 
 def files_list():
     data_path = Path("valid_data")
-    files = [f for f in data_path.rglob("*.json") if f.is_file()]
+    files = [
+        f
+        for f in data_path.rglob("*.json")
+        if f.is_file() and "node_modules" not in f.parts
+    ]
     return files
 
 
@@ -133,7 +137,7 @@ def main(similarity, split, seed, commits_file):
 if __name__ == "__main__":
     parser = argparse.ArgumentParser()
     parser.add_argument("--similarity", default=None, type=float)
-    parser.add_argument("--seed", default=94, type=int)
+    parser.add_argument("--seed", default=112, type=int)
     parser.add_argument("--split", default=0.8, type=float)
     parser.add_argument("--commits_file", default="commits.json")
     args = parser.parse_args()

validation.jsonl.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab8a6ee8067de1eae9582f7cff274b5aab9f9719460471fe45283832ae8e3abd
-size 5578087
+oid sha256:9bb8f7a2c01c670a7bba56ecea93e2d46fe2b80304f7bac422b92decb13e69a9
+size 3369671