diff --git a/training/scripts/prepare_dataset.py b/training/scripts/prepare_dataset.py index c81fc4d..e4e24eb 100755 --- a/training/scripts/prepare_dataset.py +++ b/training/scripts/prepare_dataset.py @@ -66,17 +66,17 @@ def prepare_dataset(input_file, output_file, test_size=0.05): test_data = formatted[split_point:] # Save - with open(output_dir / "train.jsonl", "w") as f: + with open(output_file / "train.jsonl", "w") as f: for item in train_data: f.write(json.dumps(item) + "\n") - with open(output_dir / "test.jsonl", "w") as f: + with open(output_file / "test.jsonl", "w") as f: for item in test_data: f.write(json.dumps(item) + "\n") print(f"Train: {len(train_data)} examples") print(f"Test: {len(test_data)} examples") - print(f"Saved to {output_file.parent}") + print(f"Saved to {output_file}") def main():