Skip to content

🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming

modify prompt

Yizhou Chi358a97e34cb3
diff --git a/expo/data/custom_task.py b/expo/data/custom_task.py
index f3cd433f5..032fcb8ac 100644
--- a/expo/data/custom_task.py
+++ b/expo/data/custom_task.py
@@ -21,11 +21,11 @@ COMPETITION INSTRUCTIONS
{task_description}
## More Instructions
-- output_dir: {output_dir}
-- Besides `submission.csv`, you should also save your output in the output directory.
- You should split the training data into train and dev set.
- You should use the dev set to improve your model. Print the final dev set score after training.
-- Save the prediction results of BOTH the dev set and test set in `dev_predictions.csv` and `test_predictions.csv` respectively in the output directory. They should be in the same format as the `submission.csv`.
+- output_dir: {output_dir}
+- Besides `submission.csv`, you should also save your `test_predictions.csv` and `dev_predictions.csv` in the output directory.
+- Note that `test_predictions.csv` should be identical to `submission.csv`.
- Perform data analysis, data preprocessing, feature engineering, and modeling to predict the target. {special_instruction}
**Do not make any plots or visualizations.**
"""