| Go through the task_descriptor.txt file to understand the task and the dataset. You can summarize it in your research logs to keep track of what all you have to do. | |
| Then fill in the provided train.py script to train a model and iterate over different models or feature selections to get a better performance. | |
| Never try to read any csv files directly. Do not forget to execute the changes you made to check for performance. |