| import pandas as pd | |
| from datasets import load_dataset | |
| dataset_remote = load_dataset("parquet", data_files={'test': '/tmp/data/data/test-00000-of-00001.parquet'}) | |
| out = [] | |
| for id in dataset_remote["test"]["id"]: | |
| out.append(dict(id = id, pred = "flux")) | |
| pd.DataFrame(out).to_csv("submission.csv",index = False) | |