| # NOTE: run with different random seed | |
| import datasets | |
| import pandas as pd | |
| data = datasets.load_dataset('program_synthesis.py', 'text')['train'] | |
| data.to_csv('_t.text.csv', index=False) | |
| data = datasets.load_dataset('program_synthesis.py', 'list')['train'] | |
| data.to_csv('_t.list.csv', index=False) | |
| df_list = pd.read_csv('_t.list.csv') | |
| df_text = pd.read_csv('_t.text.csv') | |
| df = pd.concat((df_list, df_text)) | |
| df = df.sample(frac=1) | |
| df.to_csv('sample_test.csv', index=False) | |