|
import pandas as pd |
|
|
|
n = 2 |
|
total_n = 16 |
|
|
|
|
|
csv_files = [ |
|
f"en-wikihow-qa-dataset-{n}k.csv", |
|
f"ru-wikihow-qa-dataset-{n}k.csv", |
|
f"pt-wikihow-qa-dataset-{n}k.csv", |
|
f"nl-wikihow-qa-dataset-{n}k.csv", |
|
f"it-wikihow-qa-dataset-{n}k.csv", |
|
f"fr-wikihow-qa-dataset-{n}k.csv", |
|
f"es-wikihow-qa-dataset-{n}k.csv", |
|
f"de-wikihow-qa-dataset-{n}k.csv" |
|
] |
|
|
|
|
|
merged_data = pd.DataFrame() |
|
|
|
|
|
for file in csv_files: |
|
data = pd.read_csv(file) |
|
merged_data = pd.concat([merged_data, data], ignore_index=True) |
|
|
|
|
|
merged_data.to_csv(f"multilingial-wikihow-qa-dataset-{total_n}k.csv", index=False) |
|
merged_data.to_parquet(f"multilingial-wikihow-qa-dataset-{total_n}k.parquet", row_group_size=100, engine="pyarrow", index=False) |