create_dataset_from_tsv.py

import pandas as pd
import xmltodict
from sklearn.model_selection import train_test_split
import glob
import sys
import os

filelist = glob.glob('tsv_source_target/*.tsv')

data = pd.DataFrame()

for tsvfile in filelist:
    tmp = pd.read_csv(tsvfile, sep='\t')
    tmp.columns=['source','target']
    tmp['rev_source'] = tmp['target']
    tmp['rev_target'] = tmp['source']


    path = tsvfile.split("/")
    source = path[1][0:3]
    target = path[1][3:6]
    
    prefix = f"{source}_{target}: "
    tmp['source'] = prefix + tmp['source']
    
    rev_prefix = f"{target}_{source}: "
    tmp['rev_source'] = rev_prefix + tmp['rev_source']

    data = pd.concat([data,tmp])


#Shuffle
data = data.sample(frac=1).reset_index(drop=True)

# Add both directions
original = data[['source','target']]
reverse = data[['rev_source','rev_target']]
reverse.columns=['source','target']

data = pd.concat([original,reverse])
data = data.sample(frac=1).reset_index(drop=True)

# Train - test - dev
train, test = train_test_split(data, test_size=0.2)
test, dev = train_test_split(test, test_size=0.5)

# Write the datasets to disk
train.to_csv('tsv_all_source_target/train.tsv', index=False, header=False, sep='\t')
test.to_csv('tsv_all_source_target/test.tsv', index=False, header=False, sep='\t')
dev.to_csv('tsv_all_source_target/dev.tsv', index=False, header=False, sep='\t')


print("Finished")