Commit ac98d9e1 authored by lintangsutawika's avatar lintangsutawika
Browse files

xnli reformat

parent 83683ddd
...@@ -13,91 +13,91 @@ LANGUAGES = { ...@@ -13,91 +13,91 @@ LANGUAGES = {
"QUESTION_WORD": "صحيح", "QUESTION_WORD": "صحيح",
"ENTAILMENT_LABEL": "نعم", "ENTAILMENT_LABEL": "نعم",
"NEUTRAL_LABEL": "لذا", "NEUTRAL_LABEL": "لذا",
"CONTRADICTION_LABEL": "رقم" "CONTRADICTION_LABEL": "رقم",
}, },
"bg": { # Bulgarian "bg": { # Bulgarian
"QUESTION_WORD": "правилно", "QUESTION_WORD": "правилно",
"ENTAILMENT_LABEL": "да", "ENTAILMENT_LABEL": "да",
"NEUTRAL_LABEL": "така", "NEUTRAL_LABEL": "така",
"CONTRADICTION_LABEL": "не" "CONTRADICTION_LABEL": "не",
}, },
"de": { # German "de": { # German
"QUESTION_WORD": "richtig", "QUESTION_WORD": "richtig",
"ENTAILMENT_LABEL": "Ja", "ENTAILMENT_LABEL": "Ja",
"NEUTRAL_LABEL": "Auch", "NEUTRAL_LABEL": "Auch",
"CONTRADICTION_LABEL": "Nein" "CONTRADICTION_LABEL": "Nein",
}, },
"el": { # Greek "el": { # Greek
"QUESTION_WORD": "σωστός", "QUESTION_WORD": "σωστός",
"ENTAILMENT_LABEL": "Ναί", "ENTAILMENT_LABEL": "Ναί",
"NEUTRAL_LABEL": "Έτσι", "NEUTRAL_LABEL": "Έτσι",
"CONTRADICTION_LABEL": "όχι" "CONTRADICTION_LABEL": "όχι",
}, },
"en": { # English "en": { # English
"QUESTION_WORD": "right", "QUESTION_WORD": "right",
"ENTAILMENT_LABEL": "Yes", "ENTAILMENT_LABEL": "Yes",
"NEUTRAL_LABEL": "Also", "NEUTRAL_LABEL": "Also",
"CONTRADICTION_LABEL": "No" "CONTRADICTION_LABEL": "No",
}, },
"es": { # Spanish "es": { # Spanish
"QUESTION_WORD": "correcto", "QUESTION_WORD": "correcto",
"ENTAILMENT_LABEL": "Sí", "ENTAILMENT_LABEL": "Sí",
"NEUTRAL_LABEL": "Asi que", "NEUTRAL_LABEL": "Asi que",
"CONTRADICTION_LABEL": "No" "CONTRADICTION_LABEL": "No",
}, },
"fr": { # French "fr": { # French
"QUESTION_WORD": "correct", "QUESTION_WORD": "correct",
"ENTAILMENT_LABEL": "Oui", "ENTAILMENT_LABEL": "Oui",
"NEUTRAL_LABEL": "Aussi", "NEUTRAL_LABEL": "Aussi",
"CONTRADICTION_LABEL": "Non" "CONTRADICTION_LABEL": "Non",
}, },
"hi": { # Hindi "hi": { # Hindi
"QUESTION_WORD": "सही", "QUESTION_WORD": "सही",
"ENTAILMENT_LABEL": "हाँ", "ENTAILMENT_LABEL": "हाँ",
"NEUTRAL_LABEL": "इसलिए", "NEUTRAL_LABEL": "इसलिए",
"CONTRADICTION_LABEL": "नहीं" "CONTRADICTION_LABEL": "नहीं",
}, },
"ru": { # Russian "ru": { # Russian
"QUESTION_WORD": "правильно", "QUESTION_WORD": "правильно",
"ENTAILMENT_LABEL": "Да", "ENTAILMENT_LABEL": "Да",
"NEUTRAL_LABEL": "Так", "NEUTRAL_LABEL": "Так",
"CONTRADICTION_LABEL": "Нет" "CONTRADICTION_LABEL": "Нет",
}, },
"sw": { # Swahili "sw": { # Swahili
"QUESTION_WORD": "sahihi", "QUESTION_WORD": "sahihi",
"ENTAILMENT_LABEL": "Ndiyo", "ENTAILMENT_LABEL": "Ndiyo",
"NEUTRAL_LABEL": "Hivyo", "NEUTRAL_LABEL": "Hivyo",
"CONTRADICTION_LABEL": "Hapana" "CONTRADICTION_LABEL": "Hapana",
}, },
"th": { # Thai "th": { # Thai
"QUESTION_WORD": "ถูกต้อง", "QUESTION_WORD": "ถูกต้อง",
"ENTAILMENT_LABEL": "ใช่", "ENTAILMENT_LABEL": "ใช่",
"NEUTRAL_LABEL": "ดังนั้น", "NEUTRAL_LABEL": "ดังนั้น",
"CONTRADICTION_LABEL": "ไม่" "CONTRADICTION_LABEL": "ไม่",
}, },
"tr": { # Turkish "tr": { # Turkish
"QUESTION_WORD": "doğru", "QUESTION_WORD": "doğru",
"ENTAILMENT_LABEL": "Evet", "ENTAILMENT_LABEL": "Evet",
"NEUTRAL_LABEL": "Böylece", "NEUTRAL_LABEL": "Böylece",
"CONTRADICTION_LABEL": "Hayır" "CONTRADICTION_LABEL": "Hayır",
}, },
"ur": { # Urdu "ur": { # Urdu
"QUESTION_WORD": "صحیح", "QUESTION_WORD": "صحیح",
"ENTAILMENT_LABEL": "جی ہاں", "ENTAILMENT_LABEL": "جی ہاں",
"NEUTRAL_LABEL": "اس لئے", "NEUTRAL_LABEL": "اس لئے",
"CONTRADICTION_LABEL": "نہیں" "CONTRADICTION_LABEL": "نہیں",
}, },
"vi": { # Vietnamese "vi": { # Vietnamese
"QUESTION_WORD": "đúng", "QUESTION_WORD": "đúng",
"ENTAILMENT_LABEL": "Vâng", "ENTAILMENT_LABEL": "Vâng",
"NEUTRAL_LABEL": "Vì vậy", "NEUTRAL_LABEL": "Vì vậy",
"CONTRADICTION_LABEL": "Không" "CONTRADICTION_LABEL": "Không",
}, },
"zh": { # Chinese "zh": { # Chinese
"QUESTION_WORD": "正确", "QUESTION_WORD": "正确",
"ENTAILMENT_LABEL": "是的", "ENTAILMENT_LABEL": "是的",
"NEUTRAL_LABEL": "所以", "NEUTRAL_LABEL": "所以",
"CONTRADICTION_LABEL": "不是的" "CONTRADICTION_LABEL": "不是的",
}, },
} }
...@@ -117,7 +117,9 @@ def gen_lang_yamls(output_dir: str, overwrite: bool) -> None: ...@@ -117,7 +117,9 @@ def gen_lang_yamls(output_dir: str, overwrite: bool) -> None:
ENTAILMENT_LABEL = LANGUAGES[lang]["ENTAILMENT_LABEL"] ENTAILMENT_LABEL = LANGUAGES[lang]["ENTAILMENT_LABEL"]
NEUTRAL_LABEL = LANGUAGES[lang]["NEUTRAL_LABEL"] NEUTRAL_LABEL = LANGUAGES[lang]["NEUTRAL_LABEL"]
CONTRADICTION_LABEL = LANGUAGES[lang]["CONTRADICTION_LABEL"] CONTRADICTION_LABEL = LANGUAGES[lang]["CONTRADICTION_LABEL"]
with open(f"{output_dir}/{file_name}", "w" if overwrite else "x", encoding='utf8') as f: with open(
f"{output_dir}/{file_name}", "w" if overwrite else "x", encoding="utf8"
) as f:
f.write("# Generated by utils.py\n") f.write("# Generated by utils.py\n")
yaml.dump( yaml.dump(
{ {
...@@ -129,10 +131,10 @@ def gen_lang_yamls(output_dir: str, overwrite: bool) -> None: ...@@ -129,10 +131,10 @@ def gen_lang_yamls(output_dir: str, overwrite: bool) -> None:
f"""premise+\", {QUESTION_WORD}? {ENTAILMENT_LABEL}, \"+hypothesis,""" f"""premise+\", {QUESTION_WORD}? {ENTAILMENT_LABEL}, \"+hypothesis,"""
f"""premise+\", {QUESTION_WORD}? {NEUTRAL_LABEL}, \"+hypothesis,""" f"""premise+\", {QUESTION_WORD}? {NEUTRAL_LABEL}, \"+hypothesis,"""
f"""premise+\", {QUESTION_WORD}? {CONTRADICTION_LABEL}, \"+hypothesis""" f"""premise+\", {QUESTION_WORD}? {CONTRADICTION_LABEL}, \"+hypothesis"""
f"]}}}}" f"]}}}}",
}, },
f, f,
allow_unicode=True allow_unicode=True,
) )
except FileExistsError: except FileExistsError:
err.append(file_name) err.append(file_name)
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment