Download data/scripts/test_xlit.py from ViuAI/ViuMini-Dense-360M: direct link, hf CLI and curl.
- Browser
- Download file 2.67 kB
-
https://huggingface.co/ViuAI/ViuMini-Dense-360M/resolve/main/data/scripts/test_xlit.py
- Command line
-
hf download hf://ViuAI/ViuMini-Dense-360M/data/scripts/test_xlit.py
-
curl -L -o test_xlit.py https://huggingface.co/ViuAI/ViuMini-Dense-360M/resolve/main/data/scripts/test_xlit.py
2.67 kB
| # -*- coding: utf-8 -*- | |
| import sys | |
| sys.stdout.reconfigure(encoding='utf-8') | |
| VOWELS = { | |
| 'अ': 'a', 'आ': 'aa', 'इ': 'i', 'ई': 'ee', 'उ': 'u', 'ऊ': 'oo', 'ऋ': 'ri', | |
| 'ए': 'e', 'ऐ': 'ai', 'ओ': 'o', 'औ': 'au', 'अं': 'an', 'अः': 'ah' | |
| } | |
| MATRAS = { | |
| 'ा': 'aa', 'ि': 'i', 'ी': 'ee', 'ु': 'u', 'ू': 'oo', 'ृ': 'ri', | |
| 'े': 'e', 'ै': 'ai', 'ो': 'o', 'ौ': 'au', 'ं': 'n', 'ँ': 'n', 'ः': 'h', '्': '' | |
| } | |
| CONSONANTS = { | |
| 'क': 'k', 'ख': 'kh', 'ग': 'g', 'घ': 'gh', 'ङ': 'ng', | |
| 'च': 'ch', 'छ': 'chh', 'ज': 'j', 'झ': 'jh', 'ञ': 'ny', | |
| 'ट': 't', 'ठ': 'th', 'ड': 'd', 'ढ': 'dh', 'ण': 'n', | |
| 'त': 't', 'थ': 'th', 'द': 'd', 'ध': 'dh', 'न': 'n', | |
| 'प': 'p', 'फ': 'ph', 'ब': 'b', 'भ': 'bh', 'म': 'm', | |
| 'य': 'y', 'र': 'r', 'ल': 'l', 'व': 'v', | |
| 'श': 'sh', 'ष': 'sh', 'स': 's', 'ह': 'h', | |
| 'क़': 'q', 'ख़': 'kh', 'ग़': 'gh', 'ज़': 'z', 'ड़': 'r', 'ढ़': 'rh', 'फ़': 'f' | |
| } | |
| def devanagari_to_hinglish(text): | |
| out = [] | |
| i = 0 | |
| n = len(text) | |
| while i < n: | |
| if i + 1 < n and text[i:i+2] in CONSONANTS: | |
| base = CONSONANTS[text[i:i+2]] | |
| i += 2 | |
| if i < n and text[i] in MATRAS: | |
| out.append(base + MATRAS[text[i]]) | |
| i += 1 | |
| else: | |
| out.append(base + ('a' if i < n and text[i] not in ' \t\n.,!?;:()[]' else '')) | |
| elif text[i] in CONSONANTS: | |
| base = CONSONANTS[text[i]] | |
| i += 1 | |
| if i < n and text[i] in MATRAS: | |
| out.append(base + MATRAS[text[i]]) | |
| i += 1 | |
| else: | |
| if i == n or text[i] in ' \t\n.,!?;:()[]\'"': | |
| out.append(base) | |
| else: | |
| out.append(base + 'a') | |
| elif text[i] in VOWELS: | |
| out.append(VOWELS[text[i]]) | |
| i += 1 | |
| elif text[i] in MATRAS: | |
| out.append(MATRAS[text[i]]) | |
| i += 1 | |
| else: | |
| out.append(text[i]) | |
| i += 1 | |
| return ''.join(out) | |
| if __name__ == '__main__': | |
| s1 = "भारत एक महान और प्राचीन देश है। हम सब मिलकर इसे आगे बढ़ाएंगे।" | |
| s2 = "आप आज क्या कर रहे हो? मुझे कुछ नया सिखाओ।" | |
| s3 = "8D problem solving me D4 root cause analysis sabse important stage hoti hai." | |
| print("S1:", devanagari_to_hinglish(s1)) | |
| print("S2:", devanagari_to_hinglish(s2)) | |
| print("S3:", devanagari_to_hinglish(s3)) | |