File size: 5,122 Bytes
e5cc640 57f2eec e5cc640 f34d9ab 57f2eec e5cc640 1e8689e 57f2eec e5cc640 1e8689e e5cc640 1e8689e e5cc640 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 |
import gradio as gr
import os
import torch
import gradio as gr
from transformers import M2M100Tokenizer, M2M100ForConditionalGeneration
if torch.cuda.is_available():
device = torch.device("cuda:0")
else:
device = torch.device("cpu")
tokenizer = M2M100Tokenizer.from_pretrained("facebook/m2m100_1.2B")
model = M2M100ForConditionalGeneration.from_pretrained("facebook/m2m100_1.2B").to(device)
model.eval()
lang_id = {
"":"",
"Afrikaans": "af",
"Albanian": "sq",
"Amharic": "am",
"Arabic": "ar",
"Armenian": "hy",
"Asturian": "ast",
"Azerbaijani": "az",
"Bashkir": "ba",
"Belarusian": "be",
"Bulgarian": "bg",
"Bengali": "bn",
"Breton": "br",
"Bosnian": "bs",
"Burmese": "my",
"Catalan": "ca",
"Cebuano": "ceb",
"Chinese": "zh",
"Chinese (simplified)": "zh",
"Chinese (traditional)": "zh",
"Croatian": "hr",
"Czech": "cs",
"Danish": "da",
"Dutch": "nl",
"English": "en",
"Estonian": "et",
"Fulah": "ff",
"Finnish": "fi",
"French": "fr",
"Western Frisian": "fy",
"Gaelic": "gd",
"Galician": "gl",
"Georgian": "ka",
"German": "de",
"Greek": "el",
"Gujarati": "gu",
"Hausa": "ha",
"Hebrew": "he",
"Hindi": "hi",
"Haitian": "ht",
"Hungarian": "hu",
"Irish": "ga",
"Indonesian": "id",
"Igbo": "ig",
"Iloko": "ilo",
"Icelandic": "is",
"Italian": "it",
"Japanese": "ja",
"Javanese": "jv",
"Kazakh": "kk",
"Central Khmer": "km",
"Kannada": "kn",
"Korean": "ko",
"Luxembourgish": "lb",
"Ganda": "lg",
"Lingala": "ln",
"Lao": "lo",
"Lithuanian": "lt",
"Latvian": "lv",
"Malagasy": "mg",
"Macedonian": "mk",
"Malayalam": "ml",
"Mongolian": "mn",
"Marathi": "mr",
"Malay": "ms",
"Nepali": "ne",
"Norwegian": "no",
"Northern Sotho": "ns",
"Occitan": "oc",
"Oriya": "or",
"Panjabi": "pa",
"Persian": "fa",
"Polish": "pl",
"Pushto": "ps",
"Portuguese": "pt",
"Romanian": "ro",
"Russian": "ru",
"Sindhi": "sd",
"Sinhala": "si",
"Slovak": "sk",
"Slovenian": "sl",
"Spanish": "es",
"Somali": "so",
"Serbian": "sr",
"Serbian (cyrillic)": "sr",
"Serbian (latin)": "sr",
"Swati": "ss",
"Sundanese": "su",
"Swedish": "sv",
"Swahili": "sw",
"Tamil": "ta",
"Thai": "th",
"Tagalog": "tl",
"Tswana": "tn",
"Turkish": "tr",
"Ukrainian": "uk",
"Urdu": "ur",
"Uzbek": "uz",
"Vietnamese": "vi",
"Welsh": "cy",
"Wolof": "wo",
"Xhosa": "xh",
"Yiddish": "yi",
"Yoruba": "yo",
"Zulu": "zu",
}
def trans_page(input,trg):
src_lang = lang_id["English"]
trg_lang = lang_id[trg]
if trg_lang != src_lang:
tokenizer.src_lang = src_lang
with torch.no_grad():
encoded_input = tokenizer(input, return_tensors="pt").to(device)
generated_tokens = model.generate(**encoded_input, forced_bos_token_id=tokenizer.get_lang_id(trg_lang))
translated_text = tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)[0]
else:
translated_text=input
pass
return translated_text
def trans_to(input,src,trg):
src_lang = lang_id[src]
trg_lang = lang_id[trg]
if trg_lang != src_lang:
tokenizer.src_lang = src_lang
with torch.no_grad():
encoded_input = tokenizer(input, return_tensors="pt").to(device)
generated_tokens = model.generate(**encoded_input, forced_bos_token_id=tokenizer.get_lang_id(trg_lang))
translated_text = tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)[0]
else:
translated_text=input
pass
return translated_text
md1 = "Translate - 100 Languages"
with gr.Blocks() as transbot:
this=gr.State()
with gr.Row():
gr.Column()
with gr.Column():
with gr.Row():
t_space = gr.Dropdown(label="Translate Space", choices=list(lang_id.keys()),value="English")
t_submit = gr.Button("Translate Space")
gr.Column()
with gr.Row():
gr.Column()
with gr.Column():
md = gr.Markdown("""<h1><center>Translate - 100 Languages</center></h1><h4><center>Translation may not be accurate</center></h4>""")
with gr.Row():
lang_from = gr.Dropdown(label="From:", choices=list(lang_id.keys()),value="English")
lang_to = gr.Dropdown(label="To:", choices=list(lang_id.keys()),value="Chinese")
submit = gr.Button("Go")
with gr.Row():
with gr.Column():
message = gr.Textbox(label="Prompt",placeholder="Enter Prompt",lines=4)
translated = gr.Textbox(label="Translated",lines=4,interactive=False)
gr.Column()
t_submit.click(trans_page,[this,t_space],[this])
submit.click(trans_to, inputs=[message,lang_from,lang_to], outputs=[translated])
transbot.queue(concurrency_count=20)
transbot.launch() |