Spaces:
Runtime error
Runtime error
Duplicate from kadirnar/BioGpt
Browse filesCo-authored-by: Kadir Nar <kadirnar@users.noreply.huggingface.co>
- .gitattributes +34 -0
- README.md +14 -0
- app.py +95 -0
- requirements.txt +5 -0
- utils.py +106 -0
.gitattributes
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
28 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
29 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
30 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
31 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
32 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
33 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
34 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
title: BioGpt
|
3 |
+
emoji: 🌖
|
4 |
+
colorFrom: red
|
5 |
+
colorTo: purple
|
6 |
+
sdk: gradio
|
7 |
+
sdk_version: 3.17.0
|
8 |
+
app_file: app.py
|
9 |
+
pinned: false
|
10 |
+
license: mit
|
11 |
+
duplicated_from: kadirnar/BioGpt
|
12 |
+
---
|
13 |
+
|
14 |
+
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
app.py
ADDED
@@ -0,0 +1,95 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from transformers import pipeline, set_seed
|
2 |
+
from transformers import BioGptTokenizer, BioGptForCausalLM
|
3 |
+
from multilingual_translation import translate
|
4 |
+
from utils import lang_ids
|
5 |
+
import gradio as gr
|
6 |
+
import torch
|
7 |
+
|
8 |
+
biogpt_model_list = [
|
9 |
+
"microsoft/biogpt",
|
10 |
+
"microsoft/BioGPT-Large",
|
11 |
+
"microsoft/BioGPT-Large-PubMedQA"
|
12 |
+
]
|
13 |
+
|
14 |
+
lang_model_list = [
|
15 |
+
"facebook/m2m100_1.2B",
|
16 |
+
"facebook/m2m100_418M"
|
17 |
+
]
|
18 |
+
|
19 |
+
lang_list = list(lang_ids.keys())
|
20 |
+
|
21 |
+
def translate_to_english(text, lang_model_id, base_lang):
|
22 |
+
if base_lang == "English":
|
23 |
+
return text
|
24 |
+
else:
|
25 |
+
base_lang = lang_ids[base_lang]
|
26 |
+
new_text = translate(lang_model_id, text, base_lang, "en")
|
27 |
+
return new_text[0]
|
28 |
+
|
29 |
+
def biogpt(
|
30 |
+
prompt: str,
|
31 |
+
biogpt_model_id: str,
|
32 |
+
max_length: str,
|
33 |
+
num_return_sequences: int,
|
34 |
+
base_lang: str,
|
35 |
+
lang_model_id: str
|
36 |
+
):
|
37 |
+
|
38 |
+
en_prompt = translate_to_english(prompt, lang_model_id, base_lang)
|
39 |
+
generator = pipeline("text-generation", model=biogpt_model_id, device="cuda:0")
|
40 |
+
output = generator(en_prompt, max_length=max_length, num_return_sequences=num_return_sequences, do_sample=True)
|
41 |
+
output_dict = {}
|
42 |
+
for i in range(num_return_sequences):
|
43 |
+
output_dict[str(i+1)] = output[i]['generated_text']
|
44 |
+
|
45 |
+
output_text = ""
|
46 |
+
for i in range(num_return_sequences):
|
47 |
+
output_text += f'{output_dict[str(i+1)]}\n\n'
|
48 |
+
|
49 |
+
if base_lang == "English":
|
50 |
+
base_lang_output = output_text
|
51 |
+
|
52 |
+
else:
|
53 |
+
base_lang_output_ = ""
|
54 |
+
for i in range(num_return_sequences):
|
55 |
+
base_lang_output_ += f'{translate(lang_model_id, output_dict[str(i+1)], "en", lang_ids[base_lang])[0]}\n\n'
|
56 |
+
base_lang_output = base_lang_output_
|
57 |
+
|
58 |
+
|
59 |
+
return en_prompt, output_text, base_lang_output
|
60 |
+
|
61 |
+
|
62 |
+
inputs = [
|
63 |
+
gr.Textbox(lines=5, value="COVID-19 is", label="Prompt"),
|
64 |
+
gr.Dropdown(biogpt_model_list, value="microsoft/biogpt", label="BioGPT Model ID"),
|
65 |
+
gr.Slider(minumum=1, maximum=100, value=25, step=1, label="Max Length"),
|
66 |
+
gr.Slider(minumum=1, maximum=10, value=2, step=1, label="Number of Outputs"),
|
67 |
+
gr.Dropdown(lang_list, value="English", label="Base Language"),
|
68 |
+
gr.Dropdown(lang_model_list, value="facebook/m2m100_418M", label="Language Model ID")
|
69 |
+
]
|
70 |
+
|
71 |
+
outputs = [
|
72 |
+
gr.outputs.Textbox(label="Prompt"),
|
73 |
+
gr.outputs.Textbox(label="Output"),
|
74 |
+
gr.outputs.Textbox(label="Translated Output")
|
75 |
+
]
|
76 |
+
|
77 |
+
examples = [
|
78 |
+
["COVID-19 is", "microsoft/biogpt", 25, 2, "English", "facebook/m2m100_418M"],
|
79 |
+
["Kanser", "microsoft/biogpt", 25, 2, "Turkish", "facebook/m2m100_1.2B"]
|
80 |
+
]
|
81 |
+
|
82 |
+
title = "M2M100 + BioGPT: Generative Pre-trained Transformer for Biomedical Text Generation and Mining"
|
83 |
+
|
84 |
+
description = "BioGPT is a domain-specific generative pre-trained Transformer language model for biomedical text generation and mining. BioGPT follows the Transformer language model backbone, and is pre-trained on 15M PubMed abstracts from scratch. Github: github.com/microsoft/BioGPT Paper: https://arxiv.org/abs/2210.10341"
|
85 |
+
|
86 |
+
demo_app = gr.Interface(
|
87 |
+
biogpt,
|
88 |
+
inputs,
|
89 |
+
outputs,
|
90 |
+
title=title,
|
91 |
+
description=description,
|
92 |
+
examples=examples,
|
93 |
+
cache_examples=False,
|
94 |
+
)
|
95 |
+
demo_app.launch(debug=True, enable_queue=True)
|
requirements.txt
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
sacremoses
|
2 |
+
torch
|
3 |
+
beautifulsoup4==4.11.2
|
4 |
+
multilingual_translation==0.0.3
|
5 |
+
requests==2.28.1
|
utils.py
ADDED
@@ -0,0 +1,106 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from bs4 import BeautifulSoup
|
2 |
+
import requests
|
3 |
+
|
4 |
+
|
5 |
+
lang_ids = {
|
6 |
+
"Afrikaans": "af",
|
7 |
+
"Amharic": "am",
|
8 |
+
"Arabic": "ar",
|
9 |
+
"Asturian": "ast",
|
10 |
+
"Azerbaijani": "az",
|
11 |
+
"Bashkir": "ba",
|
12 |
+
"Belarusian": "be",
|
13 |
+
"Bulgarian": "bg",
|
14 |
+
"Bengali": "bn",
|
15 |
+
"Breton": "br",
|
16 |
+
"Bosnian": "bs",
|
17 |
+
"Catalan": "ca",
|
18 |
+
"Cebuano": "ceb",
|
19 |
+
"Czech": "cs",
|
20 |
+
"Welsh": "cy",
|
21 |
+
"Danish": "da",
|
22 |
+
"German": "de",
|
23 |
+
"Greeek": "el",
|
24 |
+
"English": "en",
|
25 |
+
"Spanish": "es",
|
26 |
+
"Estonian": "et",
|
27 |
+
"Persian": "fa",
|
28 |
+
"Fulah": "ff",
|
29 |
+
"Finnish": "fi",
|
30 |
+
"French": "fr",
|
31 |
+
"Western Frisian": "fy",
|
32 |
+
"Irish": "ga",
|
33 |
+
"Gaelic": "gd",
|
34 |
+
"Galician": "gl",
|
35 |
+
"Gujarati": "gu",
|
36 |
+
"Hausa": "ha",
|
37 |
+
"Hebrew": "he",
|
38 |
+
"Hindi": "hi",
|
39 |
+
"Croatian": "hr",
|
40 |
+
"Haitian": "ht",
|
41 |
+
"Hungarian": "hu",
|
42 |
+
"Armenian": "hy",
|
43 |
+
"Indonesian": "id",
|
44 |
+
"Igbo": "ig",
|
45 |
+
"Iloko": "ilo",
|
46 |
+
"Icelandic": "is",
|
47 |
+
"Italian": "it",
|
48 |
+
"Japanese": "ja",
|
49 |
+
"Javanese": "jv",
|
50 |
+
"Georgian": "ka",
|
51 |
+
"Kazakh": "kk",
|
52 |
+
"Central Khmer": "km",
|
53 |
+
"Kannada": "kn",
|
54 |
+
"Korean": "ko",
|
55 |
+
"Luxembourgish": "lb",
|
56 |
+
"Ganda": "lg",
|
57 |
+
"Lingala": "ln",
|
58 |
+
"Lao": "lo",
|
59 |
+
"Lithuanian": "lt",
|
60 |
+
"Latvian": "lv",
|
61 |
+
"Malagasy": "mg",
|
62 |
+
"Macedonian": "mk",
|
63 |
+
"Malayalam": "ml",
|
64 |
+
"Mongolian": "mn",
|
65 |
+
"Marathi": "mr",
|
66 |
+
"Malay": "ms",
|
67 |
+
"Burmese": "my",
|
68 |
+
"Nepali": "ne",
|
69 |
+
"Dutch": "nl",
|
70 |
+
"Norwegian": "no",
|
71 |
+
"Northern Sotho": "ns",
|
72 |
+
"Occitan": "oc",
|
73 |
+
"Oriya": "or",
|
74 |
+
"Panjabi": "pa",
|
75 |
+
"Polish": "pl",
|
76 |
+
"Pushto": "ps",
|
77 |
+
"Portuguese": "pt",
|
78 |
+
"Romanian": "ro",
|
79 |
+
"Russian": "ru",
|
80 |
+
"Sindhi": "sd",
|
81 |
+
"Sinhala": "si",
|
82 |
+
"Slovak": "sk",
|
83 |
+
"Slovenian": "sl",
|
84 |
+
"Somali": "so",
|
85 |
+
"Albanian": "sq",
|
86 |
+
"Serbian": "sr",
|
87 |
+
"Swati": "ss",
|
88 |
+
"Sundanese": "su",
|
89 |
+
"Swedish": "sv",
|
90 |
+
"Swahili": "sw",
|
91 |
+
"Tamil": "ta",
|
92 |
+
"Thai": "th",
|
93 |
+
"Tagalog": "tl",
|
94 |
+
"Tswana": "tn",
|
95 |
+
"Turkish": "tr",
|
96 |
+
"Ukrainian": "uk",
|
97 |
+
"Urdu": "ur",
|
98 |
+
"Uzbek": "uz",
|
99 |
+
"Vietnamese": "vi",
|
100 |
+
"Wolof": "wo",
|
101 |
+
"Xhosa": "xh",
|
102 |
+
"Yiddish": "yi",
|
103 |
+
"Yoruba": "yo",
|
104 |
+
"Chinese": "zh",
|
105 |
+
"Zulu": "zu",
|
106 |
+
}
|