Spaces:
Sleeping
Sleeping
Add application files
Browse files- .gitignore +129 -0
- Preprocessor/__init__.py +2 -0
- Preprocessor/demojize.py +91 -0
- Preprocessor/emojis_tr_twitter.json +0 -0
- Preprocessor/preprocessor.py +75 -0
- app.py +45 -0
.gitignore
ADDED
@@ -0,0 +1,129 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# Byte-compiled / optimized / DLL files
|
2 |
+
__pycache__/
|
3 |
+
*.py[cod]
|
4 |
+
*$py.class
|
5 |
+
|
6 |
+
# C extensions
|
7 |
+
*.so
|
8 |
+
|
9 |
+
# Distribution / packaging
|
10 |
+
.Python
|
11 |
+
build/
|
12 |
+
develop-eggs/
|
13 |
+
dist/
|
14 |
+
downloads/
|
15 |
+
eggs/
|
16 |
+
.eggs/
|
17 |
+
lib/
|
18 |
+
lib64/
|
19 |
+
parts/
|
20 |
+
sdist/
|
21 |
+
var/
|
22 |
+
wheels/
|
23 |
+
pip-wheel-metadata/
|
24 |
+
share/python-wheels/
|
25 |
+
*.egg-info/
|
26 |
+
.installed.cfg
|
27 |
+
*.egg
|
28 |
+
MANIFEST
|
29 |
+
|
30 |
+
# PyInstaller
|
31 |
+
# Usually these files are written by a python script from a template
|
32 |
+
# before PyInstaller builds the exe, so as to inject date/other infos into it.
|
33 |
+
*.manifest
|
34 |
+
*.spec
|
35 |
+
|
36 |
+
# Installer logs
|
37 |
+
pip-log.txt
|
38 |
+
pip-delete-this-directory.txt
|
39 |
+
|
40 |
+
# Unit test / coverage reports
|
41 |
+
htmlcov/
|
42 |
+
.tox/
|
43 |
+
.nox/
|
44 |
+
.coverage
|
45 |
+
.coverage.*
|
46 |
+
.cache
|
47 |
+
nosetests.xml
|
48 |
+
coverage.xml
|
49 |
+
*.cover
|
50 |
+
*.py,cover
|
51 |
+
.hypothesis/
|
52 |
+
.pytest_cache/
|
53 |
+
|
54 |
+
# Translations
|
55 |
+
*.mo
|
56 |
+
*.pot
|
57 |
+
|
58 |
+
# Django stuff:
|
59 |
+
*.log
|
60 |
+
local_settings.py
|
61 |
+
db.sqlite3
|
62 |
+
db.sqlite3-journal
|
63 |
+
|
64 |
+
# Flask stuff:
|
65 |
+
instance/
|
66 |
+
.webassets-cache
|
67 |
+
|
68 |
+
# Scrapy stuff:
|
69 |
+
.scrapy
|
70 |
+
|
71 |
+
# Sphinx documentation
|
72 |
+
docs/_build/
|
73 |
+
|
74 |
+
# PyBuilder
|
75 |
+
target/
|
76 |
+
|
77 |
+
# Jupyter Notebook
|
78 |
+
.ipynb_checkpoints
|
79 |
+
|
80 |
+
# IPython
|
81 |
+
profile_default/
|
82 |
+
ipython_config.py
|
83 |
+
|
84 |
+
# pyenv
|
85 |
+
.python-version
|
86 |
+
|
87 |
+
# pipenv
|
88 |
+
# According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
|
89 |
+
# However, in case of collaboration, if having platform-specific dependencies or dependencies
|
90 |
+
# having no cross-platform support, pipenv may install dependencies that don't work, or not
|
91 |
+
# install all needed dependencies.
|
92 |
+
#Pipfile.lock
|
93 |
+
|
94 |
+
# PEP 582; used by e.g. github.com/David-OConnor/pyflow
|
95 |
+
__pypackages__/
|
96 |
+
|
97 |
+
# Celery stuff
|
98 |
+
celerybeat-schedule
|
99 |
+
celerybeat.pid
|
100 |
+
|
101 |
+
# SageMath parsed files
|
102 |
+
*.sage.py
|
103 |
+
|
104 |
+
# Environments
|
105 |
+
.env
|
106 |
+
.venv
|
107 |
+
env/
|
108 |
+
venv/
|
109 |
+
ENV/
|
110 |
+
env.bak/
|
111 |
+
venv.bak/
|
112 |
+
|
113 |
+
# Spyder project settings
|
114 |
+
.spyderproject
|
115 |
+
.spyproject
|
116 |
+
|
117 |
+
# Rope project settings
|
118 |
+
.ropeproject
|
119 |
+
|
120 |
+
# mkdocs documentation
|
121 |
+
/site
|
122 |
+
|
123 |
+
# mypy
|
124 |
+
.mypy_cache/
|
125 |
+
.dmypy.json
|
126 |
+
dmypy.json
|
127 |
+
|
128 |
+
# Pyre type checker
|
129 |
+
.pyre/
|
Preprocessor/__init__.py
ADDED
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
1 |
+
from .demojize import demojize
|
2 |
+
from .preprocessor import preprocess
|
Preprocessor/demojize.py
ADDED
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import json
|
2 |
+
import os
|
3 |
+
|
4 |
+
dir = os.path.dirname(__file__)
|
5 |
+
EMOJI_DATA_PATH = os.path.join(dir, "emojis_tr_twitter.json")
|
6 |
+
|
7 |
+
with open(EMOJI_DATA_PATH, "r") as f:
|
8 |
+
emojis = json.load(f)
|
9 |
+
|
10 |
+
_SEARCH_TREE = None
|
11 |
+
|
12 |
+
|
13 |
+
def _get_search_tree():
|
14 |
+
global _SEARCH_TREE
|
15 |
+
if _SEARCH_TREE is None:
|
16 |
+
_SEARCH_TREE = {}
|
17 |
+
for emj in emojis:
|
18 |
+
sub_tree = _SEARCH_TREE
|
19 |
+
lastidx = len(emj) - 1
|
20 |
+
for i, char in enumerate(emj):
|
21 |
+
if char not in sub_tree:
|
22 |
+
sub_tree[char] = {}
|
23 |
+
sub_tree = sub_tree[char]
|
24 |
+
if i == lastidx:
|
25 |
+
sub_tree["data"] = emojis[emj]
|
26 |
+
|
27 |
+
return _SEARCH_TREE
|
28 |
+
|
29 |
+
|
30 |
+
def demojize(
|
31 |
+
string,
|
32 |
+
delimiters=("<emoji> ", " </emoji>"),
|
33 |
+
language="tr",
|
34 |
+
version=None,
|
35 |
+
handle_version=None,
|
36 |
+
):
|
37 |
+
if language == "alias":
|
38 |
+
language = "tr"
|
39 |
+
_use_aliases = True
|
40 |
+
else:
|
41 |
+
_use_aliases = False
|
42 |
+
tree = _get_search_tree()
|
43 |
+
result = []
|
44 |
+
i = 0
|
45 |
+
length = len(string)
|
46 |
+
while i < length:
|
47 |
+
consumed = False
|
48 |
+
char = string[i]
|
49 |
+
if char in tree:
|
50 |
+
j = i + 1
|
51 |
+
sub_tree = tree[char]
|
52 |
+
while j < length and string[j] in sub_tree:
|
53 |
+
sub_tree = sub_tree[string[j]]
|
54 |
+
j += 1
|
55 |
+
if "data" in sub_tree:
|
56 |
+
emj_data = sub_tree["data"]
|
57 |
+
code_points = string[i:j]
|
58 |
+
replace_str = None
|
59 |
+
if version is not None and emj_data["E"] > version:
|
60 |
+
if callable(handle_version):
|
61 |
+
emj_data = emj_data.copy()
|
62 |
+
emj_data["match_start"] = i
|
63 |
+
emj_data["match_end"] = j
|
64 |
+
replace_str = handle_version(code_points, emj_data)
|
65 |
+
elif handle_version is not None:
|
66 |
+
replace_str = str(handle_version)
|
67 |
+
else:
|
68 |
+
replace_str = None
|
69 |
+
elif language in emj_data:
|
70 |
+
if _use_aliases and "alias" in emj_data:
|
71 |
+
replace_str = (
|
72 |
+
delimiters[0] + emj_data["alias"][0][:-1] + delimiters[1]
|
73 |
+
)
|
74 |
+
else:
|
75 |
+
replace_str = (
|
76 |
+
delimiters[0] + emj_data[language][1:-1] + delimiters[1]
|
77 |
+
)
|
78 |
+
else:
|
79 |
+
# The emoji exists, but it is not translated, so we keep the emoji
|
80 |
+
replace_str = code_points
|
81 |
+
|
82 |
+
i = j - 1
|
83 |
+
consumed = True
|
84 |
+
if replace_str:
|
85 |
+
result.append(replace_str)
|
86 |
+
|
87 |
+
if not consumed and char != "\ufe0e" and char != "\ufe0f":
|
88 |
+
result.append(char)
|
89 |
+
i += 1
|
90 |
+
|
91 |
+
return "".join(result)
|
Preprocessor/emojis_tr_twitter.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
Preprocessor/preprocessor.py
ADDED
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import urllib
|
2 |
+
import html
|
3 |
+
import re
|
4 |
+
|
5 |
+
from urlextract import URLExtract
|
6 |
+
from unicodedata import normalize
|
7 |
+
|
8 |
+
from .demojize import demojize
|
9 |
+
|
10 |
+
|
11 |
+
def hashtag_handler(text: str):
|
12 |
+
pattern = r"(#([^\s]+))"
|
13 |
+
return re.sub(pattern, " <hashtag> \\2 </hashtag> ", text)
|
14 |
+
|
15 |
+
|
16 |
+
def cashtag_handler(text: str):
|
17 |
+
pattern = r"(\$([^\s]+))"
|
18 |
+
return re.sub(pattern, " <cashtag> \\2 </cashtag> ", text)
|
19 |
+
|
20 |
+
|
21 |
+
def mention_handler(text: str):
|
22 |
+
pattern = r"(@([^\s]+))"
|
23 |
+
return re.sub(pattern, " @user ", text)
|
24 |
+
|
25 |
+
|
26 |
+
url_extractor = URLExtract()
|
27 |
+
|
28 |
+
|
29 |
+
def url_handler(text: str):
|
30 |
+
urls = list(url_extractor.gen_urls(text))
|
31 |
+
updated_urls = list(
|
32 |
+
set([url if "http" in url else f"https://{url}" for url in urls])
|
33 |
+
)
|
34 |
+
domains = [urllib.parse.urlparse(url_text).netloc for url_text in updated_urls]
|
35 |
+
for i in range(len(domains)):
|
36 |
+
text = text.replace(urls[i], f" <http> {domains[i]} </http> ")
|
37 |
+
return text
|
38 |
+
|
39 |
+
|
40 |
+
def email_handler(text: str):
|
41 |
+
pattern = r"[\w.+-]+@[\w-]+\.[\w.-]+"
|
42 |
+
match = re.findall(pattern, text)
|
43 |
+
for m in match:
|
44 |
+
text = text.replace(m, " <email> ").strip()
|
45 |
+
return text
|
46 |
+
|
47 |
+
|
48 |
+
def emoji_handler(text: str):
|
49 |
+
return demojize(text, language="tr", delimiters=(" <emoji> ", " </emoji> "))
|
50 |
+
|
51 |
+
|
52 |
+
def normalize_text(text: str):
|
53 |
+
return normalize("NFC", text)
|
54 |
+
|
55 |
+
|
56 |
+
def preprocess(text: str):
|
57 |
+
output = html.unescape(text)
|
58 |
+
output = normalize_text(output)
|
59 |
+
output = email_handler(output)
|
60 |
+
output = url_handler(output)
|
61 |
+
output = hashtag_handler(output)
|
62 |
+
output = cashtag_handler(output)
|
63 |
+
output = mention_handler(output)
|
64 |
+
output = emoji_handler(output)
|
65 |
+
output = re.sub(r"\s+", " ", output)
|
66 |
+
output = output.lower()
|
67 |
+
output = output.strip()
|
68 |
+
|
69 |
+
return output
|
70 |
+
|
71 |
+
|
72 |
+
if __name__ == "__main__":
|
73 |
+
sample_text = ""
|
74 |
+
preprocessed_text = preprocess(sample_text)
|
75 |
+
print(preprocessed_text)
|
app.py
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
import torch
|
3 |
+
from peft import (
|
4 |
+
PeftModel,
|
5 |
+
PeftConfig,
|
6 |
+
)
|
7 |
+
|
8 |
+
from transformers import AutoModelForSequenceClassification, AutoTokenizer
|
9 |
+
from Preprocessor import preprocess
|
10 |
+
|
11 |
+
|
12 |
+
peft_model = "VRLLab/TurkishBERTweet-Lora-SA"
|
13 |
+
peft_config = PeftConfig.from_pretrained(peft_model)
|
14 |
+
|
15 |
+
# loading Tokenizer
|
16 |
+
padding_side = "right"
|
17 |
+
tokenizer = AutoTokenizer.from_pretrained(
|
18 |
+
peft_config.base_model_name_or_path, padding_side=padding_side
|
19 |
+
)
|
20 |
+
if getattr(tokenizer, "pad_token_id") is None:
|
21 |
+
tokenizer.pad_token_id = tokenizer.eos_token_id
|
22 |
+
|
23 |
+
id2label_sa = {0: "negative", 2: "positive", 1: "neutral"}
|
24 |
+
turkishBERTweet_sa = AutoModelForSequenceClassification.from_pretrained(
|
25 |
+
peft_config.base_model_name_or_path,
|
26 |
+
return_dict=True,
|
27 |
+
num_labels=len(id2label_sa),
|
28 |
+
id2label=id2label_sa,
|
29 |
+
)
|
30 |
+
turkishBERTweet_sa = PeftModel.from_pretrained(turkishBERTweet_sa, peft_model)
|
31 |
+
|
32 |
+
|
33 |
+
st.title("Sentiment Analysis with HuggingFace Spaces")
|
34 |
+
st.write("Enter a sentence to analyze its sentiment:")
|
35 |
+
|
36 |
+
|
37 |
+
user_input = st.text_input("")
|
38 |
+
if user_input:
|
39 |
+
with torch.no_grad():
|
40 |
+
ids = tokenizer.encode_plus(preprocess(user_input), return_tensors="pt")
|
41 |
+
logits = turkishBERTweet_sa(**ids).logits
|
42 |
+
label_id = logits.argmax(-1).item()
|
43 |
+
confidence = logits.softmax(-1)[0, label_id].item()
|
44 |
+
st.write(f"Sentiment: {id2label_sa[label_id]}")
|
45 |
+
st.write(f"Confidence: {confidence:.2f}")
|