Spaces:
Running
Running
petrsovadina
commited on
Commit
•
f7e1e6e
1
Parent(s):
6b6076c
Update presidio_streamlit.py
Browse files- presidio_streamlit.py +76 -272
presidio_streamlit.py
CHANGED
@@ -1,12 +1,10 @@
|
|
1 |
-
"""Streamlit app
|
2 |
import logging
|
3 |
import os
|
4 |
import traceback
|
5 |
|
6 |
-
import dotenv
|
7 |
import pandas as pd
|
8 |
import streamlit as st
|
9 |
-
import streamlit.components.v1 as components
|
10 |
from annotated_text import annotated_text
|
11 |
from streamlit_tags import st_tags
|
12 |
|
@@ -21,7 +19,7 @@ from presidio_helpers import (
|
|
21 |
)
|
22 |
|
23 |
st.set_page_config(
|
24 |
-
page_title="
|
25 |
layout="wide",
|
26 |
initial_sidebar_state="expanded",
|
27 |
menu_items={
|
@@ -29,347 +27,153 @@ st.set_page_config(
|
|
29 |
},
|
30 |
)
|
31 |
|
32 |
-
dotenv.load_dotenv()
|
33 |
logger = logging.getLogger("presidio-streamlit")
|
34 |
|
35 |
-
|
36 |
-
allow_other_models = os.getenv("ALLOW_OTHER_MODELS", False)
|
37 |
-
|
38 |
-
|
39 |
# Sidebar
|
40 |
-
st.sidebar.header(
|
41 |
-
"""
|
42 |
-
PII De-Identification with [Microsoft Presidio](https://microsoft.github.io/presidio/)
|
43 |
-
"""
|
44 |
-
)
|
45 |
-
|
46 |
|
47 |
model_help_text = """
|
48 |
-
|
49 |
-
Presidio supports multiple NER packages off-the-shelf, such as spaCy, Huggingface, Stanza and Flair,
|
50 |
-
as well as service such as Azure Text Analytics PII.
|
51 |
"""
|
52 |
-
st_ta_key = st_ta_endpoint = ""
|
53 |
|
54 |
model_list = [
|
55 |
-
"spaCy/en_core_web_lg",
|
56 |
"iiiorg/piiranha-v1-detect-personal-information",
|
57 |
-
"
|
58 |
]
|
59 |
-
|
60 |
-
model_list.pop()
|
61 |
-
# Select model
|
62 |
st_model = st.sidebar.selectbox(
|
63 |
-
"NER model
|
64 |
model_list,
|
65 |
-
index=
|
66 |
help=model_help_text,
|
67 |
)
|
68 |
|
69 |
-
# Extract model package.
|
70 |
st_model_package = st_model.split("/")[0]
|
|
|
71 |
|
72 |
-
|
73 |
-
st_model = (
|
74 |
-
st_model
|
75 |
-
if st_model_package.lower() not in ("spacy","piiiranha")
|
76 |
-
else "/".join(st_model.split("/")[1:])
|
77 |
-
)
|
78 |
-
|
79 |
-
if st_model == "Other":
|
80 |
-
st_model_package = st.sidebar.selectbox(
|
81 |
-
"NER model OSS package", options=["spacy","piiiranha"]
|
82 |
-
)
|
83 |
-
st_model = st.sidebar.text_input(f"NER model name", value="")
|
84 |
-
|
85 |
-
|
86 |
-
|
87 |
-
|
88 |
-
st.sidebar.warning("Note: Models might take some time to download. ")
|
89 |
-
|
90 |
-
analyzer_params = (st_model_package, st_model, st_ta_key, st_ta_endpoint)
|
91 |
logger.debug(f"analyzer_params: {analyzer_params}")
|
92 |
|
93 |
st_operator = st.sidebar.selectbox(
|
94 |
-
"
|
95 |
-
["
|
96 |
-
index=
|
97 |
help="""
|
98 |
-
|
99 |
-
-
|
100 |
-
-
|
101 |
-
-
|
102 |
-
|
103 |
-
- Mask: Replaces a requested number of characters with an asterisk (or other mask character)\n
|
104 |
-
- Hash: Replaces with the hash of the PII string\n
|
105 |
-
- Encrypt: Replaces with an AES encryption of the PII string, allowing the process to be reversed
|
106 |
-
""",
|
107 |
)
|
108 |
-
st_mask_char = "*"
|
109 |
-
st_number_of_chars = 15
|
110 |
-
st_encrypt_key = "WmZq4t7w!z%C&F)J"
|
111 |
-
|
112 |
-
open_ai_params = None
|
113 |
-
|
114 |
-
logger.debug(f"st_operator: {st_operator}")
|
115 |
-
|
116 |
-
|
117 |
-
def set_up_openai_synthesis():
|
118 |
-
"""Set up the OpenAI API key and model for text synthesis."""
|
119 |
-
|
120 |
-
if os.getenv("OPENAI_TYPE", default="openai") == "Azure":
|
121 |
-
openai_api_type = "azure"
|
122 |
-
st_openai_api_base = st.sidebar.text_input(
|
123 |
-
"Azure OpenAI base URL",
|
124 |
-
value=os.getenv("AZURE_OPENAI_ENDPOINT", default=""),
|
125 |
-
)
|
126 |
-
openai_key = os.getenv("AZURE_OPENAI_KEY", default="")
|
127 |
-
st_deployment_id = st.sidebar.text_input(
|
128 |
-
"Deployment name", value=os.getenv("AZURE_OPENAI_DEPLOYMENT", default="")
|
129 |
-
)
|
130 |
-
st_openai_version = st.sidebar.text_input(
|
131 |
-
"OpenAI version",
|
132 |
-
value=os.getenv("OPENAI_API_VERSION", default="2023-05-15"),
|
133 |
-
)
|
134 |
-
else:
|
135 |
-
openai_api_type = "openai"
|
136 |
-
st_openai_version = st_openai_api_base = None
|
137 |
-
st_deployment_id = ""
|
138 |
-
openai_key = os.getenv("OPENAI_KEY", default="")
|
139 |
-
st_openai_key = st.sidebar.text_input(
|
140 |
-
"OPENAI_KEY",
|
141 |
-
value=openai_key,
|
142 |
-
help="See https://help.openai.com/en/articles/4936850-where-do-i-find-my-secret-api-key for more info.",
|
143 |
-
type="password",
|
144 |
-
)
|
145 |
-
st_openai_model = st.sidebar.text_input(
|
146 |
-
"OpenAI model for text synthesis",
|
147 |
-
value=os.getenv("OPENAI_MODEL", default="gpt-3.5-turbo-instruct"),
|
148 |
-
help="See more here: https://platform.openai.com/docs/models/",
|
149 |
-
)
|
150 |
-
return (
|
151 |
-
openai_api_type,
|
152 |
-
st_openai_api_base,
|
153 |
-
st_deployment_id,
|
154 |
-
st_openai_version,
|
155 |
-
st_openai_key,
|
156 |
-
st_openai_model,
|
157 |
-
)
|
158 |
|
|
|
|
|
159 |
|
160 |
-
if st_operator == "
|
161 |
st_number_of_chars = st.sidebar.number_input(
|
162 |
-
"
|
163 |
)
|
164 |
st_mask_char = st.sidebar.text_input(
|
165 |
-
"
|
166 |
-
)
|
167 |
-
elif st_operator == "encrypt":
|
168 |
-
st_encrypt_key = st.sidebar.text_input("AES key", value=st_encrypt_key)
|
169 |
-
elif st_operator == "synthesize":
|
170 |
-
(
|
171 |
-
openai_api_type,
|
172 |
-
st_openai_api_base,
|
173 |
-
st_deployment_id,
|
174 |
-
st_openai_version,
|
175 |
-
st_openai_key,
|
176 |
-
st_openai_model,
|
177 |
-
) = set_up_openai_synthesis()
|
178 |
-
|
179 |
-
open_ai_params = OpenAIParams(
|
180 |
-
openai_key=st_openai_key,
|
181 |
-
model=st_openai_model,
|
182 |
-
api_base=st_openai_api_base,
|
183 |
-
deployment_id=st_deployment_id,
|
184 |
-
api_version=st_openai_version,
|
185 |
-
api_type=openai_api_type,
|
186 |
)
|
187 |
|
188 |
st_threshold = st.sidebar.slider(
|
189 |
-
label="
|
190 |
min_value=0.0,
|
191 |
max_value=1.0,
|
192 |
value=0.35,
|
193 |
-
help="
|
194 |
-
)
|
195 |
-
|
196 |
-
st_return_decision_process = st.sidebar.checkbox(
|
197 |
-
"Add analysis explanations to findings",
|
198 |
-
value=False,
|
199 |
-
help="Add the decision process to the output table. "
|
200 |
-
"More information can be found here: https://microsoft.github.io/presidio/analyzer/decision_process/",
|
201 |
)
|
202 |
|
203 |
-
#
|
204 |
-
|
205 |
-
"Allowlists and denylists",
|
206 |
-
expanded=False,
|
207 |
-
)
|
208 |
-
|
209 |
-
with st_deny_allow_expander:
|
210 |
-
st_allow_list = st_tags(
|
211 |
-
label="Add words to the allowlist", text="Enter word and press enter."
|
212 |
-
)
|
213 |
-
st.caption(
|
214 |
-
"Allowlists contain words that are not considered PII, but are detected as such."
|
215 |
-
)
|
216 |
|
217 |
-
|
218 |
-
|
219 |
-
)
|
220 |
-
st.caption(
|
221 |
-
"Denylists contain words that are considered PII, but are not detected as such."
|
222 |
-
)
|
223 |
-
# Main panel
|
224 |
-
|
225 |
-
with st.expander("About this demo", expanded=False):
|
226 |
-
st.info(
|
227 |
-
"""Presidio is an open source customizable framework for PII detection and de-identification.
|
228 |
-
\n\n[Code](https://aka.ms/presidio) |
|
229 |
-
[Tutorial](https://microsoft.github.io/presidio/tutorial/) |
|
230 |
-
[Installation](https://microsoft.github.io/presidio/installation/) |
|
231 |
-
[FAQ](https://microsoft.github.io/presidio/faq/) |
|
232 |
-
[Feedback](https://forms.office.com/r/9ufyYjfDaY) |"""
|
233 |
-
)
|
234 |
-
|
235 |
-
st.info(
|
236 |
-
"""
|
237 |
-
Use this demo to:
|
238 |
-
- Experiment with different off-the-shelf models and NLP packages.
|
239 |
-
- Explore the different de-identification options, including redaction, masking, encryption and more.
|
240 |
-
- Generate synthetic text with Microsoft Presidio and OpenAI.
|
241 |
-
- Configure allow and deny lists.
|
242 |
-
|
243 |
-
This demo website shows some of Presidio's capabilities.
|
244 |
-
[Visit our website](https://microsoft.github.io/presidio) for more info,
|
245 |
-
samples and deployment options.
|
246 |
-
"""
|
247 |
-
)
|
248 |
|
249 |
-
|
250 |
-
"[![Pypi Downloads](https://img.shields.io/pypi/dm/presidio-analyzer.svg)](https://img.shields.io/pypi/dm/presidio-analyzer.svg)" # noqa
|
251 |
-
"[![MIT license](https://img.shields.io/badge/license-MIT-brightgreen.svg)](https://opensource.org/licenses/MIT)"
|
252 |
-
"![GitHub Repo stars](https://img.shields.io/github/stars/microsoft/presidio?style=social)"
|
253 |
-
)
|
254 |
-
|
255 |
-
analyzer_load_state = st.info("Starting Presidio analyzer...")
|
256 |
-
|
257 |
-
analyzer_load_state.empty()
|
258 |
-
|
259 |
-
# Read default text
|
260 |
-
with open("demo_text.txt") as f:
|
261 |
-
demo_text = f.readlines()
|
262 |
-
|
263 |
-
# Create two columns for before and after
|
264 |
col1, col2 = st.columns(2)
|
265 |
|
266 |
-
#
|
267 |
-
col1.subheader("
|
268 |
st_text = col1.text_area(
|
269 |
-
label="
|
270 |
)
|
271 |
|
272 |
try:
|
273 |
-
#
|
274 |
-
st_entities_expander = st.sidebar.expander("
|
275 |
st_entities = st_entities_expander.multiselect(
|
276 |
-
label="
|
277 |
options=get_supported_entities(*analyzer_params),
|
278 |
default=list(get_supported_entities(*analyzer_params)),
|
279 |
-
help="
|
280 |
-
"This list is dynamic and based on the NER model and registered recognizers. "
|
281 |
-
"More information can be found here: https://microsoft.github.io/presidio/analyzer/adding_recognizers/",
|
282 |
)
|
283 |
|
284 |
-
#
|
285 |
-
analyzer_load_state = st.info("
|
286 |
analyzer = analyzer_engine(*analyzer_params)
|
287 |
analyzer_load_state.empty()
|
288 |
|
|
|
289 |
st_analyze_results = analyze(
|
290 |
*analyzer_params,
|
291 |
text=st_text,
|
292 |
entities=st_entities,
|
293 |
-
language="
|
294 |
score_threshold=st_threshold,
|
295 |
-
return_decision_process=
|
296 |
-
allow_list=
|
297 |
-
deny_list=
|
298 |
-
)
|
299 |
-
|
300 |
-
#
|
301 |
-
|
302 |
-
|
303 |
-
|
304 |
-
|
305 |
-
|
306 |
-
|
307 |
-
|
308 |
-
|
309 |
-
|
310 |
-
|
311 |
-
|
312 |
-
|
313 |
-
label="De-identified", value=st_anonymize_results.text, height=400
|
314 |
-
)
|
315 |
-
elif st_operator == "synthesize":
|
316 |
-
with col2:
|
317 |
-
st.subheader(f"OpenAI Generated output")
|
318 |
-
fake_data = create_fake_data(
|
319 |
-
st_text,
|
320 |
-
st_analyze_results,
|
321 |
-
open_ai_params,
|
322 |
-
)
|
323 |
-
st.text_area(label="Synthetic data", value=fake_data, height=400)
|
324 |
else:
|
325 |
-
st.subheader("Highlighted")
|
326 |
annotated_tokens = annotate(text=st_text, analyze_results=st_analyze_results)
|
327 |
-
# annotated_tokens
|
328 |
annotated_text(*annotated_tokens)
|
329 |
|
330 |
-
#
|
331 |
-
st.subheader(
|
332 |
-
"Findings"
|
333 |
-
if not st_return_decision_process
|
334 |
-
else "Findings with decision factors"
|
335 |
-
)
|
336 |
if st_analyze_results:
|
337 |
df = pd.DataFrame.from_records([r.to_dict() for r in st_analyze_results])
|
338 |
df["text"] = [st_text[res.start : res.end] for res in st_analyze_results]
|
339 |
|
340 |
df_subset = df[["entity_type", "text", "start", "end", "score"]].rename(
|
341 |
{
|
342 |
-
"entity_type": "
|
343 |
"text": "Text",
|
344 |
-
"start": "
|
345 |
-
"end": "
|
346 |
-
"score": "
|
347 |
},
|
348 |
axis=1,
|
349 |
)
|
350 |
-
df_subset["Text"] = [st_text[res.start : res.end] for res in st_analyze_results]
|
351 |
-
if st_return_decision_process:
|
352 |
-
analysis_explanation_df = pd.DataFrame.from_records(
|
353 |
-
[r.analysis_explanation.to_dict() for r in st_analyze_results]
|
354 |
-
)
|
355 |
-
df_subset = pd.concat([df_subset, analysis_explanation_df], axis=1)
|
356 |
st.dataframe(df_subset.reset_index(drop=True), use_container_width=True)
|
357 |
else:
|
358 |
-
st.text("
|
359 |
|
360 |
except Exception as e:
|
361 |
print(e)
|
362 |
traceback.print_exc()
|
363 |
-
st.error(e)
|
364 |
-
|
365 |
-
|
366 |
-
|
367 |
-
|
368 |
-
|
369 |
-
|
370 |
-
|
371 |
-
|
372 |
-
})(window, document, "clarity", "script", "h7f8bp42n8");
|
373 |
-
</script>
|
374 |
-
"""
|
375 |
-
)
|
|
|
1 |
+
"""Streamlit app pro anonymizaci českých textů s využitím Presidio."""
|
2 |
import logging
|
3 |
import os
|
4 |
import traceback
|
5 |
|
|
|
6 |
import pandas as pd
|
7 |
import streamlit as st
|
|
|
8 |
from annotated_text import annotated_text
|
9 |
from streamlit_tags import st_tags
|
10 |
|
|
|
19 |
)
|
20 |
|
21 |
st.set_page_config(
|
22 |
+
page_title="Anonymizace českých textů",
|
23 |
layout="wide",
|
24 |
initial_sidebar_state="expanded",
|
25 |
menu_items={
|
|
|
27 |
},
|
28 |
)
|
29 |
|
|
|
30 |
logger = logging.getLogger("presidio-streamlit")
|
31 |
|
|
|
|
|
|
|
|
|
32 |
# Sidebar
|
33 |
+
st.sidebar.header("Anonymizace osobních údajů v českých textech")
|
|
|
|
|
|
|
|
|
|
|
34 |
|
35 |
model_help_text = """
|
36 |
+
Vyberte model pro rozpoznávání pojmenovaných entit (NER) pro detekci osobních údajů.
|
|
|
|
|
37 |
"""
|
|
|
38 |
|
39 |
model_list = [
|
|
|
40 |
"iiiorg/piiranha-v1-detect-personal-information",
|
41 |
+
"spaCy/cs_core_news_sm",
|
42 |
]
|
43 |
+
|
|
|
|
|
44 |
st_model = st.sidebar.selectbox(
|
45 |
+
"NER model",
|
46 |
model_list,
|
47 |
+
index=0,
|
48 |
help=model_help_text,
|
49 |
)
|
50 |
|
|
|
51 |
st_model_package = st_model.split("/")[0]
|
52 |
+
st_model = "/".join(st_model.split("/")[1:])
|
53 |
|
54 |
+
analyzer_params = (st_model_package, st_model, None, None)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
55 |
logger.debug(f"analyzer_params: {analyzer_params}")
|
56 |
|
57 |
st_operator = st.sidebar.selectbox(
|
58 |
+
"Metoda anonymizace",
|
59 |
+
["nahrazení", "maskování", "zvýraznění"],
|
60 |
+
index=0,
|
61 |
help="""
|
62 |
+
Vyberte způsob anonymizace textu po identifikaci osobních údajů.\n
|
63 |
+
- Nahrazení: Nahradí osobní údaj obecným označením, např. <OSOBA>\n
|
64 |
+
- Maskování: Nahradí část osobního údaje hvězdičkami\n
|
65 |
+
- Zvýraznění: Zvýrazní osobní údaje v původním textu
|
66 |
+
""",
|
|
|
|
|
|
|
|
|
67 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
68 |
|
69 |
+
st_mask_char = "*"
|
70 |
+
st_number_of_chars = 4
|
71 |
|
72 |
+
if st_operator == "maskování":
|
73 |
st_number_of_chars = st.sidebar.number_input(
|
74 |
+
"Počet znaků k maskování", value=st_number_of_chars, min_value=0, max_value=100
|
75 |
)
|
76 |
st_mask_char = st.sidebar.text_input(
|
77 |
+
"Znak pro maskování", value=st_mask_char, max_chars=1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
78 |
)
|
79 |
|
80 |
st_threshold = st.sidebar.slider(
|
81 |
+
label="Práh přijetí",
|
82 |
min_value=0.0,
|
83 |
max_value=1.0,
|
84 |
value=0.35,
|
85 |
+
help="Definujte práh pro přijetí detekce jako osobní údaj.",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
86 |
)
|
87 |
|
88 |
+
# Hlavní panel
|
89 |
+
st.title("Anonymizace českých textů")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
90 |
|
91 |
+
# Načtení ukázkového textu
|
92 |
+
with open("demo_text.txt", "r", encoding="utf-8") as f:
|
93 |
+
demo_text = f.read()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
94 |
|
95 |
+
# Vytvoření dvou sloupců pro vstup a výstup
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
96 |
col1, col2 = st.columns(2)
|
97 |
|
98 |
+
# Vstup
|
99 |
+
col1.subheader("Vstupní text")
|
100 |
st_text = col1.text_area(
|
101 |
+
label="Zadejte text", value=demo_text, height=400, key="text_input"
|
102 |
)
|
103 |
|
104 |
try:
|
105 |
+
# Výběr entit
|
106 |
+
st_entities_expander = st.sidebar.expander("Vyberte entity k detekci")
|
107 |
st_entities = st_entities_expander.multiselect(
|
108 |
+
label="Které entity hledat?",
|
109 |
options=get_supported_entities(*analyzer_params),
|
110 |
default=list(get_supported_entities(*analyzer_params)),
|
111 |
+
help="Omezte seznam detekovaných osobních údajů.",
|
|
|
|
|
112 |
)
|
113 |
|
114 |
+
# Inicializace analyzátoru
|
115 |
+
analyzer_load_state = st.info("Spouštění Presidio analyzátoru...")
|
116 |
analyzer = analyzer_engine(*analyzer_params)
|
117 |
analyzer_load_state.empty()
|
118 |
|
119 |
+
# Analýza textu
|
120 |
st_analyze_results = analyze(
|
121 |
*analyzer_params,
|
122 |
text=st_text,
|
123 |
entities=st_entities,
|
124 |
+
language="cs",
|
125 |
score_threshold=st_threshold,
|
126 |
+
return_decision_process=False,
|
127 |
+
allow_list=[],
|
128 |
+
deny_list=[],
|
129 |
+
)
|
130 |
+
|
131 |
+
# Výstup
|
132 |
+
col2.subheader("Výstup")
|
133 |
+
if st_operator != "zvýraznění":
|
134 |
+
st_anonymize_results = anonymize(
|
135 |
+
text=st_text,
|
136 |
+
operator=st_operator,
|
137 |
+
mask_char=st_mask_char,
|
138 |
+
number_of_chars=st_number_of_chars,
|
139 |
+
analyze_results=st_analyze_results,
|
140 |
+
)
|
141 |
+
col2.text_area(
|
142 |
+
label="Anonymizovaný text", value=st_anonymize_results.text, height=400
|
143 |
+
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
144 |
else:
|
|
|
145 |
annotated_tokens = annotate(text=st_text, analyze_results=st_analyze_results)
|
|
|
146 |
annotated_text(*annotated_tokens)
|
147 |
|
148 |
+
# Tabulka s výsledky
|
149 |
+
st.subheader("Nalezené osobní údaje")
|
|
|
|
|
|
|
|
|
150 |
if st_analyze_results:
|
151 |
df = pd.DataFrame.from_records([r.to_dict() for r in st_analyze_results])
|
152 |
df["text"] = [st_text[res.start : res.end] for res in st_analyze_results]
|
153 |
|
154 |
df_subset = df[["entity_type", "text", "start", "end", "score"]].rename(
|
155 |
{
|
156 |
+
"entity_type": "Typ entity",
|
157 |
"text": "Text",
|
158 |
+
"start": "Začátek",
|
159 |
+
"end": "Konec",
|
160 |
+
"score": "Důvěryhodnost",
|
161 |
},
|
162 |
axis=1,
|
163 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
164 |
st.dataframe(df_subset.reset_index(drop=True), use_container_width=True)
|
165 |
else:
|
166 |
+
st.text("Žádné osobní údaje nebyly nalezeny.")
|
167 |
|
168 |
except Exception as e:
|
169 |
print(e)
|
170 |
traceback.print_exc()
|
171 |
+
st.error(f"Došlo k chybě: {str(e)}")
|
172 |
+
|
173 |
+
# Informace o aplikaci
|
174 |
+
st.sidebar.markdown("---")
|
175 |
+
st.sidebar.subheader("O aplikaci")
|
176 |
+
st.sidebar.info(
|
177 |
+
"Tato aplikace anonymizuje osobní údaje v českých textech. "
|
178 |
+
"Využívá Microsoft Presidio a pokročilé NLP techniky pro detekci a anonymizaci PII."
|
179 |
+
)
|
|
|
|
|
|
|
|