Collabs/Veracity_AI
Collabs
/
Veracity_AI
3
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

llm optional gemacht

This commit is contained in:
Falko Victor Habel 2024-10-15 21:57:24 +02:00
parent 7a1aae3d45
commit 69fb3eb406
1 changed files with 20 additions and 52 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

72
src/controller/mainFrameController.py
View File

@ -7,12 +7,7 @@ from utils.database.database import FakeNewsChecker
from models.provider import Provider
from collections import Counter
from Ai.llm import ArticleRater
BAD_WORDS = ["FAKE", "SATIRE", "Fake", "fake", "fake news", "Fake News", "FakeNews"]
GOOD_WORDS = ["REAL", "real ", "Real", "Reale News", "reale", "reale News", "realen", "Real News"]
BAD_COLOR = "#ff8080"
GOOD_COLOR = "#80ff8f"
WORDS = BAD_WORDS + GOOD_WORDS
from Ai.Token import get_token
class MainFrameController:
@ -47,63 +42,36 @@ class MainFrameController:
return text_data
def press_check_button(self):
self.frame.result_label.configure(text="", fg_color="#333333")
self.frame.confidence_label.configure(text="", fg_color="#333333")
text_data = self.get_text_data()
if not text_data.text.strip():
return
text_data = self._predict(text_data)
self._add_to_db(text_data)
self.frame.output_textbox.configure(state="normal")
self.frame.output_textbox.delete("0.0", "end")
response_stream = self.rater.get_response(text_data.text, text_data.result, float(f"{text_data.confidence * 100:.2f}"))
confidence = text_data.confidence * 100
self.frame.confidence_label.configure(text=f"{confidence:.2f}%")
highlight_buffer = deque(maxlen=5)
result_color = "green" if text_data.result == "REAL" else "red"
self.frame.result_label.configure(text=text_data.result, fg_color=result_color)
for chunk in response_stream:
# Display the chunk immediately
self.frame.output_textbox.insert("end", chunk)
self.frame.output_textbox.see("end")
self.frame.update_idletasks()
confidence_color = "green" if confidence > 80 else ("orange" if confidence > 50 else "red")
self.frame.confidence_label.configure(fg_color=confidence_color)
if get_token().strip():
response_stream = self.rater.get_response(text_data.text, text_data.result, confidence)
# Add to highlight buffer
highlight_buffer.append(chunk)
for chunk in response_stream:
self.frame.output_textbox.insert("end", chunk)
self.frame.output_textbox.see("end")
self.frame.update_idletasks()
# Process highlighting when buffer is full
if len(highlight_buffer) == 5:
self._process_highlighting(highlight_buffer)
# Process any remaining chunks in the buffer
if highlight_buffer:
self._process_highlighting(highlight_buffer)
self.frame.output_textbox.configure(state="disabled")
self.update_provider_list()
def _process_highlighting(self, highlight_buffer):
start_index = self.frame.output_textbox.index(f"end-{sum(len(c) for c in highlight_buffer)}c")
end_index = self.frame.output_textbox.index("end")
self._highlight_words(start_index, end_index)
# Keep overlap of 2 chunks
highlight_buffer = deque(list(highlight_buffer)[-3:], maxlen=5)
def _highlight_words(self, start_index, end_index):
content = self.frame.output_textbox.get(start_index, end_index)
for word in WORDS:
start = 0
while True:
pos = content.find(word, start)
if pos == -1:
break
word_start = f"{start_index}+{pos}c"
word_end = f"{word_start}+{len(word)}c"
tag_name = f"{word.lower()}_color"
self.frame.output_textbox.tag_add(tag_name, word_start, word_end)
if word in BAD_WORDS:
self.frame.output_textbox.tag_config(tag_name, foreground=BAD_COLOR)
elif word in GOOD_WORDS:
self.frame.output_textbox.tag_config(tag_name, foreground=GOOD_COLOR)
start = pos + len(word)
def _predict(self, text_data: TextData) -> TextData:
"""
Make a prediction using the VeraMind model.