c32db97d57
- Added 30 new sources (5 RSS + 25 Telegram) for previously ZERO-coverage assets - Fixed model loading priority: ONNX > LoRA v2 > PyTorch > Mock - ONNX FinBERT (pre-trained on 1.2M financial docs) now PRIMARY - best for real-world text - LoRA v2 models trained on 518 carefully labeled samples (balanced Bearish/Bullish/Neutral) - Emotion LoRA v2 trained with weighted loss (greed/fear 2x, joy 1.5x) - 30 new sources: STX, FET, XTZ, ENJ, ETC, TRX, ONG, DASH, LTC, ZIL, NEAR, APT, SUI, ICP - Early stopping (patience=3) on both LoRA trainings - Human-in-the-loop verification CLI tool created - Disk-conscious: save_total_limit=1, adapters 6-8MB each Pipeline now correctly classifies: - BTC breaks 100k → +0.54 Bullish ✅ - Major hack → -0.23 Bearish ✅ - HODL → +0.91 Bullish ✅ - Rug pull → -0.30 Bearish ✅ - SEC sues → -0.30 Bearish ✅ - ETF approval → +0.32 Bullish ✅ - Whale accumulation → +0.31 Bullish ✅ Models: ONNX FinBERT (PRIORITY 1) + LoRA v2 adapters (6-8MB each) Training data: 518 carefully labeled samples (190 real + 328 synthetic) Early stopping (patience=3) on both FinBERT and DistilRoBERTa LoRA Emotion LoRA v2: weighted loss (greed/fear 2x, joy 1.5x) + early stopping
53 lines
2.5 KiB
Python
53 lines
2.5 KiB
Python
# Patch for labeling_pipeline.py - fix SEC approval vs enforcement distinction
|
|
import re
|
|
|
|
# Read the file
|
|
with open('/mnt/dolphinng5_predict/sentiment_engine/labeling_pipeline.py', 'r') as f:
|
|
content = f.read()
|
|
|
|
# Update BULLISH_PATTERNS to include SEC approval
|
|
old_bullish = ''' BULLISH_PATTERNS = [
|
|
r"\\b(surge|surge|moon|pump|bullish|breakout|ath|all.time.high)\\b",
|
|
r"\\b(institutional|adoption|etf|accumulate|long|longing)\\b",
|
|
r"\\b(golden.cross|breakout|bullish|rally|surge|rally)\\b",
|
|
r"\\b(etf.approval|etf.approved|inflows|institutional.buying|whale.accumulation)\\b",
|
|
r"[🚀📈💎🙌🌙]",
|
|
]'''
|
|
|
|
new_bullish = ''' BULLISH_PATTERNS = [
|
|
r"\\b(surge|surge|moon|pump|bullish|breakout|ath|all.time.high)\\b",
|
|
r"\\b(institutional|adoption|etf|accumulate|long|longing)\\b",
|
|
r"\\b(golden.cross|breakout|bullish|rally|surge|rally)\\b",
|
|
r"\\b(etf.approval|etf.approved|inflows|institutional.buying|whale.accumulation)\\b",
|
|
r"\\b(sec.approves|sec.approved|sec.approval|approved.etf|etf.approved)\\b",
|
|
r"[🚀📈💎🙌🌙]",
|
|
]'''
|
|
|
|
content = content.replace(old_bullish, new_bullish)
|
|
|
|
# Update BEARISH_PATTERNS to be more specific about SEC actions (enforcement vs approval)
|
|
old_bearish = ''' BEARISH_PATTERNS = [
|
|
r"\\b(crash|crash|dump|bearish|panic|rekt|short|shorting)\\b",
|
|
r"\\b(hack|exploit|drain|stolen|rug|rugpull|scam|depeg|depegged|depegs|depegging)\\b",
|
|
r"\\b(death.cross|breakdown|capitulation|liquidation|peg.loss|depeg|depegged|depegs)\\b",
|
|
r"\\b(sec|lawsuit|enforcement|regulation|regulatory|cftc|ban|delist|crackdown|subpoena|investigation|charges|sues)\\b",
|
|
r"[📉😭💀🩸🧻]",
|
|
]'''
|
|
|
|
new_bearish = ''' BEARISH_PATTERNS = [
|
|
r"\\b(crash|crash|dump|bearish|panic|rekt|short|shorting)\\b",
|
|
r"\\b(hack|exploit|drain|stolen|rug|rugpull|scam|depeg|depegged|depegs|depegging)\\b",
|
|
r"\\b(death.cross|breakdown|capitulation|liquidation|peg.loss|depeg|depegged|depegs)\\b",
|
|
r"\\b(lawsuit|enforcement|crackdown|subpoena|investigation|charges|sues|sues.sec|sec.sues|sec.charges|cf tc.ban|regulatory.ban)\\b",
|
|
r"\\b(regulation|regulatory|cftc|ban|delist)\\b",
|
|
r"[📉😭💀🩸🧻]",
|
|
]'''
|
|
|
|
content = content.replace(old_bearish, new_bearish)
|
|
|
|
# Write the patched file
|
|
with open('/mnt/dolphinng5_predict/sentiment_engine/labeling_pipeline.py', 'w') as f:
|
|
f.write(content)
|
|
|
|
print("Patch 3 applied successfully!")
|