Update pattern_analyzer.py

pattern_analyzer.py

@@ -1,5 +1,5 @@
 import os
-os.environ['HF_HOME'] = '/tmp/huggingface'
+os.environ['TRANSFORMERS_CACHE'] = '/tmp/transformers_cache'
 from transformers import AutoModelForCausalLM, AutoTokenizer
 import numpy as np
 import pandas as pd
@@ -9,12 +9,14 @@ from pattern_logic import PatternLogic
 class PatternAnalyzer:
     def __init__(self):
         self.model = AutoModelForCausalLM.from_pretrained(
-            "http://localhost:5000/codellama-chart-model",
-            load_in_8bit=True,
-            device_map="auto",
-            torch_dtype="auto"
+            "tmmdev/codellama-pattern-analysis",
+            load_in_8bit=True,  # Enable 8-bit quantization
+            device_map="auto",   # Optimize device usage
+            torch_dtype="auto"   # Automatic precision selection
         )
-        self.tokenizer = AutoTokenizer.from_pretrained("http://localhost:5000/codellama-chart-model")
+        self.tokenizer = AutoTokenizer.from_pretrained("tmmdev/codellama-pattern-analysis")
+
+
         self.basic_patterns = {
             'channel': {'min_points': 4, 'confidence_threshold': 0.7},
             'triangle': {'min_points': 3, 'confidence_threshold': 0.75},
@@ -26,7 +28,7 @@ class PatternAnalyzer:
         self.pattern_logic = PatternLogic()
 
     def analyze_data(self, ohlcv_data):
-        data_prompt = f"""TASK: Identify high-confidence technical patterns only.
+        data_prompt = f"""TASK: Identify high-confidence technical patterns only. 
         Minimum confidence threshold: 0.8
         Required pattern criteria:
         1. Channel: Must have at least 3 touching points
@@ -56,6 +58,7 @@ class PatternAnalyzer:
             
             for pattern in analysis_data.get('patterns', []):
                 pattern_type = pattern.get('type')
+                
                 if pattern_type in self.basic_patterns:
                     threshold = self.basic_patterns[pattern_type]['confidence_threshold']
                     if pattern.get('confidence', 0) >= threshold:
@@ -68,6 +71,8 @@ class PatternAnalyzer:
                                 'timestamp': pd.Timestamp.now().isoformat()
                             }
                         })
+            
             return patterns
+            
         except json.JSONDecodeError:
             return []