Update app.py

app.py

@@ -25,15 +25,18 @@ model_cache_dir = os.path.join(base_dir, "txagent_models")
 tool_cache_dir = os.path.join(base_dir, "tool_cache")
 file_cache_dir = os.path.join(base_dir, "cache")
 report_dir = os.path.join(base_dir, "reports")
+vllm_cache_dir = os.path.join(base_dir, "vllm_cache")
 
 os.makedirs(model_cache_dir, exist_ok=True)
 os.makedirs(tool_cache_dir, exist_ok=True)
 os.makedirs(file_cache_dir, exist_ok=True)
 os.makedirs(report_dir, exist_ok=True)
+os.makedirs(vllm_cache_dir, exist_ok=True)
 
 os.environ.update({
     "TRANSFORMERS_CACHE": model_cache_dir,
     "HF_HOME": model_cache_dir,
+    "VLLM_CACHE_DIR": vllm_cache_dir,
     "TOKENIZERS_PARALLELISM": "false",
     "CUDA_LAUNCH_BLOCKING": "1"
 })
@@ -60,7 +63,7 @@ def extract_priority_pages(file_path: str, max_pages: int = 20) -> str:
                 text_chunks.append(f"=== Page {i+1} ===\n{(page.extract_text() or '').strip()}")
             for i, page in enumerate(pdf.pages[3:max_pages], start=4):
                 page_text = page.extract_text() or ""
-                if any(re.search(rf'\\b{kw}\\b', page_text.lower()) for kw in MEDICAL_KEYWORDS):
+                if any(re.search(rf'\b{kw}\b', page_text.lower()) for kw in MEDICAL_KEYWORDS):
                     text_chunks.append(f"=== Page {i} ===\n{page_text.strip()}")
         return "\n\n".join(text_chunks)
     except Exception as e: