install cuda toolkit

app.py

@@ -9,6 +9,25 @@ import torch
 import json
 import bs4
 
+def install_cuda_toolkit():
+    CUDA_TOOLKIT_URL = "https://developer.download.nvidia.com/compute/cuda/12.4.0/local_installers/cuda_12.4.0_550.54.14_linux.run"
+    # CUDA_TOOLKIT_URL = "https://developer.download.nvidia.com/compute/cuda/12.2.0/local_installers/cuda_12.2.0_535.54.03_linux.run"
+    CUDA_TOOLKIT_FILE = "/tmp/%s" % os.path.basename(CUDA_TOOLKIT_URL)
+    subprocess.call(["wget", "-q", CUDA_TOOLKIT_URL, "-O", CUDA_TOOLKIT_FILE])
+    subprocess.call(["chmod", "+x", CUDA_TOOLKIT_FILE])
+    subprocess.call([CUDA_TOOLKIT_FILE, "--silent", "--toolkit"])
+    
+    os.environ["CUDA_HOME"] = "/usr/local/cuda"
+    os.environ["PATH"] = "%s/bin:%s" % (os.environ["CUDA_HOME"], os.environ["PATH"])
+    os.environ["LD_LIBRARY_PATH"] = "%s/lib:%s" % (
+        os.environ["CUDA_HOME"],
+        "" if "LD_LIBRARY_PATH" not in os.environ else os.environ["LD_LIBRARY_PATH"],
+    )
+    # Fix: arch_list[-1] += '+PTX'; IndexError: list index out of range
+    os.environ["TORCH_CUDA_ARCH_LIST"] = "8.0;8.6"
+
+install_cuda_toolkkit()
+
 product_strings = []
 with open('./Data/products.json', 'r', encoding='utf-8') as f:
     for product in json.load(f)['products']:
@@ -232,7 +251,6 @@ def respond(
     print(f"Frage Nutzer: {query}\n Antwort Assistent: {answer}")
     return answer
 
-
 """
 For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
 """