removed the space config

app.py

@@ -1,7 +1,7 @@
 import torch
 import gradio as gr
 from transformers import AutoTokenizer, AutoModelForCausalLM, TextIteratorStreamer
-from huggingface_hub import login, HfApi, HfFolder, Repository
+from huggingface_hub import login
 import os
 from threading import Thread
 from openai import OpenAI
@@ -10,46 +10,6 @@ TOKEN = os.getenv('HF_AUTH_TOKEN')
 login(token=TOKEN,
       add_to_git_credential=False)
 
-# Authenticate with hf api
-api = HfApi()
-
-# switch hardware function
-def space_hardware_config(instance_size: str="gpu",
-                          instance_type: str="1xL4",
-                          vcpus: int=8,
-                          memory: int=30):
-    """
-    This will manually select what hardware we'll use in the space.
-    """
-    
-    api = HfApi()
-    token = HfFolder.get_token()
-    if token is None:
-        raise ValueError("Hugging Face token not found. Please log in using huggingface-cli or set the token manually.")
-    
-    space_id = os.getenv("SPACE_ID")
-    if not space_id:
-        raise ValueError("SPACE_ID environment variable not found.")
-    
-    space_info = api.repo_info(repo_id=space_id, repo_type="space", token=token)
-    print(space_info)
-
-    # # Hardware Configuration
-    # space.config["compute"] = {
-    #     "instance_type": instance_type,
-    #     "instance_size": instance_size,
-    #     "disk_size": 50,
-    #     "vcpus": vcpus, # number of virtual CPU's
-    #     "memory": memory # amount of memory in gb
-    # }
-
-    # # Save updated space config
-    # api.push_to_hub(space)
-    # print("Hardware configuration successfull. Check the cuda command.")
-
-# Automatically place to the standard config we need for loki
-space_hardware_config()
-
 # Open ai api key
 API_KEY = os.getenv('OPEN_AI_API_KEY')
 
@@ -68,7 +28,6 @@ terminators = [
     llama_tokenizer.convert_tokens_to_ids("<|eot_id|>")
 ]
 
-
 # The output
 def output_list(output: list):
     """
@@ -122,10 +81,6 @@ def loki_generation(input_text: str,
     """
     Pass input texts, tokenize, output and back to text.
     """
-    space_hardware_config(instance_size="gpu",
-                          instance_type="1xL4",
-                          vcpus=8,
-                          memory=30)
     if mode == "llama":
         conversation = []
         for user, assistant in history:
@@ -259,10 +214,6 @@ def bot_comms(input_text: str,
             yield "".join(outputs)
 
     if llm_mode == "switch to gpt-4o":
-        space_hardware_config(instance_size="cpu",
-                              instance_type="basic",
-                              vcpus=2,
-                              memory=16)
         stream = gpt_generation(input=input_text,
                                 llama_output="",
                                 mode="gpt-4o")
@@ -275,10 +226,6 @@ def bot_comms(input_text: str,
                 yield "".join(outputs)
 
     if llm_mode == "switch to gpt-3.5-turbo":
-        space_hardware_config(instance_size="cpu",
-                              instance_type="basic",
-                              vcpus=2,
-                              memory=16)
         stream = gpt_generation(input=input_text,
                                 llama_output="",
                                 mode="gpt-3.5-turbo")