Upload app.py with huggingface_hub

app.py

@@ -1,32 +1,20 @@
 
-## required lib, required "pip install"
-# import transformers
-# import accelerate
 import openai
 import llama_index
 import torch
 import cryptography
 import cryptography.fernet
-## interface libs, required "pip install"
 import gradio
 import huggingface_hub
 import huggingface_hub.hf_api
-## standard libs, no need to install
 import json
-import requests
 import time
 import os
 import random
-import re
 import sys
 import psutil
-import threading
 import socket
-# import PIL
-# import pandas
-import matplotlib
 class HFace_Pluto(object):
-  #
   # initialize the object
   def __init__(self, name="Pluto",*args, **kwargs):
     super(HFace_Pluto, self).__init__(*args, **kwargs)
@@ -91,23 +79,6 @@ class HFace_Pluto(object):
       y = x
     return y
   #
-  # fetch huggingface file
-  def fetch_hface_files(self,
-    hf_names,
-    hf_space="duchaba/monty",
-    local_dir="/content/"):
-    f = str(hf_names) + " is not iteratable, type: " + str(type(hf_names))
-    try:
-      for f in hf_names:
-        lo = local_dir + f
-        huggingface_hub.hf_hub_download(repo_id=hf_space, filename=f,
-          use_auth_token=True,repo_type=huggingface_hub.REPO_TYPE_SPACE,
-          force_filename=lo)
-    except:
-      self._pp("*Error", f)
-    return
-  #
-  #
   def push_hface_files(self,
     hf_names,
     hf_space="duchaba/skin_cancer_diagnose",
@@ -133,50 +104,12 @@ class HFace_Pluto(object):
       repo_type="space")
     return
   #
-  # Define a function to display available CPU and RAM
-  def fetch_system_info(self):
-    s=''
-    # Get CPU usage as a percentage
-    cpu_usage = psutil.cpu_percent()
-    # Get available memory in bytes
-    mem = psutil.virtual_memory()
-    # Convert bytes to gigabytes
-    mem_total_gb = mem.total / (1024 ** 3)
-    mem_available_gb = mem.available / (1024 ** 3)
-    mem_used_gb = mem.used / (1024 ** 3)
-    # Print the results
-    s += f"CPU usage: {cpu_usage}%\n"
-    s += f"Total memory: {mem_total_gb:.2f} GB\n"
-    s += f"Available memory: {mem_available_gb:.2f} GB\n"
-    # print(f"Used memory: {mem_used_gb:.2f} GB")
-    s += f"Memory usage: {mem_used_gb/mem_total_gb:.2f}%\n"
-    return s
-  #
-  def restart_script_periodically(self):
-    while True:
-      #random_time = random.randint(540, 600)
-      random_time = random.randint(15800, 21600)
-      time.sleep(random_time)
-      os.execl(sys.executable, sys.executable, *sys.argv)
-    return
-  #
   def write_file(self,fname, txt):
     f = open(fname, "w")
     f.writelines("\n".join(txt))
     f.close()
     return
   #
-  def fetch_gpu_info(self):
-    s=''
-    try:
-      s += f'Your GPU is the {torch.cuda.get_device_name(0)}\n'
-      s += f'GPU ready staus {torch.cuda.is_available()}\n'
-      s += f'GPU allocated RAM: {round(torch.cuda.memory_allocated(0)/1024**3,1)} GB\n'
-      s += f'GPU reserved RAM {round(torch.cuda.memory_reserved(0)/1024**3,1)} GB\n'
-    except Exception as e:
-      s += f'**Warning, No GPU: {e}'
-    return s
-  #
   def _fetch_crypt(self,is_generate=False):
     s=self._fkey[::-1]
     if (is_generate):
@@ -208,25 +141,6 @@ class HFace_Pluto(object):
     self._ph()
     return
   #
-  def _fetch_version(self):
-    s = ''
-    # print(f"{'torch: 2.0.1':<25} Actual: {torch.__version__}")
-    # print(f"{'transformers: 4.29.2':<25} Actual: {transformers.__version__}")
-    s += f"{'openai: 0.27.7,':<28} Actual: {openai.__version__}\n"
-    s += f"{'huggingface_hub: 0.14.1,':<28} Actual: {huggingface_hub.__version__}\n"
-    s += f"{'gradio: 3.32.0,':<28} Actual: {gradio.__version__}\n"
-    s += f"{'cryptography: 3.34.0,':<28} Actual: {cryptography.__version__}\n"
-    s += f"{'llama_index: 0.6.21.post1,':<28} Actual: {llama_index.__version__}\n"
-    return s
-  #
-  def _fetch_host_ip(self):
-    s=''
-    hostname = socket.gethostname()
-    ip_address = socket.gethostbyname(hostname)
-    s += f"Hostname: {hostname}\n"
-    s += f"IP Address: {ip_address}\n"
-    return s
-  #
   def _setup_openai(self,key=None):
     if (key is None):
       key = self._decrypt_it(self._gpt_key)
@@ -259,28 +173,9 @@ def add_method(cls):
 #
 monty = HFace_Pluto("Monty")
 monty._login_hface()
-# print(monty._fetch_version())
-# monty._ph()
-# print(monty.fetch_system_info())
-# monty._ph()
-# print(monty.fetch_gpu_info())
-# monty._ph()
-# print(monty._fetch_host_ip())
 monty._ph()
 monty._setup_openai()
 
-@add_method(HFace_Pluto)
-def gen_llama_index(self, doc_path, vindex='vector_index', vpath='./index_storage'):
-  # load doc
-  doc = llama_index.SimpleDirectoryReader(doc_path).load_data()
-  # need openai key
-  self._llama_index_doc = llama_index.VectorStoreIndex.from_documents(doc)
-  # save index to disk
-  self._llama_index_doc.set_index_id(vindex)
-  self._llama_index_doc.storage_context.persist(vpath)
-  print(f'Index doc are: {self._fetch_index_files(self._llama_index_doc)}')
-  return
-
 @add_method(HFace_Pluto)
 def load_llama_index(self,vindex='vector_index',vpath='./index_storage'):
   try:
@@ -297,49 +192,26 @@ def load_llama_index(self,vindex='vector_index',vpath='./index_storage'):
 monty.load_llama_index()
 
 @add_method(HFace_Pluto)
-def ask_me(self, p, ll_sign_in_member='Girish'):
-  # todo: make this into a dict so not to use if/else
-  if (self._llama_query_engine is None):
-    self._llama_query_engine = self._llama_index_doc.as_query_engine()
-
+def ask_me(self, p, ll_sign_in_member='Girish' , ll_engine='Humana'):
+  self._llama_query_engine = self._llama_index_doc.as_query_engine()
   ll_engine = self._llama_query_engine
 
   px = f'My name is {ll_sign_in_member}, and I want answer to the following: {p}.'
   print("##### " + px)
+  
   resp = ll_engine.query(px)
-
   return resp
-# @add_method(HFace_Pluto)
-def ask_me(self, p):
-  if (self._llama_query_engine is None):
-    self._llama_query_engine = self._llama_index_doc.as_chat_engine()
-  resp = self._llama_query_engine.chat(p)
-  return resp
-
-p = 'How is my Humana plan chaning?'
-resp = monty.ask_me(p)
-print(resp)
-
-gradio.Interface(fn=monty.ask_me,
-  inputs=in_box,
-  outputs=out_box,
-  examples=exp,
-  title=title,
-  description=desc,
-  article=arti,
-  flagging_options=flag_options).launch(debug=True)
 
 in_box = [gradio.Textbox(lines=1, label="Your Humana request", placeholder="Your Humana request...see example if you need help.")
-  ,gradio.Radio(["Duc", "Girish", "Mayo"], label="Login Member", value='Duc', info="Who had login?")
-  ,gradio.Radio(["Humana", "Kohli Fighters", "Ally Bank"], label="Login Member", value='Humana', info="Fine-Tune LLM for:")
+  ,gradio.Radio(["Girish"], label="Login Member", value='Girish', info="Who had login?")
+  ,gradio.Radio(["Humana"], label="Login Member", value='Humana', info="Fine-Tune LLM for:")
   ]
 out_box = [gradio.Textbox(label="Humana response:")]
 #
 
-title = "Humana and YML Fine-tune LLM model"
+title = "Humana and C&T Fine-tune LLM model"
 desc = '*Note: This model is fine-tuned by YML using GPT3.5 as the base LLM.'
-arti = f'<ul><li>The documents for fine-tuning are:</li>{monty.self._fetch_index_files(monty._llama_index_doc)}'
-arti += '<li><i>**Note: You can add more documentation. The more documentation the model has the smarter it will be.</i></li></ul>'
+arti = '<li><i>**Note: You can add more documentation. The more documentation the model has the smarter it will be.</i></li></ul>'
 exp = [
   ['Tell me the Humana Gold Plus plan.'],
   ['Please write a summary in bullet point of the Humana Gold Plus SNP-DE H0028-015 (HMO-POS D-SNP) Annual Notice of Changes for 2023.'],
@@ -362,5 +234,4 @@ gradio.Interface(fn=monty.ask_me,
   examples=exp,
   title=title,
   description=desc,
-  article=arti,
-  flagging_options=flag_options).launch(debug=True)
+  article=arti).launch(debug=True)