Feat: GPT4All Model Integration

README.md

@@ -14,8 +14,11 @@ To set up your environment, you will need to generate a `utils.py` file that con
 ### Step 1. Generate Utils File
 In the `reverie/backend_server` folder (where `reverie.py` is located), create a new file titled `utils.py` and copy and paste the content below into the file:
 ```
-# Copy and paste your OpenAI API Key
-openai_api_key = "<Your OpenAI API>"
+# Select the GPT4All Model you'll use for the simulation. See: https://observablehq.com/@simonw/gpt4all-models
+gpt4all_model="orca-mini-3b.ggmlv3.q4_0.bin"
+max_tokens = 30
+temperature = 0.5
+
 # Put your name
 key_owner = "<Name>"
 
@@ -31,7 +34,7 @@ collision_block_id = "32125"
 # Verbose 
 debug = True
 ```
-Replace `<Your OpenAI API>` with your OpenAI API key, and `<name>` with your name.
+Replace `<name>` with your name.
 
 ### Step 2. Install requirements.txt
 Install everything listed in the `requirements.txt` file (I strongly recommend first setting up a virtualenv as usual). A note on Python version: we tested our environment on Python 3.9.12. 
@@ -112,12 +115,12 @@ For a more involved customization, you will need to author your own base simulat
 
 ## <img src="https://joonsungpark.s3.amazonaws.com:443/static/assets/characters/profile/Eddy_Lin.png" alt="Generative Eddy">   Authors and Citation 
 
-**Authors:** Joon Sung Park, Joseph C. O'Brien, Carrie J. Cai, Meredith Ringel Morris, Percy Liang, Michael S. Bernstein
+**Authors:** Cassini Saturn, Joon Sung Park, Joseph C. O'Brien, Carrie J. Cai, Meredith Ringel Morris, Percy Liang, Michael S. Bernstein
 
 Please cite our paper if you use the code or data in this repository. 
 ```
 @inproceedings{Park2023GenerativeAgents,  
-author = {Park, Joon Sung and O'Brien, Joseph C. and Cai, Carrie J. and Morris, Meredith Ringel and Liang, Percy and Bernstein, Michael S.},  
+author = {Cassini, Saturn and Park, Joon Sung and O'Brien, Joseph C. and Cai, Carrie J. and Morris, Meredith Ringel and Liang, Percy and Bernstein, Michael S.},  
 title = {Generative Agents: Interactive Simulacra of Human Behavior},  
 year = {2023},  
 publisher = {Association for Computing Machinery},  

requirements.txt

@@ -28,7 +28,7 @@ matplotlib==3.7.2
 multidict==6.0.4
 nltk==3.6.5
 numpy==1.25.2
-openai==0.27.0
+gpt4all==1.0.5
 outcome==1.2.0
 packaging==23.0
 pandas==2.0.3

reverie/backend_server/persona/prompt_template/gpt_structure.py

@@ -1,81 +1,70 @@
 """
-Author: Joon Sung Park ([email protected])
+Author: 
+Cassini Saturn ([email protected])
+Joon Sung Park ([email protected])
 
 File: gpt_structure.py
-Description: Wrapper functions for calling OpenAI APIs.
+Description: Wrapper functions for calling GPT4All APIs.
 """
 import json
-import random
-import openai
 import time 
 
 from utils import *
 
-openai.api_key = openai_api_key
+from gpt4all import GPT4All, Embed4All
+
+model = GPT4All(gpt4all_model)
 
 def temp_sleep(seconds=0.1):
   time.sleep(seconds)
 
 def ChatGPT_single_request(prompt): 
   temp_sleep()
 
-  completion = openai.ChatCompletion.create(
-    model="gpt-3.5-turbo", 
-    messages=[{"role": "user", "content": prompt}]
-  )
-  return completion["choices"][0]["message"]["content"]
-
+  output = model.generate(prompt, max_tokens=max_tokens)
+  return output
 
 # ============================================================================
-# #####################[SECTION 1: CHATGPT-3 STRUCTURE] ######################
+# #####################[SECTION 1: CHATGPT4All STRUCTURE] ######################
 # ============================================================================
 
-def GPT4_request(prompt): 
+def GPT4All_request(prompt): 
   """
-  Given a prompt and a dictionary of GPT parameters, make a request to OpenAI
-  server and returns the response. 
+  Given a prompt, make a request to GPT4All model and returns the response.
   ARGS:
     prompt: a str prompt
     gpt_parameter: a python dictionary with the keys indicating the names of  
                    the parameter and the values indicating the parameter 
                    values.   
   RETURNS: 
-    a str of GPT-3's response. 
+    a str of GPT4All's response. 
   """
   temp_sleep()
 
   try: 
-    completion = openai.ChatCompletion.create(
-    model="gpt-4", 
-    messages=[{"role": "user", "content": prompt}]
-    )
-    return completion["choices"][0]["message"]["content"]
-
+    output = model.generate(prompt, max_tokens=max_tokens)
+    return output
   except: 
     print ("ChatGPT ERROR")
     return "ChatGPT ERROR"
 
 
 def ChatGPT_request(prompt): 
   """
-  Given a prompt and a dictionary of GPT parameters, make a request to OpenAI
-  server and returns the response. 
+  Given a prompt, make a request to GPT4All model
+
   ARGS:
     prompt: a str prompt
     gpt_parameter: a python dictionary with the keys indicating the names of  
                    the parameter and the values indicating the parameter 
                    values.   
   RETURNS: 
-    a str of GPT-3's response. 
+    a str of GPT4All's response. 
   """
   # temp_sleep()
   try: 
-    completion = openai.ChatCompletion.create(
-    model="gpt-3.5-turbo", 
-    messages=[{"role": "user", "content": prompt}]
-    )
-    return completion["choices"][0]["message"]["content"]
-
+    output = model.generate(prompt, max_tokens=max_tokens)
+    return output
   except: 
     print ("ChatGPT ERROR")
     return "ChatGPT ERROR"
@@ -89,19 +78,19 @@ def GPT4_safe_generate_response(prompt,
                                    func_validate=None,
                                    func_clean_up=None,
                                    verbose=False): 
-  prompt = 'GPT-3 Prompt:\n"""\n' + prompt + '\n"""\n'
+  prompt = 'GPT4All Prompt:\n"""\n' + prompt + '\n"""\n'
   prompt += f"Output the response to the prompt above in json. {special_instruction}\n"
   prompt += "Example output json:\n"
   prompt += '{"output": "' + str(example_output) + '"}'
 
   if verbose: 
-    print ("CHAT GPT PROMPT")
+    print ("GPT4All PROMPT")
     print (prompt)
 
   for i in range(repeat): 
 
     try: 
-      curr_gpt_response = GPT4_request(prompt).strip()
+      curr_gpt_response = GPT4All_request(prompt).strip()
       end_index = curr_gpt_response.rfind('}') + 1
       curr_gpt_response = curr_gpt_response[:end_index]
       curr_gpt_response = json.loads(curr_gpt_response)["output"]
@@ -128,14 +117,14 @@ def ChatGPT_safe_generate_response(prompt,
                                    func_validate=None,
                                    func_clean_up=None,
                                    verbose=False): 
-  # prompt = 'GPT-3 Prompt:\n"""\n' + prompt + '\n"""\n'
+  # prompt = 'GPT4All Prompt:\n"""\n' + prompt + '\n"""\n'
   prompt = '"""\n' + prompt + '\n"""\n'
   prompt += f"Output the response to the prompt above in json. {special_instruction}\n"
   prompt += "Example output json:\n"
   prompt += '{"output": "' + str(example_output) + '"}'
 
   if verbose: 
-    print ("CHAT GPT PROMPT")
+    print ("GPT4All PROMPT")
     print (prompt)
 
   for i in range(repeat): 
@@ -171,7 +160,7 @@ def ChatGPT_safe_generate_response_OLD(prompt,
                                    func_clean_up=None,
                                    verbose=False): 
   if verbose: 
-    print ("CHAT GPT PROMPT")
+    print ("GPT4All PROMPT")
     print (prompt)
 
   for i in range(repeat): 
@@ -191,34 +180,34 @@ def ChatGPT_safe_generate_response_OLD(prompt,
 
 
 # ============================================================================
-# ###################[SECTION 2: ORIGINAL GPT-3 STRUCTURE] ###################
+# ###################[SECTION 2: ORIGINAL GPT4All STRUCTURE] ###################
 # ============================================================================
 
 def GPT_request(prompt, gpt_parameter): 
   """
-  Given a prompt and a dictionary of GPT parameters, make a request to OpenAI
+  Given a prompt and a dictionary of GPT parameters, make a request to GPT4All
   server and returns the response. 
   ARGS:
     prompt: a str prompt
     gpt_parameter: a python dictionary with the keys indicating the names of  
                    the parameter and the values indicating the parameter 
                    values.   
   RETURNS: 
-    a str of GPT-3's response. 
+    a str of GPT4All's response. 
   """
   temp_sleep()
   try: 
-    response = openai.Completion.create(
-                model=gpt_parameter["engine"],
-                prompt=prompt,
-                temperature=gpt_parameter["temperature"],
-                max_tokens=gpt_parameter["max_tokens"],
-                top_p=gpt_parameter["top_p"],
-                frequency_penalty=gpt_parameter["frequency_penalty"],
-                presence_penalty=gpt_parameter["presence_penalty"],
-                stream=gpt_parameter["stream"],
-                stop=gpt_parameter["stop"],)
-    return response.choices[0].text
+
+
+    output = model.generate(
+      prompt,
+      max_tokens=gpt_parameter["max_tokens"],
+      temp=gpt_parameter["temperature"],
+      top_p=gpt_parameter["top_p"],
+
+
+    )
+    return output
   except: 
     print ("TOKEN LIMIT EXCEEDED")
     return "TOKEN LIMIT EXCEEDED"
@@ -236,7 +225,7 @@ def generate_prompt(curr_input, prompt_lib_file):
                 INPUT, THIS CAN BE A LIST.)
     prompt_lib_file: the path to the promopt file. 
   RETURNS: 
-    a str prompt that will be sent to OpenAI's GPT server.  
+    a str prompt that will be sent to GPT4All's model.  
   """
   if type(curr_input) == type("string"): 
     curr_input = [curr_input]
@@ -273,17 +262,19 @@ def safe_generate_response(prompt,
   return fail_safe_response
 
 
-def get_embedding(text, model="text-embedding-ada-002"):
-  text = text.replace("\n", " ")
-  if not text: 
-    text = "this is blank"
-  return openai.Embedding.create(
-          input=[text], model=model)['data'][0]['embedding']
+def get_embedding(text):
+    text = text.replace("\n", " ")
+    if not text: 
+        text = "this is blank"
+    embedder = Embed4All()
+    embedding = embedder.embed(text)
+    return embedding
+
 
 
 if __name__ == '__main__':
-  gpt_parameter = {"engine": "text-davinci-003", "max_tokens": 50, 
-                   "temperature": 0, "top_p": 1, "stream": False,
+  gpt_parameter = {"max_tokens": max_tokens, 
+                   "temperature": temperature, "top_p": 1, "stream": False,
                    "frequency_penalty": 0, "presence_penalty": 0, 
                    "stop": ['"']}
   curr_input = ["driving to a friend's house"]
@@ -309,23 +300,3 @@ def __func_clean_up(gpt_response):
                                  True)
 
   print (output)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-