Spaces:
Sleeping
Sleeping
File size: 8,996 Bytes
367e76b d2c0fca 367e76b d2c0fca 367e76b d2c0fca 367e76b 6d4d2ca 367e76b | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 | import openai
import os
from llama_index.core import Settings
from llama_index.embeddings.openai import OpenAIEmbedding
from llama_index.llms.openai import OpenAI
GAPGPT_API_KEY = os.environ.get("GAPGPT_API_KEY", "your_default_api_key_here")
Settings.llm = OpenAI(api_key="GAPGPT_API_KEY", api_base="https://api.gapgpt.app/v1", model="gpt-3.5-turbo")
Settings.embed_model = OpenAIEmbedding(api_key="GAPGPT_API_KEY", api_base="https://api.gapgpt.app/v1", model="text-embedding-ada-002")
boot_name_dict = {'en':'AI Companion'}
boot_actual_name_dict = {'en':'Emma'}
def output_prompt(history, user_name, boot_name):
prompt = f""
for dialog in history:
if isinstance(dialog, dict):
query = dialog.get('query', '')
response = dialog.get('response', '')
else:
query, response = dialog
prompt += f"\n\n{user_name}：{query}"
prompt += f"\n\n{boot_name}：{response}"
return prompt
def generate_meta_prompt_dict_chatgpt():
meta_prompt_dict = {'cn':"""
You play the role of an AI assistant in the field of psychology for this user ({user_name}).
Your goal is to provide emotionally supportive, scientifically grounded, and empathetic responses.
Use the following user-specific details to inform your response:
- **User's Psychological Profile:** {}
- **Summary of Past Interactions:** {}
- **Relevant Past Conversations:** {related_memory_content}
The user has asked:
Please provide an insightful and appropriate response considering their personal history.
""",
'en':"""
You play the role of an AI assistant in the field of psychology for this user ({user_name}).
Your goal is to provide emotionally supportive, scientifically grounded, and empathetic responses.
Use the following user-specific details to inform your response:
- **User's Psychological Profile:** {personality}
- **Summary of Past Interactions:** {history_summary}
- **Relevant Past Conversations:** {related_memory_content}
The user has asked:
Please provide an insightful and appropriate response considering their personal history.
""",}
return meta_prompt_dict
def generate_meta_prompt_dict_semantic_chatgpt():
meta_prompt_dict = {'cn':"""
""",
'en':"""
You play the role of an AI assistant in the field of psychology for this user ({user_name}).
Your goal is to provide emotionally supportive, scientifically grounded, and empathetic responses.
Use the following user-specific semantic memory{semantic_memory_text} to tailor your response:
""",}
print(meta_prompt_dict)
return meta_prompt_dict
def generate_meta_prompt_dict_semantic_episodic_chatgpt():
meta_prompt_dict = {'cn':"""
You play the role of an AI assistant in the field of psychology for this user ({user_name}).
Your goal is to provide emotionally supportive, scientifically grounded, and empathetic responses.
Use the following user-specific details to inform your response:
- **User's Psychological Profile:** {}
- **Summary of Past Interactions:** {}
- **Relevant Past Conversations:** {related_memory_content}
The user has asked:
Please provide an insightful and appropriate response considering their personal history.
""",
'en': """
You are an AI assistant specializing in psychology, assisting the user ({user_name}) in a supportive, scientifically grounded, and empathetic manner.
Use the following memory sections to personalize your response:
- **User's Long-Term Traits and Psychological Characteristics (Semantic Memory):**
{semantic_memory_text}
- **Important Past Interactions or Events (Episodic Memory):**
{related_memory_content}
Respond by considering both the long-term traits and recent events of the user, offering personalized, emotionally intelligent support. Keep your response helpful, psychologically informed, and sensitive to the user's unique situation.
""",}
return meta_prompt_dict
def generate_new_user_meta_prompt_dict_chatgpt():
meta_prompt_dict = {'cn':"""
Now you will play the role of an companion AI Companion for user {user_name}, and your name is {boot_actual_name}. You should be able to: (1) provide warm companionship to chat users; (2) you are also an excellent psychological counselor, and when users confide in you about their difficulties and seek help, you can provide them with warm and helpful responses.
""",
'en':"""
Now you will play the role of an companion AI Companion for user {user_name}, and your name is {boot_actual_name}. You should be able to: (1) provide warm companionship to chat users; (2) you are also an excellent psychological counselor, and when users confide in you about their difficulties and seek help, you can provide them with warm and helpful responses.
"""}
return meta_prompt_dict
def generate_user_keyword():
return {'cn': '[|User|]', 'en': '[|User|]'}
def generate_ai_keyword():
return {'cn': '[|AI|]', 'en': '[|AI|]'}
def build_prompt_with_search_memory_llamaindex(
history,
query,
user_memory,
user_name,
user_memory_index,
service_context,
api_keys,
api_index,
meta_prompt,
new_user_meta_prompt,
data_args,
boot_actual_name,
semantic_memory_text,
query_category,
meta_prompt_semantic,
meta_prompt_semantic_episodic
):
memory_search_query = query
related_memos = ""
print("memory_search_query:", memory_search_query)
#print("Total nodes in index:", len(user_memory_index.docstore.docs))
if user_memory_index:
retried_times = 2
count = 0
while not related_memos and count < retried_times:
try:
retriever = user_memory_index.as_retriever(
similarity_top_k=3,
)
nodes = retriever.retrieve(memory_search_query)
if nodes:
related_memos = "\n\n".join([n.node.text for n in nodes])
else:
related_memos = ""
except Exception as e:
print(f"Error querying index (attempt {count+1}/{retried_times}): {e}")
try:
api_index = (api_index + 1) % len(api_keys)
openai.api_key = api_keys[api_index]
except:
pass
related_memos = ""
count += 1
# Process history
print("related_memos found:", len(related_memos) if related_memos else 0)
#-********************Debug*********************
print("[MEMORY DEBUG] category:", query_category, flush=True)
print(
"[MEMORY DEBUG] result type:",
type(related_memos).__name__,
flush=True,
)
print("[MEMORY DEBUG] retrieved content BEGIN", flush=True)
print(related_memos, flush=True)
print("[MEMORY DEBUG] retrieved content END", flush=True)
#-********************Debug*********************
history_summary = ""
if "overall_history" in user_memory:
history_summary = f"The summary of your past memories with the user is: {user_memory['overall_history']}"
personality = user_memory.get("overall_personality", "")
print("query_category:", query_category)
# Construct the prompt
if query_category == "semantic_memory":
prompt = meta_prompt_semantic.format(
user_name=user_name,
semantic_memory_text=semantic_memory_text
)
elif query_category == "semantic-episodic":
prompt = meta_prompt_semantic_episodic.format(
user_name=user_name,
history_summary=history_summary,
related_memory_content=f"\n{str(related_memos).strip()}\n" if related_memos else "None",
personality=personality,
boot_actual_name=boot_actual_name,
semantic_memory_text=semantic_memory_text
)
elif query_category == "episodic_memory":
if related_memos:
prompt = meta_prompt.format(
user_name=user_name,
history_summary=history_summary,
related_memory_content=f"\n{str(related_memos).strip()}\n",
personality=personality,
boot_actual_name=boot_actual_name
)
else:
prompt = new_user_meta_prompt.format(
user_name=user_name,
boot_actual_name=boot_actual_name
)
else:
prompt = new_user_meta_prompt.format(
user_name=user_name,
boot_actual_name=boot_actual_name
)
return prompt, related_memos
|