Spaces:
Sleeping
Sleeping
Upload 2 files
Browse files- shared/chat_history.py +80 -0
- shared/models.py +34 -0
shared/chat_history.py
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import os
|
| 2 |
+
import json
|
| 3 |
+
import time
|
| 4 |
+
from datetime import datetime
|
| 5 |
+
from typing import List, Dict, Any
|
| 6 |
+
from .models import ChatMessage
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
def save_chat_history(messages: List[ChatMessage], model_name: str, response: str, filename: str = "chat.md"):
|
| 10 |
+
"""
|
| 11 |
+
Save chat history to a markdown file with timestamp and model information
|
| 12 |
+
"""
|
| 13 |
+
timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
|
| 14 |
+
|
| 15 |
+
# Prepare the markdown content
|
| 16 |
+
history_content = f"""
|
| 17 |
+
## Chat Session: {timestamp}
|
| 18 |
+
**Model Used:** {model_name}
|
| 19 |
+
|
| 20 |
+
---
|
| 21 |
+
"""
|
| 22 |
+
|
| 23 |
+
# Add all messages to the markdown file
|
| 24 |
+
for msg in messages:
|
| 25 |
+
role_prefix = "**User:**" if msg.role.lower() == "user" else "**Assistant:**"
|
| 26 |
+
history_content += f"\n{role_prefix} {msg.content}\n\n"
|
| 27 |
+
|
| 28 |
+
# Add the final response from the assistant
|
| 29 |
+
history_content += f"\n**Assistant Response:** {response}\n\n---\n\n"
|
| 30 |
+
|
| 31 |
+
# Append to the chat history file
|
| 32 |
+
with open(filename, "a", encoding="utf-8") as file:
|
| 33 |
+
file.write(history_content)
|
| 34 |
+
|
| 35 |
+
|
| 36 |
+
def save_detailed_chat_log(request_data: Dict[str, Any], response_data: str, model_name: str, processing_time: float, filename: str = "chat.md"):
|
| 37 |
+
"""
|
| 38 |
+
Save detailed chat log with metadata
|
| 39 |
+
"""
|
| 40 |
+
timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
|
| 41 |
+
|
| 42 |
+
log_content = f"""
|
| 43 |
+
## Chat Request Log: {timestamp}
|
| 44 |
+
- **Model:** {model_name}
|
| 45 |
+
- **Processing Time:** {processing_time:.2f}s
|
| 46 |
+
- **Max Tokens:** {request_data.get('max_tokens', 512)}
|
| 47 |
+
- **Temperature:** {request_data.get('temperature', 0.8)}
|
| 48 |
+
|
| 49 |
+
### Input Messages:
|
| 50 |
+
"""
|
| 51 |
+
|
| 52 |
+
# Add the messages from the request
|
| 53 |
+
messages = request_data.get('messages', [])
|
| 54 |
+
for msg in messages:
|
| 55 |
+
role = msg.get('role', 'unknown')
|
| 56 |
+
content = msg.get('content', '')
|
| 57 |
+
role_display = "**User**" if role.lower() == 'user' else "**Assistant**"
|
| 58 |
+
log_content += f"- {role_display}: {content}\n"
|
| 59 |
+
|
| 60 |
+
log_content += f"\n### Model Response:\n{response_data}\n\n---\n\n"
|
| 61 |
+
|
| 62 |
+
# Append to the file
|
| 63 |
+
with open(filename, "a", encoding="utf-8") as file:
|
| 64 |
+
file.write(log_content)
|
| 65 |
+
|
| 66 |
+
|
| 67 |
+
def initialize_chat_file(filename: str = "chat.md"):
|
| 68 |
+
"""
|
| 69 |
+
Initialize the chat history file with header if it doesn't exist
|
| 70 |
+
"""
|
| 71 |
+
if not os.path.exists(filename):
|
| 72 |
+
header = f"""# Chat History
|
| 73 |
+
Last updated: {datetime.now().strftime("%Y-%m-%d %H:%M:%S")}
|
| 74 |
+
|
| 75 |
+
This file contains the history of all chat conversations processed by the multi-node API system.
|
| 76 |
+
|
| 77 |
+
---
|
| 78 |
+
"""
|
| 79 |
+
with open(filename, "w", encoding="utf-8") as file:
|
| 80 |
+
file.write(header)
|
shared/models.py
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from pydantic import BaseModel
|
| 2 |
+
from typing import List, Optional, Dict, Any
|
| 3 |
+
|
| 4 |
+
|
| 5 |
+
class ChatMessage(BaseModel):
|
| 6 |
+
role: str # "user" or "assistant"
|
| 7 |
+
content: str
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
class ChatRequest(BaseModel):
|
| 11 |
+
messages: List[ChatMessage]
|
| 12 |
+
model: str = "sam-x-nano"
|
| 13 |
+
max_tokens: Optional[int] = 512
|
| 14 |
+
temperature: Optional[float] = 0.8
|
| 15 |
+
top_k: Optional[int] = 40
|
| 16 |
+
top_p: Optional[float] = 0.9
|
| 17 |
+
repetition_penalty: Optional[float] = 1.1
|
| 18 |
+
stream: Optional[bool] = False
|
| 19 |
+
|
| 20 |
+
|
| 21 |
+
class ChatResponse(BaseModel):
|
| 22 |
+
id: str
|
| 23 |
+
object: str = "chat.completion"
|
| 24 |
+
created: int
|
| 25 |
+
model: str
|
| 26 |
+
choices: List[Dict[str, Any]]
|
| 27 |
+
usage: Dict[str, int]
|
| 28 |
+
|
| 29 |
+
|
| 30 |
+
class WorkerStatus(BaseModel):
|
| 31 |
+
model_name: str
|
| 32 |
+
is_active: bool
|
| 33 |
+
load: float
|
| 34 |
+
last_heartbeat: int
|