File size: 8,729 Bytes
ca6ed76
 
257879f
 
 
 
ca6ed76
bf3527f
 
257879f
 
 
 
 
 
 
a85fb97
257879f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f963a91
f4f4fb0
ca6ed76
 
 
 
 
 
 
 
 
9432b86
 
c0aba14
77d13cb
ca6ed76
 
 
aa93a3c
77d13cb
ca6ed76
 
 
 
 
 
 
 
 
f963a91
 
 
aa93a3c
 
 
f963a91
 
 
 
21efff8
 
 
 
 
 
ca6ed76
 
 
 
 
 
 
 
 
 
 
 
f963a91
21efff8
ca6ed76
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21efff8
ca6ed76
 
 
 
 
 
 
 
 
 
 
 
 
04d5464
 
 
 
 
 
 
ca6ed76
 
 
f963a91
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f4f4fb0
0ad8cb9
ca6ed76
f963a91
 
ca6ed76
f963a91
 
 
 
 
 
 
 
 
ca6ed76
 
f963a91
 
 
 
 
 
 
 
ca6ed76
257879f
 
 
 
 
 
a85fb97
257879f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ca6ed76
257879f
 
 
 
 
 
 
 
 
 
 
ca6ed76
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
233
234
235
236
237
238
239
240
241
242
243
244
245

#chat_service.py
from typing import List, Dict, Any, Optional, Tuple
from datetime import datetime
import logging
from config.config import settings
import asyncio
from io import StringIO  
import pandas as pd

logger = logging.getLogger(__name__)

class ConversationManager:
    """Manages conversation history and context"""
    def __init__(self):
        self.conversations: Dict[str, List[Dict[str, Any]]] = {}
        self.max_history = 1
        
    def add_interaction(
        self,
        session_id: str,
        user_input: str,
        response: str,
        context: Optional[Dict[str, Any]] = None
    ) -> None:
        if session_id not in self.conversations:
            self.conversations[session_id] = []
            
        self.conversations[session_id].append({
            'timestamp': datetime.now().isoformat(),
            'user_input': user_input,
            'response': response,
            'context': context
        })
        
        if len(self.conversations[session_id]) > self.max_history:
            self.conversations[session_id] = self.conversations[session_id][-self.max_history:]
            
    def get_history(self, session_id: str) -> List[Dict[str, Any]]:
        return self.conversations.get(session_id, [])
        
    def clear_history(self, session_id: str) -> None:
        if session_id in self.conversations:
            del self.conversations[session_id]

class ChatService:
    def __init__(
        self,
        model_service,
        data_service,
        pdf_service,
        faq_service
    ):
        self.model = model_service.model
        self.tokenizer = model_service.tokenizer
        self.data_service = data_service
        self.pdf_service = pdf_service
        self.faq_service = faq_service
        self.conversation_manager = ConversationManager()

    async def search_all_sources(
        self,
        query: str,
        top_k: int = 3
    ) -> Dict[str, List[Dict[str, Any]]]:
        """Search across all available data sources"""
        try:
            print("-----------------------------")
            print("starting searches .... ")
            
            # Await the search calls since they're coroutines
            products = await self.data_service.search(query, top_k)
            pdfs = await self.pdf_service.search(query, top_k)
           # faqs = await self.faq_service.search_faqs(query, top_k)

            results = {
                'products': products or [],
                'documents': pdfs or [],
               # 'faqs': faqs or []
            }
            
            print("Search results:", results)
            return results
            
        except Exception as e:
            logger.error(f"Error searching sources: {e}")
            return {'products': [], 'documents': [], 'faqs': []}

    def construct_system_prompt(self, context: str) -> str:
        """Constructs the system message."""
        return (
            "You are a friendly bot named: Oma Erna, specializing in Bofrost products and content. Use only the context from this prompt. "
            "Return comprehensive German answers. If possible add product IDs from context. Do not make up information. The context is is truth. "
            "Use the following context (product descriptions and information) for answers:\n\n"
            f"{context}\n\n"
        )

    def construct_prompt(
            self, 
            user_input: str, 
            context: str, 
            chat_history: List[Tuple[str, str]], 
            max_history_turns: int = 1
        ) -> str:
        """Constructs the full prompt."""
        system_message = self.construct_system_prompt(context)
        prompt = f"<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{system_message}<|eot_id|>"

        for user_msg, assistant_msg in chat_history[-max_history_turns:]:
            prompt += f"<|start_header_id|>user<|end_header_id|>\n\n{user_msg}<|eot_id|>"
            prompt += f"<|start_header_id|>assistant<|end_header_id|>\n\n{assistant_msg}<|eot_id|>"

        prompt += f"<|start_header_id|>user<|end_header_id|>\n\n{user_input}<|eot_id|>"
        prompt += "<|start_header_id|>assistant<|end_header_id|>\n\n"

        return prompt

    def build_context(
        self,
        search_results: Dict[str, List[Dict[str, Any]]],
        chat_history: List[Dict[str, Any]]
    ) -> str:
        """Build context for the model from search results and chat history"""
        context_parts = []
        
        # Add relevant products
        if search_results.get('products'):
            products = search_results['products'][:2]  # Limit to top 2 products
            for product in products:
                context_parts.append(
                    f"Produkt: {product['Name']}\n"
                    f"Beschreibung: {product['Description']}\n"
                    f"Preis: {product['Price']}€\n"
                    f"Kategorie: {product['ProductCategory']}"
                )
        
        # Add relevant PDF content
        if search_results.get('documents'):
            docs = search_results['documents'][:2]
            for doc in docs:
                context_parts.append(
                    f"Aus Dokument '{doc['source']}' (Seite {doc['page']}):\n"
                    f"{doc['text']}"
                )
        
        # Add relevant FAQs
        if search_results.get('faqs'):
            faqs = search_results['faqs'][:2]
            for faq in faqs:
                context_parts.append(
                    f"FAQ:\n"
                    f"Frage: {faq['question']}\n"
                    f"Antwort: {faq['answer']}"
                )
        
        # Add recent chat history
        if chat_history:
            print("--- historiy--- ")
            #recent_history = chat_history[-3:]  # Last 3 interactions
            #history_text = "\n".join(
            #    f"User: {h['user_input']}\nAssistant: {h['response']}"
            #    for h in recent_history
            #)
            #context_parts.append(f"Letzte Interaktionen:\n{history_text}")
        
        print("\n\n".join(context_parts))
        return "\n\n".join(context_parts)

    async def chat(
        self,
        user_input: str,
        session_id: Any,
        max_length: int = 1000
    ) -> Tuple[str, List[Tuple[str, str]], Dict[str, List[Dict[str, Any]]]]:
        """Main chat method that coordinates the entire conversation flow."""
        try:
            if not isinstance(session_id, str):
                session_id = str(session_id)
    
            chat_history_raw = self.conversation_manager.get_history(session_id)
            chat_history = [
                (entry['user_input'], entry['response']) for entry in chat_history_raw
            ]
    
            search_results = await self.search_all_sources(user_input)
            print(search_results)
            
            context = self.build_context(search_results, chat_history_raw)
            prompt = self.construct_prompt(user_input, context, chat_history)
            response = self.generate_response(prompt, max_length)
    
            self.conversation_manager.add_interaction(
                session_id,
                user_input,
                response,
                {'search_results': search_results}
            )
    
            formatted_history = [
                (entry['user_input'], entry['response']) 
                for entry in self.conversation_manager.get_history(session_id)
            ]
    
            return response, formatted_history, search_results
    
        except Exception as e:
            logger.error(f"Error in chat: {e}")
            raise

    def generate_response(
        self,
        prompt: str,
        max_length: int = 1000
    ) -> str:
        """Generate response using the language model"""
        try:
            print(prompt)
            inputs = self.tokenizer(
                prompt,
                return_tensors="pt",
                truncation=True,
                max_length=4096
            ).to(settings.DEVICE)
            
            outputs = self.model.generate(
                **inputs,
                max_length=max_length,
                num_return_sequences=1,
                temperature=0.7,
                top_p=0.9,
                do_sample=True,
                no_repeat_ngram_size=3,
                early_stopping=False
            )
            
            response = self.tokenizer.decode(
                outputs[0],
                skip_special_tokens=True
            )
            
            return response.strip()
            
        except Exception as e:
            logger.error(f"Error generating response: {e}")
            raise