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
246
247
248
249
250
251
252
253
254
255
256 | class ResponseService(Service):
"""
Orchestrates AI response generation by integrating retrieval results, historical context, and prompt engineering.
Coordinates plugin-managed large language models (LLMs), websockets for streaming responses,
and tracks metrics throughout the response generation pipeline.
Attributes:
plugin_manager (PluginManager): Provides access to LLM and DB plugins.
web_socket_manager (WebsocketManager): Manages live streaming over websocket.
db_document_plugin (dict): Document store plugin interface.
engram_repository (EngramRepository): Access point for loading engrams.
llm_main (dict): Plugin for executing the main LLM-based response generation.
metrics_tracker (MetricsTracker): Tracks internal response metrics.
Methods:
start() -> None:
Subscribes to service topics and initializes websocket manager.
stop() -> None:
Shuts down the websocket manager and stops the service.
init_async() -> None:
Initializes the DB plugin connection asynchronously.
on_retrieve_complete(retrieve_result_in: dict[str, Any]) -> None:
Processes retrieval results and initiates engram and history fetch.
_fetch_history(prompt: Prompt) -> dict[str, Any]:
Asynchronously fetches historical conversation context.
_fetch_retrieval(prompt: Prompt, source_id: str, analysis: PromptAnalysis, retrieve_result: RetrieveResult) -> dict[str, Any]:
Loads engrams using retrieve result.
on_fetch_data_complete(fut: Future[Any]) -> None:
Launches main prompt generation after history and engrams are loaded.
main_prompt(prompt_in: Prompt, source_id: str, analysis: PromptAnalysis, engram_array: list[Engram], retrieve_result: RetrieveResult, history_array: dict[str, Any]) -> Response:
Constructs and submits the main prompt to the LLM plugin.
on_main_prompt_complete(fut: Future[Any]) -> None:
Sends generated response and updates metrics.
on_acknowledge(message_in: str) -> None:
Sends current metrics snapshot to monitoring topics.
"""
def __init__(self, host: Host) -> None:
super().__init__(host)
self.plugin_manager: PluginManager = host.plugin_manager
self.web_socket_manager: WebsocketManager = WebsocketManager(host)
self.db_document_plugin = self.plugin_manager.get_plugin('db', 'document')
self.engram_repository: EngramRepository = EngramRepository(self.db_document_plugin)
self.llm_main = self.plugin_manager.get_plugin('llm', 'response_main')
self.metrics_tracker: MetricsTracker[ResponseMetric] = MetricsTracker[ResponseMetric]()
##
# Many methods are not ready to be until their async component is running.
# Do not call async context methods in the constructor.
def start(self) -> None:
self.subscribe(Service.Topic.ACKNOWLEDGE, self.on_acknowledge)
self.subscribe(Service.Topic.RETRIEVE_COMPLETE, self.on_retrieve_complete)
self.web_socket_manager.init_async()
super().start()
async def stop(self) -> None:
await self.web_socket_manager.shutdown()
def init_async(self) -> None:
self.db_document_plugin['func'].connect(args=None)
return super().init_async()
def on_retrieve_complete(self, retrieve_result_in: dict[str, Any]) -> None:
if __debug__:
self.host.update_mock_data_input(self, retrieve_result_in)
prompt = Prompt(**retrieve_result_in['prompt'])
prompt_analysis = PromptAnalysis(**retrieve_result_in['analysis'])
retrieve_result = RetrieveResult(**retrieve_result_in['retrieve_response'])
source_id = retrieve_result.source_id
self.metrics_tracker.increment(ResponseMetric.RETRIEVES_RECIEVED)
fetch_engrams_task = self.run_tasks([
self._fetch_retrieval(
prompt=prompt, source_id=source_id, analysis=prompt_analysis, retrieve_result=retrieve_result
),
self._fetch_history(prompt),
])
fetch_engrams_task.add_done_callback(self.on_fetch_data_complete)
"""
### Fetch History & Engram
Fetch engrams based on the IDs provided by the retrieve service.
"""
async def _fetch_history(self, prompt: Prompt) -> dict[str, Any]:
plugin = self.db_document_plugin
args = plugin['args']
args['history'] = 1
args['repo_ids_filters'] = prompt.repo_ids_filters
ret_val = await asyncio.to_thread(plugin['func'].fetch, table=DB.DBTables.HISTORY, ids=[], args=args)
history: dict[str, Any] = ret_val[0]
return history
async def _fetch_retrieval(
self, prompt: Prompt, source_id: str, analysis: PromptAnalysis, retrieve_result: RetrieveResult
) -> dict[str, Any]:
engram_array: list[Engram] = await asyncio.to_thread(
self.engram_repository.load_batch_retrieve_result, retrieve_result
)
# assembled main_prompt, render engrams.
return {
'prompt': prompt,
'source_id': source_id,
'analysis': analysis,
'retrieve_result': retrieve_result,
'engram_array': engram_array,
}
def on_fetch_data_complete(self, fut: Future[Any]) -> None:
exc = fut.exception()
if exc is not None:
raise exc
result = fut.result()
retrieval = result['_fetch_retrieval'][0]
history = result['_fetch_history'][0]
main_prompt_task = self.run_task(
self.main_prompt(
retrieval['prompt'],
retrieval['source_id'],
retrieval['analysis'],
retrieval['engram_array'],
retrieval['retrieve_result'],
history,
)
)
main_prompt_task.add_done_callback(self.on_main_prompt_complete)
"""
### Main Prompt
Combine the previous stages to generate the response.
"""
async def main_prompt(
self,
prompt_in: Prompt,
source_id: str,
analysis: PromptAnalysis,
engram_array: list[Engram],
retrieve_result: RetrieveResult,
history_array: dict[str, Any],
) -> Response:
self.metrics_tracker.increment(ResponseMetric.ENGRAMS_FETCHED, len(engram_array))
engram_dict_list = [asdict(engram) for engram in engram_array]
# build main prompt here
prompt = PromptMainPrompt(
prompt_str=prompt_in.prompt_str,
is_lesson=prompt_in.is_lesson,
training_mode=prompt_in.training_mode,
repo_ids_filters=prompt_in.repo_ids_filters,
input_data={
'engram_list': engram_dict_list,
'history': history_array['history'],
'working_memory': retrieve_result.conversation_direction,
'analysis': retrieve_result.analysis,
},
)
plugin = self.llm_main
args = self.host.mock_update_args(plugin)
if prompt_in.is_lesson:
response = await asyncio.to_thread(
plugin['func'].submit, prompt=prompt, args=args, images=None, structured_schema=None
)
else:
response = await asyncio.to_thread(
plugin['func'].submit_streaming,
prompt=prompt,
websocket_manager=self.web_socket_manager,
args=args,
)
if __debug__:
main_prompt = prompt.render_prompt()
self.send_message_async(
Service.Topic.DEBUG_MAIN_PROMPT_INPUT, {'main_prompt': main_prompt, 'ask_id': retrieve_result.ask_id}
)
self.host.update_mock_data(self.llm_main, response)
model = ''
if plugin['args'].get('model'):
model = plugin['args']['model']
response = response[0]['llm_response'].replace('$', 'USD ').replace('<context>', '').replace('</context>', '')
response_inst = Response(str(uuid.uuid4()), source_id, response, retrieve_result, prompt_in, analysis, model)
return response_inst
def on_main_prompt_complete(self, fut: Future[Any]) -> None:
result = fut.result()
self.metrics_tracker.increment(ResponseMetric.MAIN_PROMPTS_RUN)
self.send_message_async(Service.Topic.MAIN_PROMPT_COMPLETE, asdict(result))
if __debug__:
self.host.update_mock_data_output(self, asdict(result))
"""
### Ack
Acknowledge and return metrics
"""
def on_acknowledge(self, message_in: str) -> None:
del message_in
metrics_packet: MetricPacket = self.metrics_tracker.get_and_reset_packet()
self.send_message_async(
Service.Topic.STATUS,
{'id': self.id, 'name': self.__class__.__name__, 'timestamp': time.time(), 'metrics': metrics_packet},
)
|