内置可观察性工具¶

在 LlamaIndex 内部，许多事件 (event) 和 span 都通过我们的工具集成系统创建和记录。

本 Notebook 将引导您了解如何 Hook 到这些事件和 span 中，以创建您自己的可观察性工具。

In [ ]

%pip install llama-index treelib

%pip install llama-index treelib

事件¶

LlamaIndex 记录了几种类型的事件。事件是运行时发生的单一数据点，通常属于某个父 span。

下面详细列出了记录的内容，以及如何创建事件处理程序来读取这些事件。

In [ ]

from typing import Dict, List
from treelib import Tree

from llama_index.core.instrumentation.events import BaseEvent
from llama_index.core.instrumentation.event_handlers import BaseEventHandler

from llama_index.core.instrumentation.events.agent import (
    AgentChatWithStepStartEvent,
    AgentChatWithStepEndEvent,
    AgentRunStepStartEvent,
    AgentRunStepEndEvent,
    AgentToolCallEvent,
)
from llama_index.core.instrumentation.events.chat_engine import (
    StreamChatErrorEvent,
    StreamChatDeltaReceivedEvent,
)
from llama_index.core.instrumentation.events.embedding import (
    EmbeddingStartEvent,
    EmbeddingEndEvent,
)
from llama_index.core.instrumentation.events.llm import (
    LLMPredictEndEvent,
    LLMPredictStartEvent,
    LLMStructuredPredictEndEvent,
    LLMStructuredPredictStartEvent,
    LLMCompletionEndEvent,
    LLMCompletionStartEvent,
    LLMChatEndEvent,
    LLMChatStartEvent,
    LLMChatInProgressEvent,
)
from llama_index.core.instrumentation.events.query import (
    QueryStartEvent,
    QueryEndEvent,
)
from llama_index.core.instrumentation.events.rerank import (
    ReRankStartEvent,
    ReRankEndEvent,
)
from llama_index.core.instrumentation.events.retrieval import (
    RetrievalStartEvent,
    RetrievalEndEvent,
)
from llama_index.core.instrumentation.events.span import (
    SpanDropEvent,
)
from llama_index.core.instrumentation.events.synthesis import (
    SynthesizeStartEvent,
    SynthesizeEndEvent,
    GetResponseEndEvent,
    GetResponseStartEvent,
)


class ExampleEventHandler(BaseEventHandler):
    """Example event handler.

    This event handler is an example of how to create a custom event handler.

    In general, logged events are treated as single events in a point in time,
    that link to a span. The span is a collection of events that are related to
    a single task. The span is identified by a unique span_id.

    While events are independent, there is some hierarchy.
    For example, in query_engine.query() call with a reranker attached:
    - QueryStartEvent
    - RetrievalStartEvent
    - EmbeddingStartEvent
    - EmbeddingEndEvent
    - RetrievalEndEvent
    - RerankStartEvent
    - RerankEndEvent
    - SynthesizeStartEvent
    - GetResponseStartEvent
    - LLMPredictStartEvent
    - LLMChatStartEvent
    - LLMChatEndEvent
    - LLMPredictEndEvent
    - GetResponseEndEvent
    - SynthesizeEndEvent
    - QueryEndEvent
    """

    events: List[BaseEvent] = []

    @classmethod
    def class_name(cls) -> str:
        """Class name."""
        return "ExampleEventHandler"

    def handle(self, event: BaseEvent) -> None:
        """Logic for handling event."""
        print("-----------------------")
        # all events have these attributes
        print(event.id_)
        print(event.timestamp)
        print(event.span_id)

        # event specific attributes
        print(f"Event type: {event.class_name()}")
        if isinstance(event, AgentRunStepStartEvent):
            print(event.task_id)
            print(event.step)
            print(event.input)
        if isinstance(event, AgentRunStepEndEvent):
            print(event.step_output)
        if isinstance(event, AgentChatWithStepStartEvent):
            print(event.user_msg)
        if isinstance(event, AgentChatWithStepEndEvent):
            print(event.response)
        if isinstance(event, AgentToolCallEvent):
            print(event.arguments)
            print(event.tool.name)
            print(event.tool.description)
            print(event.tool.to_openai_tool())
        if isinstance(event, StreamChatDeltaReceivedEvent):
            print(event.delta)
        if isinstance(event, StreamChatErrorEvent):
            print(event.exception)
        if isinstance(event, EmbeddingStartEvent):
            print(event.model_dict)
        if isinstance(event, EmbeddingEndEvent):
            print(event.chunks)
            print(event.embeddings[0][:5])  # avoid printing all embeddings
        if isinstance(event, LLMPredictStartEvent):
            print(event.template)
            print(event.template_args)
        if isinstance(event, LLMPredictEndEvent):
            print(event.output)
        if isinstance(event, LLMStructuredPredictStartEvent):
            print(event.template)
            print(event.template_args)
            print(event.output_cls)
        if isinstance(event, LLMStructuredPredictEndEvent):
            print(event.output)
        if isinstance(event, LLMCompletionStartEvent):
            print(event.model_dict)
            print(event.prompt)
            print(event.additional_kwargs)
        if isinstance(event, LLMCompletionEndEvent):
            print(event.response)
            print(event.prompt)
        if isinstance(event, LLMChatInProgressEvent):
            print(event.messages)
            print(event.response)
        if isinstance(event, LLMChatStartEvent):
            print(event.messages)
            print(event.additional_kwargs)
            print(event.model_dict)
        if isinstance(event, LLMChatEndEvent):
            print(event.messages)
            print(event.response)
        if isinstance(event, RetrievalStartEvent):
            print(event.str_or_query_bundle)
        if isinstance(event, RetrievalEndEvent):
            print(event.str_or_query_bundle)
            print(event.nodes)
        if isinstance(event, ReRankStartEvent):
            print(event.query)
            print(event.nodes)
            print(event.top_n)
            print(event.model_name)
        if isinstance(event, ReRankEndEvent):
            print(event.nodes)
        if isinstance(event, QueryStartEvent):
            print(event.query)
        if isinstance(event, QueryEndEvent):
            print(event.response)
            print(event.query)
        if isinstance(event, SpanDropEvent):
            print(event.err_str)
        if isinstance(event, SynthesizeStartEvent):
            print(event.query)
        if isinstance(event, SynthesizeEndEvent):
            print(event.response)
            print(event.query)
        if isinstance(event, GetResponseStartEvent):
            print(event.query_str)

        self.events.append(event)
        print("-----------------------")

    def _get_events_by_span(self) -> Dict[str, List[BaseEvent]]:
        events_by_span: Dict[str, List[BaseEvent]] = {}
        for event in self.events:
            if event.span_id in events_by_span:
                events_by_span[event.span_id].append(event)
            else:
                events_by_span[event.span_id] = [event]
        return events_by_span

    def _get_event_span_trees(self) -> List[Tree]:
        events_by_span = self._get_events_by_span()

        trees = []
        tree = Tree()

        for span, sorted_events in events_by_span.items():
            # create root node i.e. span node
            tree.create_node(
                tag=f"{span} (SPAN)",
                identifier=span,
                parent=None,
                data=sorted_events[0].timestamp,
            )

            for event in sorted_events:
                tree.create_node(
                    tag=f"{event.class_name()}: {event.id_}",
                    identifier=event.id_,
                    parent=event.span_id,
                    data=event.timestamp,
                )

            trees.append(tree)
            tree = Tree()
        return trees

    def print_event_span_trees(self) -> None:
        """Method for viewing trace trees."""
        trees = self._get_event_span_trees()
        for tree in trees:
            print(
                tree.show(
                    stdout=False, sorting=True, key=lambda node: node.data
                )
            )
            print("")

from typing import Dict, List from treelib import Tree from llama_index.core.instrumentation.events import BaseEvent from llama_index.core.instrumentation.event_handlers import BaseEventHandler from llama_index.core.instrumentation.events.agent import ( AgentChatWithStepStartEvent, AgentChatWithStepEndEvent, AgentRunStepStartEvent, AgentRunStepEndEvent, AgentToolCallEvent, ) from llama_index.core.instrumentation.events.chat_engine import ( StreamChatErrorEvent, StreamChatDeltaReceivedEvent, ) from llama_index.core.instrumentation.events.embedding import ( EmbeddingStartEvent, EmbeddingEndEvent, ) from llama_index.core.instrumentation.events.llm import ( LLMPredictEndEvent, LLMPredictStartEvent, LLMStructuredPredictEndEvent, LLMStructuredPredictStartEvent, LLMCompletionEndEvent, LLMCompletionStartEvent, LLMChatEndEvent, LLMChatStartEvent, LLMChatInProgressEvent, ) from llama_index.core.instrumentation.events.query import ( QueryStartEvent, QueryEndEvent, ) from llama_index.core.instrumentation.events.rerank import ( ReRankStartEvent, ReRankEndEvent, ) from llama_index.core.instrumentation.events.retrieval import ( RetrievalStartEvent, RetrievalEndEvent, ) from llama_index.core.instrumentation.events.span import ( SpanDropEvent, ) from llama_index.core.instrumentation.events.synthesis import ( SynthesizeStartEvent, SynthesizeEndEvent, GetResponseEndEvent, GetResponseStartEvent, ) class ExampleEventHandler(BaseEventHandler): """示例事件处理器。此事件处理器是创建自定义事件处理器的示例。一般来说，记录的事件被视为时间中的单个点事件，这些事件链接到一个跨度（span）。跨度是与单个任务相关的事件集合。跨度由唯一的 span_id 标识。虽然事件是独立的，但它们之间存在一定的层级关系。例如，在带有 reranker 的 query_engine.query() 调用中： - QueryStartEvent - RetrievalStartEvent - EmbeddingStartEvent - EmbeddingEndEvent - RetrievalEndEvent - RerankStartEvent - ReRankEndEvent - SynthesizeStartEvent - GetResponseStartEvent - LLMPredictStartEvent - LLMChatStartEvent - LLMChatEndEvent - LLMPredictEndEvent - GetResponseEndEvent - SynthesizeEndEvent - QueryEndEvent """ events: List[BaseEvent] = [] @classmethod def class_name(cls) -> str: """类名。""" return "ExampleEventHandler" def handle(self, event: BaseEvent) -> None: """处理事件的逻辑。""" print("-----------------------") # 所有事件都具有这些属性 print(event.id_) print(event.timestamp) print(event.span_id) # 事件特有的属性 print(f"Event type: {event.class_name()}") if isinstance(event, AgentRunStepStartEvent): print(event.task_id) print(event.step) print(event.input) if isinstance(event, AgentRunStepEndEvent): print(event.step_output) if isinstance(event, AgentChatWithStepStartEvent): print(event.user_msg) if isinstance(event, AgentChatWithStepEndEvent): print(event.response) if isinstance(event, AgentToolCallEvent): print(event.arguments) print(event.tool.name) print(event.tool.description) print(event.tool.to_openai_tool()) if isinstance(event, StreamChatDeltaReceivedEvent): print(event.delta) if isinstance(event, StreamChatErrorEvent): print(event.exception) if isinstance(event, EmbeddingStartEvent): print(event.model_dict) if isinstance(event, EmbeddingEndEvent): print(event.chunks) print(event.embeddings[0][:5]) # 避免打印所有 embeddings if isinstance(event, LLMPredictStartEvent): print(event.template) print(event.template_args) if isinstance(event, LLMPredictEndEvent): print(event.output) if isinstance(event, LLMStructuredPredictStartEvent): print(event.template) print(event.template_args) print(event.output_cls) if isinstance(event, LLMStructuredPredictEndEvent): print(event.output) if isinstance(event, LLMCompletionStartEvent): print(event.model_dict) print(event.prompt) print(event.additional_kwargs) if isinstance(event, LLMCompletionEndEvent): print(event.response) print(event.prompt) if isinstance(event, LLMChatInProgressEvent): print(event.messages) print(event.response) if isinstance(event, LLMChatStartEvent): print(event.messages) print(event.additional_kwargs) print(event.model_dict) if isinstance(event, LLMChatEndEvent): print(event.messages) print(event.response) if isinstance(event, RetrievalStartEvent): print(event.str_or_query_bundle) if isinstance(event, RetrievalEndEvent): print(event.str_or_query_bundle) print(event.nodes) if isinstance(event, ReRankStartEvent): print(event.query) print(event.nodes) print(event.top_n) print(event.model_name) if isinstance(event, ReRankEndEvent): print(event.nodes) if isinstance(event, QueryStartEvent): print(event.query) if isinstance(event, QueryEndEvent): print(event.response) print(event.query) if isinstance(event, SpanDropEvent): print(event.err_str) if isinstance(event, SynthesizeStartEvent): print(event.query) if isinstance(event, SynthesizeEndEvent): print(event.response) print(event.query) if isinstance(event, GetResponseStartEvent): print(event.query_str) self.events.append(event) print("-----------------------") def _get_events_by_span(self) -> Dict[str, List[BaseEvent]]: events_by_span: Dict[str, List[BaseEvent]] = {} for event in self.events: if event.span_id in events_by_span: events_by_span[event.span_id].append(event) else: events_by_span[event.span_id] = [event] return events_by_span def _get_event_span_trees(self) -> List[Tree]: events_by_span = self._get_events_by_span() trees = [] tree = Tree() for span, sorted_events in events_by_span.items(): # create root node i.e. span node tree.create_node( tag=f"{span} (SPAN)", identifier=span, parent=None, data=sorted_events[0].timestamp, ) for event in sorted_events: tree.create_node( tag=f"{event.class_name()}: {event.id_}", identifier=event.id_, parent=event.span_id, data=event.timestamp, ) trees.append(tree) tree = Tree() return trees def print_event_span_trees(self) -> None: """用于查看追踪树的方法。""" trees = self._get_event_span_trees() for tree in trees: print( tree.show( stdout=False, sorting=True, key=lambda node: node.data ) ) print("")

跨度（Spans）¶

跨度（Spans）是 LlamaIndex 中的“操作”（通常是函数调用）。跨度可以包含更多的跨度，并且每个跨度都包含关联的事件。

以下代码展示了如何观察 LlamaIndex 中发生的跨度

In [ ]

from typing import Any, Optional

from llama_index.core.instrumentation.span import SimpleSpan
from llama_index.core.instrumentation.span_handlers.base import BaseSpanHandler


class ExampleSpanHandler(BaseSpanHandler[SimpleSpan]):
    span_dict = {}

    @classmethod
    def class_name(cls) -> str:
        """Class name."""
        return "ExampleSpanHandler"

    def new_span(
        self,
        id_: str,
        bound_args: Any,
        instance: Optional[Any] = None,
        parent_span_id: Optional[str] = None,
        tags: Optional[Dict[str, Any]] = None,
        **kwargs: Any,
    ) -> Optional[SimpleSpan]:
        """Create a span."""
        # logic for creating a new MyCustomSpan
        if id_ not in self.span_dict:
            self.span_dict[id_] = []
        self.span_dict[id_].append(
            SimpleSpan(id_=id_, parent_id=parent_span_id)
        )

    def prepare_to_exit_span(
        self,
        id_: str,
        bound_args: Any,
        instance: Optional[Any] = None,
        result: Optional[Any] = None,
        **kwargs: Any,
    ) -> Any:
        """Logic for preparing to exit a span."""
        pass
        # if id in self.span_dict:
        #    return self.span_dict[id].pop()

    def prepare_to_drop_span(
        self,
        id_: str,
        bound_args: Any,
        instance: Optional[Any] = None,
        err: Optional[BaseException] = None,
        **kwargs: Any,
    ) -> Any:
        """Logic for preparing to drop a span."""
        pass
        # if id in self.span_dict:
        #    return self.span_dict[id].pop()

from typing import Any, Optional from llama_index.core.instrumentation.span import SimpleSpan from llama_index.core.instrumentation.span_handlers.base import BaseSpanHandler class ExampleSpanHandler(BaseSpanHandler[SimpleSpan]): span_dict = {} @classmethod def class_name(cls) -> str: """类名。""" return "ExampleSpanHandler" def new_span( self, id_: str, bound_args: Any, instance: Optional[Any] = None, parent_span_id: Optional[str] = None, tags: Optional[Dict[str, Any]] = None, **kwargs: Any, ) -> Optional[SimpleSpan]: """创建跨度。""" # 创建新 MyCustomSpan 的逻辑 if id_ not in self.span_dict: self.span_dict[id_] = [] self.span_dict[id_].append( SimpleSpan(id_=id_, parent_id=parent_span_id) ) def prepare_to_exit_span( self, id_: str, bound_args: Any, instance: Optional[Any] = None, result: Optional[Any] = None, **kwargs: Any, ) -> Any: """准备退出跨度的逻辑。""" pass # if id in self.span_dict: # return self.span_dict[id].pop() def prepare_to_drop_span( self, id_: str, bound_args: Any, instance: Optional[Any] = None, err: Optional[BaseException] = None, **kwargs: Any, ) -> Any: """准备丢弃跨度的逻辑。""" pass # if id in self.span_dict: # return self.span_dict[id].pop()

整合到一起¶

定义好我们的跨度处理器和事件处理器后，我们可以将其附加到调度器（dispatcher）上，来观察事件和跨度的到来。

并不强制要求同时拥有跨度处理器和事件处理器，你可以只使用其中一个，或者两者都用。

In [ ]

from llama_index.core.instrumentation import get_dispatcher
from llama_index.core.instrumentation.span_handlers import SimpleSpanHandler

# root dispatcher
root_dispatcher = get_dispatcher()

# register span handler
event_handler = ExampleEventHandler()
span_handler = ExampleSpanHandler()
simple_span_handler = SimpleSpanHandler()
root_dispatcher.add_span_handler(span_handler)
root_dispatcher.add_span_handler(simple_span_handler)
root_dispatcher.add_event_handler(event_handler)

from llama_index.core.instrumentation import get_dispatcher from llama_index.core.instrumentation.span_handlers import SimpleSpanHandler # root dispatcher root_dispatcher = get_dispatcher() # register span handler event_handler = ExampleEventHandler() span_handler = ExampleSpanHandler() simple_span_handler = SimpleSpanHandler() root_dispatcher.add_span_handler(span_handler) root_dispatcher.add_span_handler(simple_span_handler) root_dispatcher.add_event_handler(event_handler)

In [ ]

import os

os.environ["OPENAI_API_KEY"] = "sk-..."

import os os.environ["OPENAI_API_KEY"] = "sk-..."

In [ ]

from llama_index.core import Document, VectorStoreIndex

index = VectorStoreIndex.from_documents([Document.example()])

query_engine = index.as_query_engine()

query_engine.query("Tell me about LLMs?")

from llama_index.core import Document, VectorStoreIndex index = VectorStoreIndex.from_documents([Document.example()]) query_engine = index.as_query_engine() query_engine.query("Tell me about LLMs?")

-----------------------
7182e98f-1b8a-4aba-af18-3982b862c794
2024-05-06 14:00:35.931813
BaseEmbedding.get_text_embedding_batch-632972aa-3345-49cb-ae2f-46f3166e3afc
Event type: EmbeddingStartEvent
{'model_name': 'text-embedding-ada-002', 'embed_batch_size': 100, 'num_workers': None, 'additional_kwargs': {}, 'api_base': 'https://api.openai.com/v1', 'api_version': '', 'max_retries': 10, 'timeout': 60.0, 'default_headers': None, 'reuse_client': True, 'dimensions': None, 'class_name': 'OpenAIEmbedding'}
-----------------------
-----------------------
ba86e41f-cadf-4f1f-8908-8ee90404d668
2024-05-06 14:00:36.256237
BaseEmbedding.get_text_embedding_batch-632972aa-3345-49cb-ae2f-46f3166e3afc
Event type: EmbeddingEndEvent
['filename: README.md\ncategory: codebase\n\nContext\nLLMs are a phenomenal piece of technology for knowledge generation and reasoning.\nThey are pre-trained on large amounts of publicly available data.\nHow do we best augment LLMs with our own private data?\nWe need a comprehensive toolkit to help perform this data augmentation for LLMs.\n\nProposed Solution\nThat\'s where LlamaIndex comes in. LlamaIndex is a "data framework" to help\nyou build LLM  apps. It provides the following tools:\n\nOffers data connectors to ingest your existing data sources and data formats\n(APIs, PDFs, docs, SQL, etc.)\nProvides ways to structure your data (indices, graphs) so that this data can be\neasily used with LLMs.\nProvides an advanced retrieval/query interface over your data:\nFeed in any LLM input prompt, get back retrieved context and knowledge-augmented output.\nAllows easy integrations with your outer application framework\n(e.g. with LangChain, Flask, Docker, ChatGPT, anything else).\nLlamaIndex provides tools for both beginner users and advanced users.\nOur high-level API allows beginner users to use LlamaIndex to ingest and\nquery their data in 5 lines of code. Our lower-level APIs allow advanced users to\ncustomize and extend any module (data connectors, indices, retrievers, query engines,\nreranking modules), to fit their needs.']
[-0.005768016912043095, 0.02242799662053585, -0.020438531413674355, -0.040361806750297546, -0.01757599227130413]
-----------------------
-----------------------
06935377-f1e4-4fb9-b866-86f7520dfe2b
2024-05-06 14:00:36.305798
BaseQueryEngine.query-a766ae6c-6445-43b4-b1fc-9c29bae99556
Event type: QueryStartEvent
Tell me about LLMs?
-----------------------
-----------------------
62608f4f-67a1-4e2c-a653-24a4430529bb
2024-05-06 14:00:36.305998
BaseRetriever.retrieve-4e25a2a3-43a9-45e3-a7b9-59f4d54e8f00
Event type: RetrievalStartEvent
Tell me about LLMs?
-----------------------
-----------------------
e984c840-919b-4dc7-943d-5c49fbff48b8
2024-05-06 14:00:36.306265
BaseEmbedding.get_query_embedding-d30934f4-7bd2-4425-beda-12b5f55bc38b
Event type: EmbeddingStartEvent
{'model_name': 'text-embedding-ada-002', 'embed_batch_size': 100, 'num_workers': None, 'additional_kwargs': {}, 'api_base': 'https://api.openai.com/v1', 'api_version': '', 'max_retries': 10, 'timeout': 60.0, 'default_headers': None, 'reuse_client': True, 'dimensions': None, 'class_name': 'OpenAIEmbedding'}
-----------------------
-----------------------
c09fa993-a892-4efe-9f1b-7238ff6e5c62
2024-05-06 14:00:36.481459
BaseEmbedding.get_query_embedding-d30934f4-7bd2-4425-beda-12b5f55bc38b
Event type: EmbeddingEndEvent
['Tell me about LLMs?']
[0.00793155562132597, 0.011421983130276203, -0.010342259891331196, -0.03294854983687401, -0.03647972270846367]
-----------------------
-----------------------
b076d239-628d-4b4c-94ed-25aa2ca4b02b
2024-05-06 14:00:36.484080
BaseRetriever.retrieve-4e25a2a3-43a9-45e3-a7b9-59f4d54e8f00
Event type: RetrievalEndEvent
Tell me about LLMs?
[NodeWithScore(node=TextNode(id_='8de2b6b2-3fda-4f9b-95a8-a3ced6cfb0e5', embedding=None, metadata={'filename': 'README.md', 'category': 'codebase'}, excluded_embed_metadata_keys=[], excluded_llm_metadata_keys=[], relationships={<NodeRelationship.SOURCE: '1'>: RelatedNodeInfo(node_id='29e2bc8f-b62c-4752-b5eb-11346c5cbe50', node_type=<ObjectType.DOCUMENT: '4'>, metadata={'filename': 'README.md', 'category': 'codebase'}, hash='3183371414f6a23e9a61e11b45ec45f808b148f9973166cfed62226e3505eb05')}, text='Context\nLLMs are a phenomenal piece of technology for knowledge generation and reasoning.\nThey are pre-trained on large amounts of publicly available data.\nHow do we best augment LLMs with our own private data?\nWe need a comprehensive toolkit to help perform this data augmentation for LLMs.\n\nProposed Solution\nThat\'s where LlamaIndex comes in. LlamaIndex is a "data framework" to help\nyou build LLM  apps. It provides the following tools:\n\nOffers data connectors to ingest your existing data sources and data formats\n(APIs, PDFs, docs, SQL, etc.)\nProvides ways to structure your data (indices, graphs) so that this data can be\neasily used with LLMs.\nProvides an advanced retrieval/query interface over your data:\nFeed in any LLM input prompt, get back retrieved context and knowledge-augmented output.\nAllows easy integrations with your outer application framework\n(e.g. with LangChain, Flask, Docker, ChatGPT, anything else).\nLlamaIndex provides tools for both beginner users and advanced users.\nOur high-level API allows beginner users to use LlamaIndex to ingest and\nquery their data in 5 lines of code. Our lower-level APIs allow advanced users to\ncustomize and extend any module (data connectors, indices, retrievers, query engines,\nreranking modules), to fit their needs.', start_char_idx=1, end_char_idx=1279, text_template='{metadata_str}\n\n{content}', metadata_template='{key}: {value}', metadata_seperator='\n'), score=0.807312731672428)]
-----------------------
-----------------------
5e3289be-c597-48e7-ad3f-787722b766ea
2024-05-06 14:00:36.484436
BaseSynthesizer.synthesize-23d8d12d-a36e-423b-8776-042f1ff62546
Event type: SynthesizeStartEvent
Tell me about LLMs?
-----------------------
-----------------------
e9d9fe28-16d5-4301-8510-61aa11fa4951
2024-05-06 14:00:36.486070
Refine.get_response-e085393a-5510-4c3a-ba35-535caf58e159
Event type: GetResponseStartEvent
Tell me about LLMs?
-----------------------
-----------------------
29ce3778-d7cc-4095-b6b7-c811cd61ca5d
2024-05-06 14:00:36.486837
LLM.predict-007a74e7-34ff-488b-81b1-4ffb69df68a0
Event type: LLMPredictStartEvent
metadata={'prompt_type': <PromptType.QUESTION_ANSWER: 'text_qa'>} template_vars=['context_str', 'query_str'] kwargs={'query_str': 'Tell me about LLMs?'} output_parser=None template_var_mappings={} function_mappings={} default_template=PromptTemplate(metadata={'prompt_type': <PromptType.QUESTION_ANSWER: 'text_qa'>}, template_vars=['context_str', 'query_str'], kwargs={'query_str': 'Tell me about LLMs?'}, output_parser=None, template_var_mappings=None, function_mappings=None, template='Context information is below.\n---------------------\n{context_str}\n---------------------\nGiven the context information and not prior knowledge, answer the query.\nQuery: {query_str}\nAnswer: ') conditionals=[(<function is_chat_model at 0x13a72af80>, ChatPromptTemplate(metadata={'prompt_type': <PromptType.CUSTOM: 'custom'>}, template_vars=['context_str', 'query_str'], kwargs={'query_str': 'Tell me about LLMs?'}, output_parser=None, template_var_mappings=None, function_mappings=None, message_templates=[ChatMessage(role=<MessageRole.SYSTEM: 'system'>, content="You are an expert Q&A system that is trusted around the world.\nAlways answer the query using the provided context information, and not prior knowledge.\nSome rules to follow:\n1. Never directly reference the given context in your answer.\n2. Avoid statements like 'Based on the context, ...' or 'The context information ...' or anything along those lines.", additional_kwargs={}), ChatMessage(role=<MessageRole.USER: 'user'>, content='Context information is below.\n---------------------\n{context_str}\n---------------------\nGiven the context information and not prior knowledge, answer the query.\nQuery: {query_str}\nAnswer: ', additional_kwargs={})]))]
{'context_str': 'filename: README.md\ncategory: codebase\n\nContext\nLLMs are a phenomenal piece of technology for knowledge generation and reasoning.\nThey are pre-trained on large amounts of publicly available data.\nHow do we best augment LLMs with our own private data?\nWe need a comprehensive toolkit to help perform this data augmentation for LLMs.\n\nProposed Solution\nThat\'s where LlamaIndex comes in. LlamaIndex is a "data framework" to help\nyou build LLM  apps. It provides the following tools:\n\nOffers data connectors to ingest your existing data sources and data formats\n(APIs, PDFs, docs, SQL, etc.)\nProvides ways to structure your data (indices, graphs) so that this data can be\neasily used with LLMs.\nProvides an advanced retrieval/query interface over your data:\nFeed in any LLM input prompt, get back retrieved context and knowledge-augmented output.\nAllows easy integrations with your outer application framework\n(e.g. with LangChain, Flask, Docker, ChatGPT, anything else).\nLlamaIndex provides tools for both beginner users and advanced users.\nOur high-level API allows beginner users to use LlamaIndex to ingest and\nquery their data in 5 lines of code. Our lower-level APIs allow advanced users to\ncustomize and extend any module (data connectors, indices, retrievers, query engines,\nreranking modules), to fit their needs.'}
-----------------------
-----------------------
2042b4ab-99b4-410d-a997-ed97dda7a7d1
2024-05-06 14:00:36.487359
LLM.predict-007a74e7-34ff-488b-81b1-4ffb69df68a0
Event type: LLMChatStartEvent
[ChatMessage(role=<MessageRole.SYSTEM: 'system'>, content="You are an expert Q&A system that is trusted around the world.\nAlways answer the query using the provided context information, and not prior knowledge.\nSome rules to follow:\n1. Never directly reference the given context in your answer.\n2. Avoid statements like 'Based on the context, ...' or 'The context information ...' or anything along those lines.", additional_kwargs={}), ChatMessage(role=<MessageRole.USER: 'user'>, content='Context information is below.\n---------------------\nfilename: README.md\ncategory: codebase\n\nContext\nLLMs are a phenomenal piece of technology for knowledge generation and reasoning.\nThey are pre-trained on large amounts of publicly available data.\nHow do we best augment LLMs with our own private data?\nWe need a comprehensive toolkit to help perform this data augmentation for LLMs.\n\nProposed Solution\nThat\'s where LlamaIndex comes in. LlamaIndex is a "data framework" to help\nyou build LLM  apps. It provides the following tools:\n\nOffers data connectors to ingest your existing data sources and data formats\n(APIs, PDFs, docs, SQL, etc.)\nProvides ways to structure your data (indices, graphs) so that this data can be\neasily used with LLMs.\nProvides an advanced retrieval/query interface over your data:\nFeed in any LLM input prompt, get back retrieved context and knowledge-augmented output.\nAllows easy integrations with your outer application framework\n(e.g. with LangChain, Flask, Docker, ChatGPT, anything else).\nLlamaIndex provides tools for both beginner users and advanced users.\nOur high-level API allows beginner users to use LlamaIndex to ingest and\nquery their data in 5 lines of code. Our lower-level APIs allow advanced users to\ncustomize and extend any module (data connectors, indices, retrievers, query engines,\nreranking modules), to fit their needs.\n---------------------\nGiven the context information and not prior knowledge, answer the query.\nQuery: Tell me about LLMs?\nAnswer: ', additional_kwargs={})]
{}
{'system_prompt': None, 'pydantic_program_mode': <PydanticProgramMode.DEFAULT: 'default'>, 'model': 'gpt-3.5-turbo', 'temperature': 0.1, 'max_tokens': None, 'logprobs': None, 'top_logprobs': 0, 'additional_kwargs': {}, 'max_retries': 3, 'timeout': 60.0, 'default_headers': None, 'reuse_client': True, 'api_base': 'https://api.openai.com/v1', 'api_version': '', 'class_name': 'openai_llm'}
-----------------------
-----------------------
67b5c0f5-135e-4571-86a4-6e7efa6a40ff
2024-05-06 14:00:37.627923
LLM.predict-007a74e7-34ff-488b-81b1-4ffb69df68a0
Event type: LLMChatEndEvent
[ChatMessage(role=<MessageRole.SYSTEM: 'system'>, content="You are an expert Q&A system that is trusted around the world.\nAlways answer the query using the provided context information, and not prior knowledge.\nSome rules to follow:\n1. Never directly reference the given context in your answer.\n2. Avoid statements like 'Based on the context, ...' or 'The context information ...' or anything along those lines.", additional_kwargs={}), ChatMessage(role=<MessageRole.USER: 'user'>, content='Context information is below.\n---------------------\nfilename: README.md\ncategory: codebase\n\nContext\nLLMs are a phenomenal piece of technology for knowledge generation and reasoning.\nThey are pre-trained on large amounts of publicly available data.\nHow do we best augment LLMs with our own private data?\nWe need a comprehensive toolkit to help perform this data augmentation for LLMs.\n\nProposed Solution\nThat\'s where LlamaIndex comes in. LlamaIndex is a "data framework" to help\nyou build LLM  apps. It provides the following tools:\n\nOffers data connectors to ingest your existing data sources and data formats\n(APIs, PDFs, docs, SQL, etc.)\nProvides ways to structure your data (indices, graphs) so that this data can be\neasily used with LLMs.\nProvides an advanced retrieval/query interface over your data:\nFeed in any LLM input prompt, get back retrieved context and knowledge-augmented output.\nAllows easy integrations with your outer application framework\n(e.g. with LangChain, Flask, Docker, ChatGPT, anything else).\nLlamaIndex provides tools for both beginner users and advanced users.\nOur high-level API allows beginner users to use LlamaIndex to ingest and\nquery their data in 5 lines of code. Our lower-level APIs allow advanced users to\ncustomize and extend any module (data connectors, indices, retrievers, query engines,\nreranking modules), to fit their needs.\n---------------------\nGiven the context information and not prior knowledge, answer the query.\nQuery: Tell me about LLMs?\nAnswer: ', additional_kwargs={})]
assistant: LLMs are a type of technology used for knowledge generation and reasoning. They are pre-trained on large amounts of publicly available data.
-----------------------
-----------------------
42cb1fc6-3d8a-4dce-81f1-de43617a37fd
2024-05-06 14:00:37.628432
LLM.predict-007a74e7-34ff-488b-81b1-4ffb69df68a0
Event type: LLMPredictEndEvent
LLMs are a type of technology used for knowledge generation and reasoning. They are pre-trained on large amounts of publicly available data.
-----------------------
-----------------------
4498248d-d07a-4460-87c7-3a6f310c4cb3
2024-05-06 14:00:37.628634
Refine.get_response-e085393a-5510-4c3a-ba35-535caf58e159
Event type: GetResponseEndEvent
LLMs are a type of technology used for knowledge generation and reasoning. They are pre-trained on large amounts of publicly available data.
-----------------------
-----------------------
f1d7fda7-de82-4149-8cd9-b9a17dba169b
2024-05-06 14:00:37.628826
BaseSynthesizer.synthesize-23d8d12d-a36e-423b-8776-042f1ff62546
Event type: SynthesizeEndEvent
LLMs are a type of technology used for knowledge generation and reasoning. They are pre-trained on large amounts of publicly available data.
Tell me about LLMs?
-----------------------
-----------------------
2f564649-dbbb-4adc-a552-552f54358112
2024-05-06 14:00:37.629251
BaseQueryEngine.query-a766ae6c-6445-43b4-b1fc-9c29bae99556
Event type: QueryEndEvent
LLMs are a type of technology used for knowledge generation and reasoning. They are pre-trained on large amounts of publicly available data.
Tell me about LLMs?
-----------------------

Out[ ]

Response(response='LLMs are a type of technology used for knowledge generation and reasoning. They are pre-trained on large amounts of publicly available data.', source_nodes=[NodeWithScore(node=TextNode(id_='8de2b6b2-3fda-4f9b-95a8-a3ced6cfb0e5', embedding=None, metadata={'filename': 'README.md', 'category': 'codebase'}, excluded_embed_metadata_keys=[], excluded_llm_metadata_keys=[], relationships={<NodeRelationship.SOURCE: '1'>: RelatedNodeInfo(node_id='29e2bc8f-b62c-4752-b5eb-11346c5cbe50', node_type=<ObjectType.DOCUMENT: '4'>, metadata={'filename': 'README.md', 'category': 'codebase'}, hash='3183371414f6a23e9a61e11b45ec45f808b148f9973166cfed62226e3505eb05')}, text='Context\nLLMs are a phenomenal piece of technology for knowledge generation and reasoning.\nThey are pre-trained on large amounts of publicly available data.\nHow do we best augment LLMs with our own private data?\nWe need a comprehensive toolkit to help perform this data augmentation for LLMs.\n\nProposed Solution\nThat\'s where LlamaIndex comes in. LlamaIndex is a "data framework" to help\nyou build LLM  apps. It provides the following tools:\n\nOffers data connectors to ingest your existing data sources and data formats\n(APIs, PDFs, docs, SQL, etc.)\nProvides ways to structure your data (indices, graphs) so that this data can be\neasily used with LLMs.\nProvides an advanced retrieval/query interface over your data:\nFeed in any LLM input prompt, get back retrieved context and knowledge-augmented output.\nAllows easy integrations with your outer application framework\n(e.g. with LangChain, Flask, Docker, ChatGPT, anything else).\nLlamaIndex provides tools for both beginner users and advanced users.\nOur high-level API allows beginner users to use LlamaIndex to ingest and\nquery their data in 5 lines of code. Our lower-level APIs allow advanced users to\ncustomize and extend any module (data connectors, indices, retrievers, query engines,\nreranking modules), to fit their needs.', start_char_idx=1, end_char_idx=1279, text_template='{metadata_str}\n\n{content}', metadata_template='{key}: {value}', metadata_seperator='\n'), score=0.807312731672428)], metadata={'8de2b6b2-3fda-4f9b-95a8-a3ced6cfb0e5': {'filename': 'README.md', 'category': 'codebase'}})

In [ ]

event_handler.print_event_span_trees()

event_handler.print_event_span_trees()

BaseEmbedding.get_text_embedding_batch-632972aa-3345-49cb-ae2f-46f3166e3afc (SPAN)
├── EmbeddingStartEvent: 7182e98f-1b8a-4aba-af18-3982b862c794
└── EmbeddingEndEvent: ba86e41f-cadf-4f1f-8908-8ee90404d668


BaseQueryEngine.query-a766ae6c-6445-43b4-b1fc-9c29bae99556 (SPAN)
├── QueryStartEvent: 06935377-f1e4-4fb9-b866-86f7520dfe2b
└── QueryEndEvent: 2f564649-dbbb-4adc-a552-552f54358112


BaseRetriever.retrieve-4e25a2a3-43a9-45e3-a7b9-59f4d54e8f00 (SPAN)
├── RetrievalStartEvent: 62608f4f-67a1-4e2c-a653-24a4430529bb
└── RetrievalEndEvent: b076d239-628d-4b4c-94ed-25aa2ca4b02b


BaseEmbedding.get_query_embedding-d30934f4-7bd2-4425-beda-12b5f55bc38b (SPAN)
├── EmbeddingStartEvent: e984c840-919b-4dc7-943d-5c49fbff48b8
└── EmbeddingEndEvent: c09fa993-a892-4efe-9f1b-7238ff6e5c62


BaseSynthesizer.synthesize-23d8d12d-a36e-423b-8776-042f1ff62546 (SPAN)
├── SynthesizeStartEvent: 5e3289be-c597-48e7-ad3f-787722b766ea
└── SynthesizeEndEvent: f1d7fda7-de82-4149-8cd9-b9a17dba169b


Refine.get_response-e085393a-5510-4c3a-ba35-535caf58e159 (SPAN)
├── GetResponseStartEvent: e9d9fe28-16d5-4301-8510-61aa11fa4951
└── GetResponseEndEvent: 4498248d-d07a-4460-87c7-3a6f310c4cb3


LLM.predict-007a74e7-34ff-488b-81b1-4ffb69df68a0 (SPAN)
├── LLMPredictStartEvent: 29ce3778-d7cc-4095-b6b7-c811cd61ca5d
├── LLMChatStartEvent: 2042b4ab-99b4-410d-a997-ed97dda7a7d1
├── LLMChatEndEvent: 67b5c0f5-135e-4571-86a4-6e7efa6a40ff
└── LLMPredictEndEvent: 42cb1fc6-3d8a-4dce-81f1-de43617a37fd

In [ ]

simple_span_handler.print_trace_trees()

simple_span_handler.print_trace_trees()

BaseEmbedding.get_text_embedding_batch-632972aa-3345-49cb-ae2f-46f3166e3afc (0.326418)


BaseQueryEngine.query-a766ae6c-6445-43b4-b1fc-9c29bae99556 (1.323617)
└── RetrieverQueryEngine._query-40135aed-9aa5-4197-a05d-d461afb524d0 (1.32328)
    ├── BaseRetriever.retrieve-4e25a2a3-43a9-45e3-a7b9-59f4d54e8f00 (0.178294)
    │   └── VectorIndexRetriever._retrieve-8ead50e0-7243-42d1-b1ed-d2a2f2ceea48 (0.177893)
    │       └── BaseEmbedding.get_query_embedding-d30934f4-7bd2-4425-beda-12b5f55bc38b (0.176907)
    └── BaseSynthesizer.synthesize-23d8d12d-a36e-423b-8776-042f1ff62546 (1.144761)
        └── CompactAndRefine.get_response-ec49a727-bf17-4d80-bf82-80ec2a906063 (1.144148)
            └── Refine.get_response-e085393a-5510-4c3a-ba35-535caf58e159 (1.142698)
                └── LLM.predict-007a74e7-34ff-488b-81b1-4ffb69df68a0 (1.141744)