From ec77e445e05e525df4511ce6a3082ec09b6c07ff Mon Sep 17 00:00:00 2001
From: Simon Suo <simonsdsuo@gmail.com>
Date: Thu, 24 Aug 2023 15:07:44 -0700
Subject: [PATCH] Prompt refactor (#7319)
---
CHANGELOG.md | 10 +
docs/api_reference/prompts.rst | 26 +-
.../model_modules/llms/usage_custom.md | 4 +-
docs/core_modules/model_modules/prompts.md | 36 +-
.../chat_engines/usage_pattern.md | 4 +-
.../structured_outputs/output_parser.md | 12 +-
.../terms_definitions_tutorial.md | 38 +-
.../SimpleIndexDemo-Huggingface_camel.ipynb | 4 +-
...SimpleIndexDemo-Huggingface_stablelm.ipynb | 4 +-
.../customization/prompts/chat_prompts.ipynb | 91 ++--
.../prompts/completion_prompts.ipynb | 34 +-
.../output_parsing/GuardrailsDemo.ipynb | 6 +-
.../LangchainOutputParserDemo.ipynb | 6 +-
.../query_engine/ensemble_query_engine.ipynb | 8 +-
.../SimpleIndexDemoLlama-Local.ipynb | 4 +-
examples/paul_graham_essay/TestEssay.ipynb | 4 +-
.../test_wiki/TestNYC-Benchmark-GPT4.ipynb | 4 +-
experimental/classifier/utils.py | 21 +-
llama_index/__init__.py | 13 +-
llama_index/agent/context_retriever_agent.py | 12 +-
llama_index/chat_engine/condense_question.py | 10 +-
llama_index/evaluation/base.py | 22 +-
llama_index/evaluation/dataset_generation.py | 8 +-
llama_index/evaluation/guideline_eval.py | 6 +-
.../indices/common/struct_store/base.py | 31 +-
.../indices/common/struct_store/sql.py | 4 +-
llama_index/indices/common_tree/base.py | 11 +-
llama_index/indices/document_summary/base.py | 8 +-
.../indices/document_summary/retrievers.py | 6 +-
llama_index/indices/empty/retrievers.py | 6 +-
llama_index/indices/keyword_table/base.py | 8 +-
.../indices/keyword_table/retrievers.py | 14 +-
llama_index/indices/knowledge_graph/base.py | 6 +-
.../indices/knowledge_graph/retrievers.py | 21 +-
llama_index/indices/list/base.py | 4 +-
llama_index/indices/list/retrievers.py | 8 +-
.../indices/postprocessor/llm_rerank.py | 6 +-
llama_index/indices/postprocessor/node.py | 6 +-
.../indices/postprocessor/node_recency.py | 34 +-
llama_index/indices/postprocessor/pii.py | 4 +-
llama_index/indices/prompt_helper.py | 24 +-
.../indices/query/query_transform/base.py | 6 +-
.../query_transform/feedback_transform.py | 8 +-
.../indices/query/query_transform/prompts.py | 20 +-
llama_index/indices/service_context.py | 10 +-
llama_index/indices/struct_store/base.py | 4 +-
.../indices/struct_store/json_query.py | 10 +-
llama_index/indices/struct_store/sql_query.py | 29 +-
.../indices/tree/all_leaf_retriever.py | 2 +-
llama_index/indices/tree/base.py | 12 +-
llama_index/indices/tree/inserter.py | 6 +-
.../tree/select_leaf_embedding_retriever.py | 13 +-
.../indices/tree/select_leaf_retriever.py | 22 +-
.../retrievers/auto_retriever/prompts.py | 6 +-
llama_index/llm_predictor/base.py | 77 +++-
llama_index/llm_predictor/mock.py | 25 +-
llama_index/llm_predictor/structured.py | 20 +-
llama_index/llm_predictor/vellum/predictor.py | 38 +-
.../llm_predictor/vellum/prompt_registry.py | 34 +-
llama_index/llms/generic_utils.py | 4 +-
llama_index/llms/palm.py | 4 +-
.../extractors/metadata_extractors.py | 19 +-
llama_index/program/llm_program.py | 14 +-
llama_index/program/openai_program.py | 6 +-
llama_index/program/predefined/df.py | 11 +-
.../program/predefined/evaporate/extractor.py | 4 +-
.../program/predefined/evaporate/prompts.py | 24 +-
llama_index/prompts/__init__.py | 24 +-
llama_index/prompts/base.py | 408 ++++++++++++------
llama_index/prompts/chat_prompts.py | 157 ++++---
llama_index/prompts/choice_select.py | 34 --
.../prompts/default_prompt_selectors.py | 62 +--
llama_index/prompts/default_prompts.py | 74 +++-
llama_index/prompts/prompt_selector.py | 26 --
llama_index/prompts/prompt_utils.py | 31 ++
llama_index/prompts/prompts.py | 65 +--
llama_index/prompts/utils.py | 39 +-
.../query_engine/citation_query_engine.py | 16 +-
.../query_engine/flare/answer_inserter.py | 6 +-
llama_index/query_engine/flare/base.py | 8 +-
.../knowledge_graph_query_engine.py | 12 +-
.../query_engine/pandas_query_engine.py | 10 +-
.../query_engine/retriever_query_engine.py | 20 +-
.../query_engine/sql_join_query_engine.py | 65 +--
.../query_engine/sql_vector_query_engine.py | 35 +-
llama_index/question_gen/llm_generators.py | 6 +-
llama_index/question_gen/prompts.py | 4 +-
.../response_synthesizers/accumulate.py | 10 +-
.../compact_and_refine.py | 2 +-
llama_index/response_synthesizers/factory.py | 19 +-
.../response_synthesizers/generation.py | 4 +-
llama_index/response_synthesizers/refine.py | 8 +-
.../response_synthesizers/simple_summarize.py | 4 +-
.../response_synthesizers/tree_summarize.py | 4 +-
llama_index/selectors/prompts.py | 14 +-
llama_index/types.py | 13 +-
tests/indices/list/test_retrievers.py | 8 +-
.../indices/postprocessor/test_llm_rerank.py | 17 +-
.../indices/response/test_response_builder.py | 26 +-
tests/indices/response/test_tree_summarize.py | 12 +-
tests/indices/test_prompt_helper.py | 30 +-
tests/llm_predictor/test_base.py | 10 +-
tests/llm_predictor/vellum/conftest.py | 18 +-
tests/llm_predictor/vellum/test_predictor.py | 10 +-
.../vellum/test_prompt_registry.py | 10 +-
tests/mock_utils/mock_predict.py | 60 +--
tests/mock_utils/mock_prompts.py | 40 +-
tests/prompts/test_base.py | 173 ++++----
tests/prompts/test_utils.py | 7 +
109 files changed, 1424 insertions(+), 1193 deletions(-)
delete mode 100644 llama_index/prompts/choice_select.py
delete mode 100644 llama_index/prompts/prompt_selector.py
create mode 100644 llama_index/prompts/prompt_utils.py
create mode 100644 tests/prompts/test_utils.py
diff --git a/CHANGELOG.md b/CHANGELOG.md
index ea419b6af6..60f56b6e49 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -12,6 +12,16 @@
- Ensure temperature is a float for openai (#7382)
- Remove duplicate subjects in knowledge graph retriever (#7378)
+### Breaking/Deprecated API Changes
+- Refactor prompt template (#7319)
+ - Use `BasePromptTemplate` for generic typing
+ - Use `PromptTemplate`, `ChatPromptTemplate`, `SelectorPromptTemplate` as core implementations
+ - Use `LangchainPromptTemplate` for compatibility with Langchain prompt templates
+ - Fully replace specific prompt classes (e.g. `SummaryPrompt`) with generic `BasePromptTemplate` for typing in codebase.
+ - Keep `Prompt` as an alias for `PromptTemplate` for backwards compatibility.
+ - BREAKING CHANGE: remove support for `Prompt.from_langchain_prompt`, please use `template=LangchainPromptTemplate(lc_template)` instead.
+
+
## [0.8.8] - 2023-08-23
### New Features
diff --git a/docs/api_reference/prompts.rst b/docs/api_reference/prompts.rst
index 9bc92c46b0..1b64c24455 100644
--- a/docs/api_reference/prompts.rst
+++ b/docs/api_reference/prompts.rst
@@ -7,26 +7,30 @@ These are the reference prompt templates.
We first show links to default prompts.
-We then show the base prompt class,
-derived from `Langchain <https://langchain.readthedocs.io/en/latest/modules/prompt.html>`_.
+We then show the base prompt template class and its subclasses.
Default Prompts
^^^^^^^^^^^^^^^^^
-The list of default prompts can be `found here <https://github.com/jerryjliu/llama_index/blob/main/llama_index/prompts/default_prompts.py>`_.
-**NOTE**: we've also curated a set of refine prompts for ChatGPT use cases.
-The list of ChatGPT refine prompts can be
-`found here <https://github.com/jerryjliu/llama_index/blob/main/llama_index/prompts/chat_prompts.py>`_.
+* `Completion prompt templates <https://github.com/jerryjliu/llama_index/blob/main/llama_index/prompts/default_prompts.py>`_.
+* `Chat prompt templates <https://github.com/jerryjliu/llama_index/blob/main/llama_index/prompts/chat_prompts.py>`_.
+* `Selector prompt templates <https://github.com/jerryjliu/llama_index/blob/main/llama_index/prompts/default_prompt_selectors.py>`_.
-Base Prompt Class
+
+Prompt Classes
^^^^^^^^^^^^^^^^^
-.. automodule:: llama_index.prompts
- :members:
- :inherited-members:
- :exclude-members: Config, construct, copy, dict, from_examples, from_file, get_full_format_args, output_parser, save, template, template_format, update_forward_refs, validate_variable_names, json, template_is_valid
+.. autopydantic_model:: llama_index.prompts.base.BasePromptTemplate
+
+.. autopydantic_model:: llama_index.prompts.base.PromptTemplate
+
+.. autopydantic_model:: llama_index.prompts.base.ChatPromptTemplate
+
+.. autopydantic_model:: llama_index.prompts.base.SelectorPromptTemplate
+
+.. autopydantic_model:: llama_index.prompts.base.LangchainPromptTemplate
Subclass Prompts (deprecated)
diff --git a/docs/core_modules/model_modules/llms/usage_custom.md b/docs/core_modules/model_modules/llms/usage_custom.md
index 9b92ac4865..f24a12d793 100644
--- a/docs/core_modules/model_modules/llms/usage_custom.md
+++ b/docs/core_modules/model_modules/llms/usage_custom.md
@@ -119,7 +119,7 @@ Many open-source models from HuggingFace require either some preamble before eac
Below, this example uses both the `system_prompt` and `query_wrapper_prompt`, using specific prompts from the model card found [here](https://huggingface.co/stabilityai/stablelm-tuned-alpha-3b).
```python
-from llama_index.prompts.prompts import SimpleInputPrompt
+from llama_index.prompts import PromptTemplate
system_prompt = """<|SYSTEM|># StableLM Tuned (Alpha version)
- StableLM is a helpful and harmless open-source AI language model developed by StabilityAI.
@@ -129,7 +129,7 @@ system_prompt = """<|SYSTEM|># StableLM Tuned (Alpha version)
"""
# This will wrap the default prompts that are internal to llama-index
-query_wrapper_prompt = SimpleInputPrompt("<|USER|>{query_str}<|ASSISTANT|>")
+query_wrapper_prompt = PromptTemplate("<|USER|>{query_str}<|ASSISTANT|>")
import torch
from llama_index.llms import HuggingFaceLLM
diff --git a/docs/core_modules/model_modules/prompts.md b/docs/core_modules/model_modules/prompts.md
index ca551647e5..1d078d2a78 100644
--- a/docs/core_modules/model_modules/prompts.md
+++ b/docs/core_modules/model_modules/prompts.md
@@ -18,7 +18,7 @@ Users may also provide their own prompt templates to further customize the behav
Defining a custom prompt is as simple as creating a format string
```python
-from llama_index import Prompt
+from llama_index.prompts import PromptTemplate
template = (
"We have provided context information below. \n"
@@ -27,10 +27,36 @@ template = (
"\n---------------------\n"
"Given this information, please answer the question: {query_str}\n"
)
-qa_template = Prompt(template)
+qa_template = PromptTemplate(template)
+
+# you can create text prompt (for completion API)
+prompt = qa_template.format(context_str=..., query_str=...)
+
+# or easily convert to message prompts (for chat API)
+messages = qa_template.format_messages(context_str=..., query_str=...)
```
-> Note: you may see references to legacy prompt subclasses such as `QuestionAnswerPrompt`, `RefinePrompt`. These have been deprecated (and now are type aliases of `Prompt`). Now you can directly specify `Prompt(template)` to construct custom prompts. But you still have to make sure the template string contains the expected parameters (e.g. `{context_str}` and `{query_str}`) when replacing a default question answer prompt.
+> Note: you may see references to legacy prompt subclasses such as `QuestionAnswerPrompt`, `RefinePrompt`. These have been deprecated (and now are type aliases of `PromptTemplate`). Now you can directly specify `PromptTemplate(template)` to construct custom prompts. But you still have to make sure the template string contains the expected parameters (e.g. `{context_str}` and `{query_str}`) when replacing a default question answer prompt.
+
+You can also define a template from chat messages
+```python
+from llama_index.prompts import ChatPromptTemplate, ChatMessage, MessageRole
+
+message_templates = [
+ ChatMessage(content="You are an expert system.", role=MessageRole.SYSTEM),
+ ChatMessage(
+ content="Generate a short story about {topic}",
+ role=MessageRole.USER,
+ ),
+]
+chat_template = ChatPromptTemplate(message_templates=message_templates)
+
+# you can create message prompts (for chat API)
+messages = chat_template.format_messages(topic=...)
+
+# or easily convert to text prompt (for completion API)
+prompt = chat_template.format(topic=...)
+```
### Passing custom prompts into the pipeline
@@ -45,8 +71,8 @@ The most commonly used prompts will be the `text_qa_template` and the `refine_te
#### Modify prompts used in index construction
Different indices use different types of prompts during construction (some don't use prompts at all).
-For instance, `TreeIndex` uses a `SummaryPrompt` to hierarchically
-summarize the nodes, and `KeywordTableIndex` uses a `KeywordExtractPrompt` to extract keywords.
+For instance, `TreeIndex` uses a summary prompt to hierarchically
+summarize the nodes, and `KeywordTableIndex` uses a keyword extract prompt to extract keywords.
There are two equivalent ways to override the prompts:
diff --git a/docs/core_modules/query_modules/chat_engines/usage_pattern.md b/docs/core_modules/query_modules/chat_engines/usage_pattern.md
index 63fefcc6f1..883a57fc7e 100644
--- a/docs/core_modules/query_modules/chat_engines/usage_pattern.md
+++ b/docs/core_modules/query_modules/chat_engines/usage_pattern.md
@@ -63,10 +63,10 @@ Here's an example where we configure the following:
* print verbose debug message.
```python
-from llama_index.prompts import Prompt
+from llama_index.prompts import PromptTemplate
from llama_index.llms import ChatMessage, MessageRole
-custom_prompt = Prompt("""\
+custom_prompt = PromptTemplate("""\
Given a conversation (between Human and Assistant) and a follow up message from Human, \
rewrite the message to be a standalone question that captures all relevant context \
from the conversation.
diff --git a/docs/core_modules/query_modules/structured_outputs/output_parser.md b/docs/core_modules/query_modules/structured_outputs/output_parser.md
index c5fcc15e7d..f2ba5328b2 100644
--- a/docs/core_modules/query_modules/structured_outputs/output_parser.md
+++ b/docs/core_modules/query_modules/structured_outputs/output_parser.md
@@ -15,7 +15,7 @@ Guardrails is an open-source Python package for specification/validation/correct
from llama_index import VectorStoreIndex, SimpleDirectoryReader
from llama_index.output_parsers import GuardrailsOutputParser
from llama_index.llm_predictor import StructuredLLMPredictor
-from llama_index.prompts.prompts import QuestionAnswerPrompt, RefinePrompt
+from llama_index.prompts import PromptTemplate
from llama_index.prompts.default_prompts import DEFAULT_TEXT_QA_PROMPT_TMPL, DEFAULT_REFINE_PROMPT_TMPL
@@ -62,8 +62,8 @@ output_parser = GuardrailsOutputParser.from_rail_string(rail_spec, llm=llm_predi
fmt_qa_tmpl = output_parser.format(DEFAULT_TEXT_QA_PROMPT_TMPL)
fmt_refine_tmpl = output_parser.format(DEFAULT_REFINE_PROMPT_TMPL)
-qa_prompt = QuestionAnswerPrompt(fmt_qa_tmpl, output_parser=output_parser)
-refine_prompt = RefinePrompt(fmt_refine_tmpl, output_parser=output_parser)
+qa_prompt = PromptTemplate(fmt_qa_tmpl, output_parser=output_parser)
+refine_prompt = PromptTemplate(fmt_refine_tmpl, output_parser=output_parser)
# obtain a structured response
query_engine = index.as_query_engine(
@@ -94,7 +94,7 @@ Langchain also offers output parsing modules that you can use within LlamaIndex.
from llama_index import VectorStoreIndex, SimpleDirectoryReader
from llama_index.output_parsers import LangchainOutputParser
from llama_index.llm_predictor import StructuredLLMPredictor
-from llama_index.prompts.prompts import QuestionAnswerPrompt, RefinePrompt
+from llama_index.prompts import PromptTemplate
from llama_index.prompts.default_prompts import DEFAULT_TEXT_QA_PROMPT_TMPL, DEFAULT_REFINE_PROMPT_TMPL
from langchain.output_parsers import StructuredOutputParser, ResponseSchema
@@ -117,8 +117,8 @@ output_parser = LangchainOutputParser(lc_output_parser)
# format each prompt with output parser instructions
fmt_qa_tmpl = output_parser.format(DEFAULT_TEXT_QA_PROMPT_TMPL)
fmt_refine_tmpl = output_parser.format(DEFAULT_REFINE_PROMPT_TMPL)
-qa_prompt = QuestionAnswerPrompt(fmt_qa_tmpl, output_parser=output_parser)
-refine_prompt = RefinePrompt(fmt_refine_tmpl, output_parser=output_parser)
+qa_prompt = PromptTemplate(fmt_qa_tmpl, output_parser=output_parser)
+refine_prompt = PromptTemplate(fmt_refine_tmpl, output_parser=output_parser)
# query index
query_engine = index.as_query_engine(
diff --git a/docs/end_to_end_tutorials/question_and_answer/terms_definitions_tutorial.md b/docs/end_to_end_tutorials/question_and_answer/terms_definitions_tutorial.md
index 98f79ee682..0f576b5ffc 100644
--- a/docs/end_to_end_tutorials/question_and_answer/terms_definitions_tutorial.md
+++ b/docs/end_to_end_tutorials/question_and_answer/terms_definitions_tutorial.md
@@ -298,14 +298,9 @@ This is due to the concept of "refining" answers in Llama Index. Since we are qu
So, the refine process seems to be messing with our results! Rather than appending extra instructions to the `query_str`, remove that, and Llama Index will let us provide our own custom prompts! Let's create those now, using the [default prompts](https://github.com/jerryjliu/llama_index/blob/main/llama_index/prompts/default_prompts.py) and [chat specific prompts](https://github.com/jerryjliu/llama_index/blob/main/llama_index/prompts/chat_prompts.py) as a guide. Using a new file `constants.py`, let's create some new query templates:
```python
-from langchain.chains.prompt_selector import ConditionalPromptSelector, is_chat_model
-from langchain.prompts.chat import (
- AIMessagePromptTemplate,
- ChatPromptTemplate,
- HumanMessagePromptTemplate,
-)
-
-from llama_index.prompts.prompts import QuestionAnswerPrompt, RefinePrompt
+from llama_index.prompts import PromptTemplate, SelectorPromptTemplate, ChatPromptTemplate
+from llama_index.prompts.utils import is_chat_model
+from llama_index.llms.base import ChatMessage, MessageRole
# Text QA templates
DEFAULT_TEXT_QA_PROMPT_TMPL = (
@@ -316,7 +311,7 @@ DEFAULT_TEXT_QA_PROMPT_TMPL = (
"Given the context information answer the following question "
"(if you don't know the answer, use the best of your knowledge): {query_str}\n"
)
-TEXT_QA_TEMPLATE = QuestionAnswerPrompt(DEFAULT_TEXT_QA_PROMPT_TMPL)
+TEXT_QA_TEMPLATE = PromptTemplate(DEFAULT_TEXT_QA_PROMPT_TMPL)
# Refine templates
DEFAULT_REFINE_PROMPT_TMPL = (
@@ -330,32 +325,29 @@ DEFAULT_REFINE_PROMPT_TMPL = (
"Given the new context and using the best of your knowledge, improve the existing answer. "
"If you can't improve the existing answer, just repeat it again."
)
-DEFAULT_REFINE_PROMPT = RefinePrompt(DEFAULT_REFINE_PROMPT_TMPL)
+DEFAULT_REFINE_PROMPT = PromptTemplate(DEFAULT_REFINE_PROMPT_TMPL)
CHAT_REFINE_PROMPT_TMPL_MSGS = [
- HumanMessagePromptTemplate.from_template("{query_str}"),
- AIMessagePromptTemplate.from_template("{existing_answer}"),
- HumanMessagePromptTemplate.from_template(
- "We have the opportunity to refine the above answer "
+ ChatMessage(content="{query_str}", role=MessageRole.USER),
+ ChatMessage(content="{existing_answer}", role=MessageRole.ASSISTANT),
+ ChatMessage(
+ content="We have the opportunity to refine the above answer "
"(only if needed) with some more context below.\n"
"------------\n"
"{context_msg}\n"
"------------\n"
"Given the new context and using the best of your knowledge, improve the existing answer. "
- "If you can't improve the existing answer, just repeat it again."
+ "If you can't improve the existing answer, just repeat it again.",
+ role=MessageRole.USER,
),
]
-CHAT_REFINE_PROMPT_LC = ChatPromptTemplate.from_messages(CHAT_REFINE_PROMPT_TMPL_MSGS)
-CHAT_REFINE_PROMPT = RefinePrompt.from_langchain_prompt(CHAT_REFINE_PROMPT_LC)
+CHAT_REFINE_PROMPT = ChatPromptTemplate(CHAT_REFINE_PROMPT_TMPL_MSGS)
# refine prompt selector
-DEFAULT_REFINE_PROMPT_SEL_LC = ConditionalPromptSelector(
- default_prompt=DEFAULT_REFINE_PROMPT.get_langchain_prompt(),
- conditionals=[(is_chat_model, CHAT_REFINE_PROMPT.get_langchain_prompt())],
-)
-REFINE_TEMPLATE = RefinePrompt(
- langchain_prompt_selector=DEFAULT_REFINE_PROMPT_SEL_LC
+REFINE_TEMPLATE = SelectorPromptTemplate(
+ default_template=DEFAULT_REFINE_PROMPT,
+ conditionals=[(is_chat_model, CHAT_REFINE_PROMPT)],
)
```
diff --git a/docs/examples/customization/llms/SimpleIndexDemo-Huggingface_camel.ipynb b/docs/examples/customization/llms/SimpleIndexDemo-Huggingface_camel.ipynb
index a91fb6fdd0..0fce5ea509 100644
--- a/docs/examples/customization/llms/SimpleIndexDemo-Huggingface_camel.ipynb
+++ b/docs/examples/customization/llms/SimpleIndexDemo-Huggingface_camel.ipynb
@@ -75,11 +75,11 @@
"outputs": [],
"source": [
"# setup prompts - specific to StableLM\n",
- "from llama_index.prompts.prompts import SimpleInputPrompt\n",
+ "from llama_index.prompts import PromptTemplate\n",
"\n",
"# This will wrap the default prompts that are internal to llama-index\n",
"# taken from https://huggingface.co/Writer/camel-5b-hf\n",
- "query_wrapper_prompt = SimpleInputPrompt(\n",
+ "query_wrapper_prompt = PromptTemplate(\n",
" \"Below is an instruction that describes a task. \"\n",
" \"Write a response that appropriately completes the request.\\n\\n\"\n",
" \"### Instruction:\\n{query_str}\\n\\n### Response:\"\n",
diff --git a/docs/examples/customization/llms/SimpleIndexDemo-Huggingface_stablelm.ipynb b/docs/examples/customization/llms/SimpleIndexDemo-Huggingface_stablelm.ipynb
index 77955e9da5..6919c882c1 100644
--- a/docs/examples/customization/llms/SimpleIndexDemo-Huggingface_stablelm.ipynb
+++ b/docs/examples/customization/llms/SimpleIndexDemo-Huggingface_stablelm.ipynb
@@ -75,7 +75,7 @@
"outputs": [],
"source": [
"# setup prompts - specific to StableLM\n",
- "from llama_index.prompts.prompts import SimpleInputPrompt\n",
+ "from llama_index.prompts import PromptTemplate\n",
"\n",
"system_prompt = \"\"\"<|SYSTEM|># StableLM Tuned (Alpha version)\n",
"- StableLM is a helpful and harmless open-source AI language model developed by StabilityAI.\n",
@@ -85,7 +85,7 @@
"\"\"\"\n",
"\n",
"# This will wrap the default prompts that are internal to llama-index\n",
- "query_wrapper_prompt = SimpleInputPrompt(\"<|USER|>{query_str}<|ASSISTANT|>\")"
+ "query_wrapper_prompt = PromptTemplate(\"<|USER|>{query_str}<|ASSISTANT|>\")"
]
},
{
diff --git a/docs/examples/customization/prompts/chat_prompts.ipynb b/docs/examples/customization/prompts/chat_prompts.ipynb
index eae5407cc8..1ff7f444ce 100644
--- a/docs/examples/customization/prompts/chat_prompts.ipynb
+++ b/docs/examples/customization/prompts/chat_prompts.ipynb
@@ -20,54 +20,55 @@
},
{
"cell_type": "code",
- "execution_count": 2,
+ "execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
- "from langchain.prompts.chat import (\n",
- " ChatPromptTemplate,\n",
- " HumanMessagePromptTemplate,\n",
- " SystemMessagePromptTemplate,\n",
- ")\n",
- "from llama_index.prompts import Prompt\n",
+ "from llama_index.llms import ChatMessage, MessageRole\n",
+ "from llama_index.prompts import ChatPromptTemplate\n",
"\n",
+ "# Text QA Prompt\n",
"chat_text_qa_msgs = [\n",
- " SystemMessagePromptTemplate.from_template(\n",
- " \"Always answer the question, even if the context isn't helpful.\"\n",
+ " ChatMessage(\n",
+ " role=MessageRole.SYSTEM,\n",
+ " content=\"Always answer the question, even if the context isn't helpful.\",\n",
" ),\n",
- " HumanMessagePromptTemplate.from_template(\n",
- " \"Context information is below.\\n\"\n",
- " \"---------------------\\n\"\n",
- " \"{context_str}\\n\"\n",
- " \"---------------------\\n\"\n",
- " \"Given the context information and not prior knowledge, \"\n",
- " \"answer the question: {query_str}\\n\"\n",
+ " ChatMessage(\n",
+ " role=MessageRole.USER,\n",
+ " content=(\n",
+ " \"Context information is below.\\n\"\n",
+ " \"---------------------\\n\"\n",
+ " \"{context_str}\\n\"\n",
+ " \"---------------------\\n\"\n",
+ " \"Given the context information and not prior knowledge, \"\n",
+ " \"answer the question: {query_str}\\n\"\n",
+ " ),\n",
" ),\n",
"]\n",
- "chat_text_qa_msgs_lc = ChatPromptTemplate.from_messages(chat_text_qa_msgs)\n",
- "text_qa_template = Prompt.from_langchain_prompt(chat_text_qa_msgs_lc)\n",
+ "text_qa_template = ChatPromptTemplate(chat_text_qa_msgs)\n",
"\n",
"# Refine Prompt\n",
"chat_refine_msgs = [\n",
- " SystemMessagePromptTemplate.from_template(\n",
- " \"Always answer the question, even if the context isn't helpful.\"\n",
+ " ChatMessage(\n",
+ " role=MessageRole.SYSTEM,\n",
+ " content=\"Always answer the question, even if the context isn't helpful.\",\n",
" ),\n",
- " HumanMessagePromptTemplate.from_template(\n",
- " \"We have the opportunity to refine the original answer \"\n",
- " \"(only if needed) with some more context below.\\n\"\n",
- " \"------------\\n\"\n",
- " \"{context_msg}\\n\"\n",
- " \"------------\\n\"\n",
- " \"Given the new context, refine the original answer to better \"\n",
- " \"answer the question: {query_str}. \"\n",
- " \"If the context isn't useful, output the original answer again.\\n\"\n",
- " \"Original Answer: {existing_answer}\"\n",
+ " ChatMessage(\n",
+ " role=MessageRole.USER,\n",
+ " content=(\n",
+ " \"We have the opportunity to refine the original answer \"\n",
+ " \"(only if needed) with some more context below.\\n\"\n",
+ " \"------------\\n\"\n",
+ " \"{context_msg}\\n\"\n",
+ " \"------------\\n\"\n",
+ " \"Given the new context, refine the original answer to better \"\n",
+ " \"answer the question: {query_str}. \"\n",
+ " \"If the context isn't useful, output the original answer again.\\n\"\n",
+ " \"Original Answer: {existing_answer}\"\n",
+ " ),\n",
" ),\n",
"]\n",
- "\n",
- "\n",
- "chat_refine_msgs_lc = ChatPromptTemplate.from_messages(chat_refine_msgs)\n",
- "refine_template = Prompt.from_langchain_prompt(chat_refine_msgs_lc)"
+ "refine_template = ChatPromptTemplate(chat_refine_msgs)"
]
},
{
@@ -82,20 +83,20 @@
},
{
"cell_type": "code",
- "execution_count": 3,
+ "execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"import openai\n",
"import os\n",
"\n",
- "os.environ[\"OPENAI_API_KEY\"] = \"YOUR_API_KEY\"\n",
+ "os.environ[\"OPENAI_API_KEY\"] = \"sk-...\"\n",
"openai.api_key = os.environ[\"OPENAI_API_KEY\"]"
]
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
@@ -106,7 +107,7 @@
"\n",
"# Create an index using a chat model, so that we can use the chat prompts!\n",
"service_context = ServiceContext.from_defaults(\n",
- " llm=OpenAI(model=\"gpt-3.5-turbo\", temperature=0.0)\n",
+ " llm=OpenAI(model=\"gpt-3.5-turbo\", temperature=0.1)\n",
")\n",
"\n",
"index = VectorStoreIndex.from_documents(documents, service_context=service_context)"
@@ -122,14 +123,14 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 10,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "Based on the given context information, there is no mention of Joe Biden. Therefore, it is not possible to determine who Joe Biden is based on this information alone.\n"
+ "I'm sorry, but the given context does not provide any information about Joe Biden.\n"
]
}
],
@@ -147,14 +148,14 @@
},
{
"cell_type": "code",
- "execution_count": 6,
+ "execution_count": 11,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "Joe Biden is a politician who served as the 46th President of the United States.\n"
+ "Joe Biden is the 46th President of the United States.\n"
]
}
],
@@ -176,9 +177,9 @@
],
"metadata": {
"kernelspec": {
- "display_name": "venv",
+ "display_name": "llama-index",
"language": "python",
- "name": "python3"
+ "name": "llama-index"
},
"language_info": {
"codemirror_mode": {
@@ -190,7 +191,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.6"
+ "version": "3.11.0"
},
"orig_nbformat": 4
},
diff --git a/docs/examples/customization/prompts/completion_prompts.ipynb b/docs/examples/customization/prompts/completion_prompts.ipynb
index c900a7184e..a4adf4cf22 100644
--- a/docs/examples/customization/prompts/completion_prompts.ipynb
+++ b/docs/examples/customization/prompts/completion_prompts.ipynb
@@ -20,11 +20,11 @@
},
{
"cell_type": "code",
- "execution_count": 28,
+ "execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
- "from llama_index.prompts import Prompt\n",
+ "from llama_index.prompts import PromptTemplate\n",
"\n",
"text_qa_template_str = (\n",
" \"Context information is below.\\n\"\n",
@@ -35,7 +35,7 @@
" \"answer the question: {query_str}\\n\"\n",
" \"If the context isn't helpful, you can also answer the question on your own.\\n\"\n",
")\n",
- "text_qa_template = Prompt(text_qa_template_str)\n",
+ "text_qa_template = PromptTemplate(text_qa_template_str)\n",
"\n",
"refine_template_str = (\n",
" \"The original question is as follows: {query_str}\\n\"\n",
@@ -47,7 +47,7 @@
" \"------------\\n\"\n",
" \"Using both the new context and your own knowledege, update or repeat the existing answer.\\n\"\n",
")\n",
- "refine_template = Prompt(refine_template_str)"
+ "refine_template = PromptTemplate(refine_template_str)"
]
},
{
@@ -62,28 +62,31 @@
},
{
"cell_type": "code",
- "execution_count": 29,
+ "execution_count": 21,
"metadata": {},
"outputs": [],
"source": [
"import openai\n",
"import os\n",
"\n",
- "os.environ[\"OPENAI_API_KEY\"] = \"YOU_API_KEY\"\n",
+ "os.environ[\"OPENAI_API_KEY\"] = \"sk-...\"\n",
"openai.api_key = os.environ[\"OPENAI_API_KEY\"]"
]
},
{
"cell_type": "code",
- "execution_count": 30,
+ "execution_count": 18,
"metadata": {},
"outputs": [],
"source": [
- "from llama_index import VectorStoreIndex, SimpleDirectoryReader\n",
+ "from llama_index import VectorStoreIndex, SimpleDirectoryReader, ServiceContext\n",
+ "from llama_index.llms import OpenAI\n",
+ "\n",
+ "service_context = ServiceContext.from_defaults(llm=OpenAI(model=\"text-davinci-003\"))\n",
"\n",
"documents = SimpleDirectoryReader(\"../../data/paul_graham/\").load_data()\n",
"\n",
- "index = VectorStoreIndex.from_documents(documents)"
+ "index = VectorStoreIndex.from_documents(documents, service_context=service_context)"
]
},
{
@@ -96,15 +99,14 @@
},
{
"cell_type": "code",
- "execution_count": 31,
+ "execution_count": 19,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "\n",
- "Joe Biden is not mentioned in the context information.\n"
+ " Joe Biden is not mentioned in the context information.\n"
]
}
],
@@ -122,7 +124,7 @@
},
{
"cell_type": "code",
- "execution_count": 33,
+ "execution_count": 20,
"metadata": {},
"outputs": [
{
@@ -152,9 +154,9 @@
],
"metadata": {
"kernelspec": {
- "display_name": "venv",
+ "display_name": "llama-index",
"language": "python",
- "name": "python3"
+ "name": "llama-index"
},
"language_info": {
"codemirror_mode": {
@@ -166,7 +168,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.6"
+ "version": "3.11.0"
},
"orig_nbformat": 4
},
diff --git a/docs/examples/output_parsing/GuardrailsDemo.ipynb b/docs/examples/output_parsing/GuardrailsDemo.ipynb
index 4edd43c4a5..b381852506 100644
--- a/docs/examples/output_parsing/GuardrailsDemo.ipynb
+++ b/docs/examples/output_parsing/GuardrailsDemo.ipynb
@@ -117,7 +117,7 @@
},
"outputs": [],
"source": [
- "from llama_index.prompts.prompts import QuestionAnswerPrompt, RefinePrompt\n",
+ "from llama_index.prompts import PromptTemplate\n",
"from llama_index.prompts.default_prompts import (\n",
" DEFAULT_TEXT_QA_PROMPT_TMPL,\n",
" DEFAULT_REFINE_PROMPT_TMPL,\n",
@@ -190,8 +190,8 @@
"fmt_qa_tmpl = output_parser.format(DEFAULT_TEXT_QA_PROMPT_TMPL)\n",
"fmt_refine_tmpl = output_parser.format(DEFAULT_REFINE_PROMPT_TMPL)\n",
"\n",
- "qa_prompt = QuestionAnswerPrompt(fmt_qa_tmpl, output_parser=output_parser)\n",
- "refine_prompt = RefinePrompt(fmt_refine_tmpl, output_parser=output_parser)"
+ "qa_prompt = PromptTemplate(fmt_qa_tmpl, output_parser=output_parser)\n",
+ "refine_prompt = PromptTemplate(fmt_refine_tmpl, output_parser=output_parser)"
]
},
{
diff --git a/docs/examples/output_parsing/LangchainOutputParserDemo.ipynb b/docs/examples/output_parsing/LangchainOutputParserDemo.ipynb
index 907a24a563..85fe9231f2 100644
--- a/docs/examples/output_parsing/LangchainOutputParserDemo.ipynb
+++ b/docs/examples/output_parsing/LangchainOutputParserDemo.ipynb
@@ -119,7 +119,7 @@
},
"outputs": [],
"source": [
- "from llama_index.prompts.prompts import QuestionAnswerPrompt, RefinePrompt\n",
+ "from llama_index.prompts import PromptTemplate\n",
"from llama_index.prompts.default_prompts import (\n",
" DEFAULT_TEXT_QA_PROMPT_TMPL,\n",
" DEFAULT_REFINE_PROMPT_TMPL,\n",
@@ -174,8 +174,8 @@
"fmt_qa_tmpl = output_parser.format(DEFAULT_TEXT_QA_PROMPT_TMPL)\n",
"fmt_refine_tmpl = output_parser.format(DEFAULT_REFINE_PROMPT_TMPL)\n",
"\n",
- "qa_prompt = QuestionAnswerPrompt(fmt_qa_tmpl, output_parser=output_parser)\n",
- "refine_prompt = RefinePrompt(fmt_refine_tmpl, output_parser=output_parser)"
+ "qa_prompt = PromptTemplate(fmt_qa_tmpl, output_parser=output_parser)\n",
+ "refine_prompt = PromptTemplate(fmt_refine_tmpl, output_parser=output_parser)"
]
},
{
diff --git a/docs/examples/query_engine/ensemble_query_engine.ipynb b/docs/examples/query_engine/ensemble_query_engine.ipynb
index 33e623fe22..7825e1a526 100644
--- a/docs/examples/query_engine/ensemble_query_engine.ipynb
+++ b/docs/examples/query_engine/ensemble_query_engine.ipynb
@@ -208,7 +208,7 @@
},
"outputs": [],
"source": [
- "from llama_index import Prompt\n",
+ "from llama_index.prompts import PromptTemplate\n",
"\n",
"QA_PROMPT_TMPL = (\n",
" \"Context information is below.\\n\"\n",
@@ -223,7 +223,7 @@
" \"Question: {query_str}\\n\"\n",
" \"Answer (including relevance score): \"\n",
")\n",
- "QA_PROMPT = Prompt(QA_PROMPT_TMPL)\n",
+ "QA_PROMPT = PromptTemplate(QA_PROMPT_TMPL)\n",
"\n",
"keyword_query_engine = keyword_index.as_query_engine(text_qa_template=QA_PROMPT)\n",
"vector_query_engine = vector_index.as_query_engine(text_qa_template=QA_PROMPT)"
@@ -368,7 +368,9 @@
" \"Answer: \"\n",
")\n",
"\n",
- "tree_summarize = TreeSummarize(summary_template=Prompt(TREE_SUMMARIZE_PROMPT_TMPL))\n",
+ "tree_summarize = TreeSummarize(\n",
+ " summary_template=PromptTemplate(TREE_SUMMARIZE_PROMPT_TMPL)\n",
+ ")\n",
"\n",
"query_engine = RouterQueryEngine(\n",
" selector=LLMMultiSelector.from_defaults(),\n",
diff --git a/docs/examples/vector_stores/SimpleIndexDemoLlama-Local.ipynb b/docs/examples/vector_stores/SimpleIndexDemoLlama-Local.ipynb
index 3fe80977ed..060d3bef0e 100644
--- a/docs/examples/vector_stores/SimpleIndexDemoLlama-Local.ipynb
+++ b/docs/examples/vector_stores/SimpleIndexDemoLlama-Local.ipynb
@@ -177,7 +177,7 @@
"source": [
"import torch\n",
"from llama_index.llms import HuggingFaceLLM\n",
- "from llama_index.prompts.prompts import SimpleInputPrompt\n",
+ "from llama_index.prompts import PromptTemplate\n",
"\n",
"# Model names (make sure you have access on HF)\n",
"LLAMA2_7B = \"meta-llama/Llama-2-7b-hf\"\n",
@@ -197,7 +197,7 @@
"- Never generate offensive or foul language.\n",
"\"\"\"\n",
"\n",
- "query_wrapper_prompt = SimpleInputPrompt(\n",
+ "query_wrapper_prompt = PromptTemplate(\n",
" \"[INST]<<SYS>>\\n\" + SYSTEM_PROMPT + \"<</SYS>>\\n\\n{query_str}[/INST]\"\n",
")\n",
"\n",
diff --git a/examples/paul_graham_essay/TestEssay.ipynb b/examples/paul_graham_essay/TestEssay.ipynb
index 4a1a20116c..404778d020 100644
--- a/examples/paul_graham_essay/TestEssay.ipynb
+++ b/examples/paul_graham_essay/TestEssay.ipynb
@@ -292,7 +292,7 @@
"metadata": {},
"outputs": [],
"source": [
- "from llama_index import Prompt"
+ "from llama_index.prompts import PromptTemplate"
]
},
{
@@ -328,7 +328,7 @@
" \"Given the context information and not prior knowledge, \"\n",
" f\"answer the question: {query_str}\\n\"\n",
")\n",
- "SUMMARY_PROMPT = Prompt(SUMMARY_PROMPT_TMPL)\n",
+ "SUMMARY_PROMPT = PromptTemplate(SUMMARY_PROMPT_TMPL)\n",
"index_with_query = TreeIndex.from_documents(documents, summary_template=SUMMARY_PROMPT)"
]
},
diff --git a/examples/test_wiki/TestNYC-Benchmark-GPT4.ipynb b/examples/test_wiki/TestNYC-Benchmark-GPT4.ipynb
index 01dd4e954e..25bfc407c2 100644
--- a/examples/test_wiki/TestNYC-Benchmark-GPT4.ipynb
+++ b/examples/test_wiki/TestNYC-Benchmark-GPT4.ipynb
@@ -55,7 +55,7 @@
" LLMPredictor,\n",
" VectorStoreIndex,\n",
" ListIndex,\n",
- " Prompt,\n",
+ " PromptTemplate,\n",
" ServiceContext,\n",
")\n",
"from llama_index.indices.base import BaseIndex\n",
@@ -324,7 +324,7 @@
" \"and explain why.\"\n",
")\n",
"\n",
- "DEFAULT_EVAL_PROMPT = Prompt(EVAL_PROMPT_TMPL)"
+ "DEFAULT_EVAL_PROMPT = PromptTemplate(EVAL_PROMPT_TMPL)"
]
},
{
diff --git a/experimental/classifier/utils.py b/experimental/classifier/utils.py
index ece63a75db..2f65448921 100644
--- a/experimental/classifier/utils.py
+++ b/experimental/classifier/utils.py
@@ -9,7 +9,7 @@ from sklearn.model_selection import train_test_split
from llama_index.indices.utils import extract_numbers_given_response
from llama_index.llm_predictor import LLMPredictor
-from llama_index.prompts.base import Prompt
+from llama_index.prompts import BasePromptTemplate, PromptTemplate
def get_train_and_eval_data(
@@ -74,7 +74,7 @@ def extract_float_given_response(response: str, n: int = 1) -> Optional[float]:
def get_eval_preds(
- train_prompt: Prompt, train_str: str, eval_df: pd.DataFrame, n: int = 20
+ train_prompt: BasePromptTemplate, train_str: str, eval_df: pd.DataFrame, n: int = 20
) -> List:
"""Get eval preds."""
llm_predictor = LLMPredictor()
@@ -111,9 +111,7 @@ train_prompt_str = (
"Survived: "
)
-train_prompt = Prompt(
- input_variables=["train_str", "eval_str"], template=train_prompt_str
-)
+train_prompt = PromptTemplate(template=train_prompt_str)
# prompt to summarize the data
@@ -130,9 +128,7 @@ qa_data_str = (
"Given this, answer the question: {query_str}"
)
-qa_data_prompt = Prompt(
- input_variables=["context_str", "query_str"], template=qa_data_str
-)
+qa_data_prompt = PromptTemplate(template=qa_data_str)
# prompt to refine the answer
refine_str = (
@@ -151,10 +147,7 @@ refine_str = (
"answer the question. "
"If the context isn't useful, return the original answer."
)
-refine_prompt = Prompt(
- input_variables=["query_str", "existing_answer", "context_msg"],
- template=refine_str,
-)
+refine_prompt = PromptTemplate(template=refine_str)
# train prompt with refined context
@@ -174,6 +167,4 @@ train_prompt_with_context_str = (
"Survived: "
)
-train_prompt_with_context = Prompt(
- input_variables=["train_str", "eval_str"], template=train_prompt_with_context_str
-)
+train_prompt_with_context = PromptTemplate(template=train_prompt_with_context_str)
diff --git a/llama_index/__init__.py b/llama_index/__init__.py
index 27e426ff33..19ff6e0d89 100644
--- a/llama_index/__init__.py
+++ b/llama_index/__init__.py
@@ -76,7 +76,14 @@ from llama_index.langchain_helpers.memory_wrapper import GPTIndexMemory
from llama_index.langchain_helpers.sql_wrapper import SQLDatabase
# prompts
-from llama_index.prompts.base import Prompt
+from llama_index.prompts import (
+ BasePromptTemplate,
+ PromptTemplate,
+ ChatPromptTemplate,
+ SelectorPromptTemplate,
+ # backwards compatibility
+ Prompt,
+)
from llama_index.prompts.prompts import (
KeywordExtractPrompt,
QueryKeywordExtractPrompt,
@@ -168,6 +175,10 @@ __all__ = [
"GPTSQLStructStoreIndex",
"GPTDocumentSummaryIndex",
"Prompt",
+ "PromptTemplate",
+ "BasePromptTemplate",
+ "ChatPromptTemplate",
+ "SelectorPromptTemplate",
"LangchainEmbedding",
"OpenAIEmbedding",
"SummaryPrompt",
diff --git a/llama_index/agent/context_retriever_agent.py b/llama_index/agent/context_retriever_agent.py
index 0bc92662aa..7c59adc4ee 100644
--- a/llama_index/agent/context_retriever_agent.py
+++ b/llama_index/agent/context_retriever_agent.py
@@ -17,7 +17,7 @@ from llama_index.llms.base import LLM, ChatMessage
from llama_index.llms.openai import OpenAI
from llama_index.llms.openai_utils import is_function_calling_model
from llama_index.memory import BaseMemory, ChatMemoryBuffer
-from llama_index.prompts.prompts import QuestionAnswerPrompt
+from llama_index.prompts import PromptTemplate
from llama_index.schema import NodeWithScore
from llama_index.tools import BaseTool
@@ -30,7 +30,7 @@ DEFAULT_QA_PROMPT_TMPL = (
"Given the context information and not prior knowledge, "
"either pick the corresponding tool or answer the function: {query_str}\n"
)
-DEFAULT_QA_PROMPT = QuestionAnswerPrompt(DEFAULT_QA_PROMPT_TMPL)
+DEFAULT_QA_PROMPT = PromptTemplate(DEFAULT_QA_PROMPT_TMPL)
class ContextRetrieverOpenAIAgent(BaseOpenAIAgent):
@@ -44,7 +44,7 @@ class ContextRetrieverOpenAIAgent(BaseOpenAIAgent):
Args:
tools (List[BaseTool]): A list of tools.
retriever (BaseRetriever): A retriever.
- qa_prompt (Optional[QuestionAnswerPrompt]): A QA prompt.
+ qa_prompt (Optional[PromptTemplate]): A QA prompt.
context_separator (str): A context separator.
llm (Optional[OpenAI]): An OpenAI LLM.
chat_history (Optional[List[ChatMessage]]): A chat history.
@@ -59,7 +59,7 @@ class ContextRetrieverOpenAIAgent(BaseOpenAIAgent):
self,
tools: List[BaseTool],
retriever: BaseRetriever,
- qa_prompt: QuestionAnswerPrompt,
+ qa_prompt: PromptTemplate,
context_separator: str,
llm: OpenAI,
memory: BaseMemory,
@@ -86,7 +86,7 @@ class ContextRetrieverOpenAIAgent(BaseOpenAIAgent):
cls,
tools: List[BaseTool],
retriever: BaseRetriever,
- qa_prompt: Optional[QuestionAnswerPrompt] = None,
+ qa_prompt: Optional[PromptTemplate] = None,
context_separator: str = "\n",
llm: Optional[LLM] = None,
chat_history: Optional[List[ChatMessage]] = None,
@@ -102,7 +102,7 @@ class ContextRetrieverOpenAIAgent(BaseOpenAIAgent):
Args:
retriever (BaseRetriever): A retriever.
- qa_prompt (Optional[QuestionAnswerPrompt]): A QA prompt.
+ qa_prompt (Optional[PromptTemplate]): A QA prompt.
context_separator (str): A context separator.
llm (Optional[OpenAI]): An OpenAI LLM.
chat_history (Optional[ChatMessageHistory]): A chat history.
diff --git a/llama_index/chat_engine/condense_question.py b/llama_index/chat_engine/condense_question.py
index e00cf0a9f9..79f5cd88c7 100644
--- a/llama_index/chat_engine/condense_question.py
+++ b/llama_index/chat_engine/condense_question.py
@@ -2,6 +2,7 @@ import logging
from threading import Thread
from typing import Any, List, Optional, Type
+from llama_index.callbacks import CallbackManager, trace_method
from llama_index.chat_engine.types import (
AgentChatResponse,
BaseChatEngine,
@@ -13,10 +14,9 @@ from llama_index.indices.service_context import ServiceContext
from llama_index.llms.base import ChatMessage, MessageRole
from llama_index.llms.generic_utils import messages_to_history_str
from llama_index.memory import BaseMemory, ChatMemoryBuffer
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import BasePromptTemplate, PromptTemplate
from llama_index.response.schema import RESPONSE_TYPE, StreamingResponse
from llama_index.tools import ToolOutput
-from llama_index.callbacks import CallbackManager, trace_method
logger = logging.getLogger(__name__)
@@ -35,7 +35,7 @@ from the conversation.
<Standalone question>
"""
-DEFAULT_PROMPT = Prompt(DEFAULT_TEMPLATE)
+DEFAULT_PROMPT = PromptTemplate(DEFAULT_TEMPLATE)
class CondenseQuestionChatEngine(BaseChatEngine):
@@ -48,7 +48,7 @@ class CondenseQuestionChatEngine(BaseChatEngine):
def __init__(
self,
query_engine: BaseQueryEngine,
- condense_question_prompt: Prompt,
+ condense_question_prompt: BasePromptTemplate,
memory: BaseMemory,
service_context: ServiceContext,
verbose: bool = False,
@@ -65,7 +65,7 @@ class CondenseQuestionChatEngine(BaseChatEngine):
def from_defaults(
cls,
query_engine: BaseQueryEngine,
- condense_question_prompt: Optional[Prompt] = None,
+ condense_question_prompt: Optional[BasePromptTemplate] = None,
chat_history: Optional[List[ChatMessage]] = None,
memory: Optional[BaseMemory] = None,
memory_cls: Type[BaseMemory] = ChatMemoryBuffer,
diff --git a/llama_index/evaluation/base.py b/llama_index/evaluation/base.py
index 85d0378cdd..047fec057b 100644
--- a/llama_index/evaluation/base.py
+++ b/llama_index/evaluation/base.py
@@ -7,7 +7,7 @@ from typing import List, Optional
from llama_index.indices.base import ServiceContext
from llama_index.indices.list.base import ListIndex
-from llama_index.prompts.prompts import QuestionAnswerPrompt, RefinePrompt
+from llama_index.prompts import PromptTemplate
from llama_index.schema import Document
from llama_index.response.schema import Response
@@ -155,8 +155,8 @@ class ResponseEvaluator:
index = ListIndex.from_documents(context, service_context=self.service_context)
response_txt = ""
- EVAL_PROMPT_TMPL = QuestionAnswerPrompt(DEFAULT_EVAL_PROMPT)
- REFINE_PROMPT_TMPL = RefinePrompt(DEFAULT_REFINE_PROMPT)
+ EVAL_PROMPT_TMPL = PromptTemplate(DEFAULT_EVAL_PROMPT)
+ REFINE_PROMPT_TMPL = PromptTemplate(DEFAULT_REFINE_PROMPT)
query_engine = index.as_query_engine(
text_qa_template=EVAL_PROMPT_TMPL,
@@ -199,8 +199,8 @@ class ResponseEvaluator:
)
response_txt = ""
- EVAL_PROMPT_TMPL = QuestionAnswerPrompt(DEFAULT_EVAL_PROMPT)
- REFINE_PROMPT_TMPL = RefinePrompt(DEFAULT_REFINE_PROMPT)
+ EVAL_PROMPT_TMPL = PromptTemplate(DEFAULT_EVAL_PROMPT)
+ REFINE_PROMPT_TMPL = PromptTemplate(DEFAULT_REFINE_PROMPT)
query_engine = index.as_query_engine(
text_qa_template=EVAL_PROMPT_TMPL,
@@ -285,10 +285,8 @@ class QueryResponseEvaluator(BaseEvaluator):
context = self.get_context(response)
index = ListIndex.from_documents(context, service_context=self.service_context)
- QUERY_RESPONSE_EVAL_PROMPT_TMPL = QuestionAnswerPrompt(
- QUERY_RESPONSE_EVAL_PROMPT
- )
- QUERY_RESPONSE_REFINE_PROMPT_TMPL = RefinePrompt(QUERY_RESPONSE_REFINE_PROMPT)
+ QUERY_RESPONSE_EVAL_PROMPT_TMPL = PromptTemplate(QUERY_RESPONSE_EVAL_PROMPT)
+ QUERY_RESPONSE_REFINE_PROMPT_TMPL = PromptTemplate(QUERY_RESPONSE_REFINE_PROMPT)
query_response = f"Question: {query}\nResponse: {answer}"
@@ -337,10 +335,8 @@ class QueryResponseEvaluator(BaseEvaluator):
)
response_txt = ""
- QUERY_RESPONSE_EVAL_PROMPT_TMPL = QuestionAnswerPrompt(
- QUERY_RESPONSE_EVAL_PROMPT
- )
- QUERY_RESPONSE_REFINE_PROMPT_TMPL = RefinePrompt(
+ QUERY_RESPONSE_EVAL_PROMPT_TMPL = PromptTemplate(QUERY_RESPONSE_EVAL_PROMPT)
+ QUERY_RESPONSE_REFINE_PROMPT_TMPL = PromptTemplate(
QUERY_RESPONSE_REFINE_PROMPT
)
diff --git a/llama_index/evaluation/dataset_generation.py b/llama_index/evaluation/dataset_generation.py
index 566ca81490..195714d484 100644
--- a/llama_index/evaluation/dataset_generation.py
+++ b/llama_index/evaluation/dataset_generation.py
@@ -8,10 +8,10 @@ from typing import List, Optional
from llama_index import (
Document,
ListIndex,
- QuestionAnswerPrompt,
ServiceContext,
)
from llama_index.llms.openai import OpenAI
+from llama_index.prompts.base import BasePromptTemplate, PromptTemplate
from llama_index.schema import BaseNode, NodeWithScore, MetadataMode
from llama_index.indices.postprocessor.node import KeywordNodePostprocessor
@@ -49,7 +49,7 @@ class DatasetGenerator:
nodes: List[BaseNode],
service_context: Optional[ServiceContext] = None,
num_questions_per_chunk: int = 10,
- text_question_template: Optional[QuestionAnswerPrompt] = None,
+ text_question_template: Optional[BasePromptTemplate] = None,
question_gen_query: Optional[str] = None,
required_keywords: Optional[List[str]] = None,
exclude_keywords: Optional[List[str]] = None,
@@ -58,7 +58,7 @@ class DatasetGenerator:
if service_context is None:
service_context = _get_default_service_context()
self.service_context = service_context
- self.text_question_template = text_question_template or QuestionAnswerPrompt(
+ self.text_question_template = text_question_template or PromptTemplate(
DEFAULT_QUESTION_GENERATION_PROMPT
)
self.question_gen_query = (
@@ -77,7 +77,7 @@ class DatasetGenerator:
documents: List[Document],
service_context: Optional[ServiceContext] = None,
num_questions_per_chunk: int = 10,
- text_question_template: Optional[QuestionAnswerPrompt] = None,
+ text_question_template: Optional[BasePromptTemplate] = None,
question_gen_query: Optional[str] = None,
required_keywords: Optional[List[str]] = None,
exclude_keywords: Optional[List[str]] = None,
diff --git a/llama_index/evaluation/guideline_eval.py b/llama_index/evaluation/guideline_eval.py
index 8142efdc94..0e378133b8 100644
--- a/llama_index/evaluation/guideline_eval.py
+++ b/llama_index/evaluation/guideline_eval.py
@@ -1,12 +1,12 @@
import logging
from typing import Optional
-from llama_index.bridge.langchain import PydanticOutputParser
from pydantic import BaseModel, Field
+from llama_index.bridge.langchain import PydanticOutputParser
from llama_index.evaluation.base import BaseEvaluator, Evaluation
from llama_index.indices.base import ServiceContext
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import PromptTemplate
from llama_index.response.schema import Response
logger = logging.getLogger(__name__)
@@ -38,7 +38,7 @@ class GuidelineEvaluator(BaseEvaluator):
)
format_instructions = parser.get_format_instructions()
response_str = response.response
- prompt = Prompt(self.eval_template)
+ prompt = PromptTemplate(self.eval_template)
logger.debug("prompt: %s", prompt)
logger.debug("query: %s", query)
logger.debug("response: %s", response_str)
diff --git a/llama_index/indices/common/struct_store/base.py b/llama_index/indices/common/struct_store/base.py
index beff18c034..9afa5b187f 100644
--- a/llama_index/indices/common/struct_store/base.py
+++ b/llama_index/indices/common/struct_store/base.py
@@ -9,6 +9,7 @@ from llama_index.data_structs.table import StructDatapoint
from llama_index.indices.service_context import ServiceContext
from llama_index.langchain_helpers.sql_wrapper import SQLDatabase
from llama_index.llm_predictor.base import BaseLLMPredictor
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompt_selectors import (
DEFAULT_REFINE_TABLE_CONTEXT_PROMPT_SEL,
)
@@ -17,13 +18,6 @@ from llama_index.prompts.default_prompts import (
DEFAULT_TABLE_CONTEXT_QUERY,
)
from llama_index.prompts.prompt_type import PromptType
-from llama_index.prompts.prompts import (
- QuestionAnswerPrompt,
- RefinePrompt,
- RefineTableContextPrompt,
- SchemaExtractPrompt,
- TableContextPrompt,
-)
from llama_index.response_synthesizers import get_response_synthesizer
from llama_index.schema import BaseNode, MetadataMode
from llama_index.text_splitter import TextSplitter
@@ -40,9 +34,9 @@ class SQLDocumentContextBuilder:
llm_predictor (Optional[BaseLLMPredictor]): LLM Predictor to use.
prompt_helper (Optional[PromptHelper]): Prompt Helper to use.
text_splitter (Optional[TextSplitter]): Text Splitter to use.
- table_context_prompt (Optional[TableContextPrompt]): A
+ table_context_prompt (Optional[BasePromptTemplate]): A
Table Context Prompt (see :ref:`Prompt-Templates`).
- refine_table_context_prompt (Optional[RefineTableContextPrompt]):
+ refine_table_context_prompt (Optional[BasePromptTemplate]):
A Refine Table Context Prompt (see :ref:`Prompt-Templates`).
table_context_task (Optional[str]): The query to perform
on the table context. A default query string is used
@@ -54,8 +48,8 @@ class SQLDocumentContextBuilder:
sql_database: SQLDatabase,
service_context: Optional[ServiceContext] = None,
text_splitter: Optional[TextSplitter] = None,
- table_context_prompt: Optional[TableContextPrompt] = None,
- refine_table_context_prompt: Optional[RefineTableContextPrompt] = None,
+ table_context_prompt: Optional[BasePromptTemplate] = None,
+ refine_table_context_prompt: Optional[BasePromptTemplate] = None,
table_context_task: Optional[str] = None,
) -> None:
"""Initialize params."""
@@ -92,14 +86,13 @@ class SQLDocumentContextBuilder:
) -> str:
"""Build context from documents for a single table."""
schema = self._sql_database.get_single_table_info(table_name)
- prompt_with_schema = QuestionAnswerPrompt.from_prompt(
- self._table_context_prompt.partial_format(schema=schema),
- prompt_type=PromptType.QUESTION_ANSWER,
- )
- refine_prompt_with_schema = RefinePrompt.from_prompt(
- self._refine_table_context_prompt.partial_format(schema=schema),
- prompt_type=PromptType.REFINE,
+ prompt_with_schema = self._table_context_prompt.partial_format(schema=schema)
+ prompt_with_schema.metadata["prompt_type"] = PromptType.QUESTION_ANSWER
+ refine_prompt_with_schema = self._refine_table_context_prompt.partial_format(
+ schema=schema
)
+ refine_prompt_with_schema.metadata["prompt_type"] = PromptType.REFINE
+
text_splitter = (
self._text_splitter
or self._service_context.prompt_helper.get_text_splitter_given_prompt(
@@ -143,7 +136,7 @@ class BaseStructDatapointExtractor:
def __init__(
self,
llm_predictor: BaseLLMPredictor,
- schema_extract_prompt: SchemaExtractPrompt,
+ schema_extract_prompt: BasePromptTemplate,
output_parser: OUTPUT_PARSER_TYPE,
) -> None:
"""Initialize params."""
diff --git a/llama_index/indices/common/struct_store/sql.py b/llama_index/indices/common/struct_store/sql.py
index a7524d231e..5ec501fcd5 100644
--- a/llama_index/indices/common/struct_store/sql.py
+++ b/llama_index/indices/common/struct_store/sql.py
@@ -11,7 +11,7 @@ from llama_index.indices.common.struct_store.base import (
)
from llama_index.langchain_helpers.sql_wrapper import SQLDatabase
from llama_index.llm_predictor.base import BaseLLMPredictor
-from llama_index.prompts.prompts import SchemaExtractPrompt
+from llama_index.prompts import BasePromptTemplate
class SQLStructDatapointExtractor(BaseStructDatapointExtractor):
@@ -20,7 +20,7 @@ class SQLStructDatapointExtractor(BaseStructDatapointExtractor):
def __init__(
self,
llm_predictor: BaseLLMPredictor,
- schema_extract_prompt: SchemaExtractPrompt,
+ schema_extract_prompt: BasePromptTemplate,
output_parser: OUTPUT_PARSER_TYPE,
sql_database: SQLDatabase,
table_name: Optional[str] = None,
diff --git a/llama_index/indices/common_tree/base.py b/llama_index/indices/common_tree/base.py
index 454327ecd5..e59a27673b 100644
--- a/llama_index/indices/common_tree/base.py
+++ b/llama_index/indices/common_tree/base.py
@@ -8,14 +8,13 @@ from typing import Dict, List, Optional, Sequence, Tuple
from llama_index.async_utils import run_async_tasks
from llama_index.callbacks.schema import CBEventType, EventPayload
from llama_index.data_structs.data_structs import IndexGraph
-from llama_index.schema import BaseNode, TextNode
-from llama_index.storage.docstore import BaseDocumentStore
-from llama_index.storage.docstore.registry import get_default_docstore
from llama_index.indices.service_context import ServiceContext
from llama_index.indices.utils import get_sorted_node_list, truncate_text
-from llama_index.prompts.prompts import SummaryPrompt
+from llama_index.prompts import BasePromptTemplate
+from llama_index.schema import BaseNode, MetadataMode, TextNode
+from llama_index.storage.docstore import BaseDocumentStore
+from llama_index.storage.docstore.registry import get_default_docstore
from llama_index.utils import get_tqdm_iterable
-from llama_index.schema import MetadataMode
logger = logging.getLogger(__name__)
@@ -31,7 +30,7 @@ class GPTTreeIndexBuilder:
def __init__(
self,
num_children: int,
- summary_prompt: SummaryPrompt,
+ summary_prompt: BasePromptTemplate,
service_context: ServiceContext,
docstore: Optional[BaseDocumentStore] = None,
show_progress: bool = False,
diff --git a/llama_index/indices/document_summary/base.py b/llama_index/indices/document_summary/base.py
index cd6d973b06..d6db59ee50 100644
--- a/llama_index/indices/document_summary/base.py
+++ b/llama_index/indices/document_summary/base.py
@@ -9,7 +9,6 @@ import logging
from collections import defaultdict
from enum import Enum
from typing import Any, Dict, Optional, Sequence, Union, cast
-from llama_index.utils import get_tqdm_iterable
from llama_index.data_structs.document_summary import IndexDocumentSummary
from llama_index.indices.base import BaseIndex
@@ -18,17 +17,18 @@ from llama_index.indices.service_context import ServiceContext
from llama_index.response.schema import Response
from llama_index.response_synthesizers import (
BaseSynthesizer,
- get_response_synthesizer,
ResponseMode,
+ get_response_synthesizer,
)
from llama_index.schema import (
BaseNode,
- NodeWithScore,
NodeRelationship,
+ NodeWithScore,
RelatedNodeInfo,
TextNode,
)
from llama_index.storage.docstore.types import RefDocInfo
+from llama_index.utils import get_tqdm_iterable
logger = logging.getLogger(__name__)
@@ -51,7 +51,7 @@ class DocumentSummaryIndex(BaseIndex[IndexDocumentSummary]):
"""Document Summary Index.
Args:
- summary_template (Optional[SummaryPrompt]): A Summary Prompt
+ summary_template (Optional[BasePromptTemplate]): A Summary Prompt
(see :ref:`Prompt-Templates`).
show_progress (bool): Whether to show tqdm progress bars. Defaults to False.
diff --git a/llama_index/indices/document_summary/retrievers.py b/llama_index/indices/document_summary/retrievers.py
index c0e4a21b65..8e729e901f 100644
--- a/llama_index/indices/document_summary/retrievers.py
+++ b/llama_index/indices/document_summary/retrievers.py
@@ -16,9 +16,9 @@ from llama_index.indices.utils import (
default_format_node_batch_fn,
default_parse_choice_select_answer_fn,
)
-from llama_index.prompts.choice_select import (
+from llama_index.prompts import PromptTemplate
+from llama_index.prompts.default_prompts import (
DEFAULT_CHOICE_SELECT_PROMPT,
- ChoiceSelectPrompt,
)
from llama_index.schema import BaseNode, MetadataMode, NodeWithScore
@@ -38,7 +38,7 @@ class DocumentSummaryIndexRetriever(BaseRetriever):
def __init__(
self,
index: DocumentSummaryIndex,
- choice_select_prompt: Optional[ChoiceSelectPrompt] = None,
+ choice_select_prompt: Optional[PromptTemplate] = None,
choice_batch_size: int = 10,
format_node_batch_fn: Optional[Callable] = None,
parse_choice_select_answer_fn: Optional[Callable] = None,
diff --git a/llama_index/indices/empty/retrievers.py b/llama_index/indices/empty/retrievers.py
index 5c2cb66525..f4c09a7631 100644
--- a/llama_index/indices/empty/retrievers.py
+++ b/llama_index/indices/empty/retrievers.py
@@ -4,8 +4,8 @@ from typing import Any, List, Optional
from llama_index.indices.base_retriever import BaseRetriever
from llama_index.indices.empty.base import EmptyIndex
from llama_index.indices.query.schema import QueryBundle
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompts import DEFAULT_SIMPLE_INPUT_PROMPT
-from llama_index.prompts.prompts import SimpleInputPrompt
from llama_index.schema import NodeWithScore
@@ -15,7 +15,7 @@ class EmptyIndexRetriever(BaseRetriever):
Passes the raw LLM call to the underlying LLM model.
Args:
- input_prompt (Optional[SimpleInputPrompt]): A Simple Input Prompt
+ input_prompt (Optional[BasePromptTemplate]): A Simple Input Prompt
(see :ref:`Prompt-Templates`).
"""
@@ -23,7 +23,7 @@ class EmptyIndexRetriever(BaseRetriever):
def __init__(
self,
index: EmptyIndex,
- input_prompt: Optional[SimpleInputPrompt] = None,
+ input_prompt: Optional[BasePromptTemplate] = None,
**kwargs: Any,
) -> None:
"""Initialize params."""
diff --git a/llama_index/indices/keyword_table/base.py b/llama_index/indices/keyword_table/base.py
index f8ec8a9d10..2655b69c0d 100644
--- a/llama_index/indices/keyword_table/base.py
+++ b/llama_index/indices/keyword_table/base.py
@@ -11,7 +11,6 @@ existing keywords in the table.
from abc import abstractmethod
from enum import Enum
from typing import Any, Dict, Optional, Sequence, Set, Union
-from llama_index.utils import get_tqdm_iterable
from llama_index.async_utils import run_async_tasks
from llama_index.data_structs.data_structs import KeywordTable
@@ -19,13 +18,14 @@ from llama_index.indices.base import BaseIndex
from llama_index.indices.base_retriever import BaseRetriever
from llama_index.indices.keyword_table.utils import extract_keywords_given_response
from llama_index.indices.service_context import ServiceContext
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompts import (
DEFAULT_KEYWORD_EXTRACT_TEMPLATE,
DEFAULT_QUERY_KEYWORD_EXTRACT_TEMPLATE,
)
-from llama_index.prompts.prompts import KeywordExtractPrompt
from llama_index.schema import BaseNode, MetadataMode
from llama_index.storage.docstore.types import RefDocInfo
+from llama_index.utils import get_tqdm_iterable
DQKET = DEFAULT_QUERY_KEYWORD_EXTRACT_TEMPLATE
@@ -49,7 +49,7 @@ class BaseKeywordTableIndex(BaseIndex[KeywordTable]):
are then used to answer the query.
Args:
- keyword_extract_template (Optional[KeywordExtractPrompt]): A Keyword
+ keyword_extract_template (Optional[BasePromptTemplate]): A Keyword
Extraction Prompt
(see :ref:`Prompt-Templates`).
use_async (bool): Whether to use asynchronous calls. Defaults to False.
@@ -64,7 +64,7 @@ class BaseKeywordTableIndex(BaseIndex[KeywordTable]):
nodes: Optional[Sequence[BaseNode]] = None,
index_struct: Optional[KeywordTable] = None,
service_context: Optional[ServiceContext] = None,
- keyword_extract_template: Optional[KeywordExtractPrompt] = None,
+ keyword_extract_template: Optional[BasePromptTemplate] = None,
max_keywords_per_chunk: int = 10,
use_async: bool = False,
show_progress: bool = False,
diff --git a/llama_index/indices/keyword_table/retrievers.py b/llama_index/indices/keyword_table/retrievers.py
index 07a7ea2c9e..2c03664b52 100644
--- a/llama_index/indices/keyword_table/retrievers.py
+++ b/llama_index/indices/keyword_table/retrievers.py
@@ -12,11 +12,11 @@ from llama_index.indices.keyword_table.utils import (
simple_extract_keywords,
)
from llama_index.indices.query.schema import QueryBundle
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompts import (
DEFAULT_KEYWORD_EXTRACT_TEMPLATE,
DEFAULT_QUERY_KEYWORD_EXTRACT_TEMPLATE,
)
-from llama_index.prompts.prompts import KeywordExtractPrompt, QueryKeywordExtractPrompt
from llama_index.schema import NodeWithScore
from llama_index.utils import truncate_text
@@ -31,15 +31,15 @@ class BaseKeywordTableRetriever(BaseRetriever):
Arguments are shared among subclasses.
Args:
- keyword_extract_template (Optional[KeywordExtractPrompt]): A Keyword
+ keyword_extract_template (Optional[BasePromptTemplate]): A Keyword
Extraction Prompt
(see :ref:`Prompt-Templates`).
- query_keyword_extract_template (Optional[QueryKeywordExtractPrompt]): A Query
+ query_keyword_extract_template (Optional[BasePromptTemplate]): A Query
Keyword Extraction
Prompt (see :ref:`Prompt-Templates`).
- refine_template (Optional[RefinePrompt]): A Refinement Prompt
+ refine_template (Optional[BasePromptTemplate]): A Refinement Prompt
(see :ref:`Prompt-Templates`).
- text_qa_template (Optional[QuestionAnswerPrompt]): A Question Answering Prompt
+ text_qa_template (Optional[BasePromptTemplate]): A Question Answering Prompt
(see :ref:`Prompt-Templates`).
max_keywords_per_query (int): Maximum number of keywords to extract from query.
num_chunks_per_query (int): Maximum number of text chunks to query.
@@ -49,8 +49,8 @@ class BaseKeywordTableRetriever(BaseRetriever):
def __init__(
self,
index: BaseKeywordTableIndex,
- keyword_extract_template: Optional[KeywordExtractPrompt] = None,
- query_keyword_extract_template: Optional[QueryKeywordExtractPrompt] = None,
+ keyword_extract_template: Optional[BasePromptTemplate] = None,
+ query_keyword_extract_template: Optional[BasePromptTemplate] = None,
max_keywords_per_query: int = 10,
num_chunks_per_query: int = 10,
**kwargs: Any,
diff --git a/llama_index/indices/knowledge_graph/base.py b/llama_index/indices/knowledge_graph/base.py
index e807b22b74..7e1a0fae5e 100644
--- a/llama_index/indices/knowledge_graph/base.py
+++ b/llama_index/indices/knowledge_graph/base.py
@@ -18,8 +18,8 @@ from llama_index.graph_stores.types import GraphStore
from llama_index.indices.base import BaseIndex
from llama_index.indices.base_retriever import BaseRetriever
from llama_index.indices.service_context import ServiceContext
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompts import DEFAULT_KG_TRIPLET_EXTRACT_PROMPT
-from llama_index.prompts.prompts import KnowledgeGraphPrompt
from llama_index.schema import BaseNode, MetadataMode
from llama_index.storage.docstore.types import RefDocInfo
from llama_index.storage.storage_context import StorageContext
@@ -34,7 +34,7 @@ class KnowledgeGraphIndex(BaseIndex[KG]):
Build a KG by extracting triplets, and leveraging the KG during query-time.
Args:
- kg_triple_extract_template (KnowledgeGraphPrompt): The prompt to use for
+ kg_triple_extract_template (BasePromptTemplate): The prompt to use for
extracting triplets.
max_triplets_per_chunk (int): The maximum number of triplets to extract.
service_context (Optional[ServiceContext]): The service context to use.
@@ -58,7 +58,7 @@ class KnowledgeGraphIndex(BaseIndex[KG]):
index_struct: Optional[KG] = None,
service_context: Optional[ServiceContext] = None,
storage_context: Optional[StorageContext] = None,
- kg_triple_extract_template: Optional[KnowledgeGraphPrompt] = None,
+ kg_triple_extract_template: Optional[BasePromptTemplate] = None,
max_triplets_per_chunk: int = 10,
include_embeddings: bool = False,
show_progress: bool = False,
diff --git a/llama_index/indices/knowledge_graph/retrievers.py b/llama_index/indices/knowledge_graph/retrievers.py
index a060d66222..fa8205ccc8 100644
--- a/llama_index/indices/knowledge_graph/retrievers.py
+++ b/llama_index/indices/knowledge_graph/retrievers.py
@@ -11,9 +11,8 @@ from llama_index.indices.knowledge_graph.base import KnowledgeGraphIndex
from llama_index.indices.query.embedding_utils import get_top_k_embeddings
from llama_index.indices.query.schema import QueryBundle
from llama_index.indices.service_context import ServiceContext
-from llama_index.prompts.base import Prompt, PromptType
+from llama_index.prompts import BasePromptTemplate, PromptTemplate, PromptType
from llama_index.prompts.default_prompts import DEFAULT_QUERY_KEYWORD_EXTRACT_TEMPLATE
-from llama_index.prompts.prompts import QueryKeywordExtractPrompt
from llama_index.schema import BaseNode, MetadataMode, NodeWithScore, TextNode
from llama_index.storage.storage_context import StorageContext
from llama_index.utils import truncate_text
@@ -53,9 +52,9 @@ class KGTableRetriever(BaseRetriever):
query_keyword_extract_template (Optional[QueryKGExtractPrompt]): A Query
KG Extraction
Prompt (see :ref:`Prompt-Templates`).
- refine_template (Optional[RefinePrompt]): A Refinement Prompt
+ refine_template (Optional[BasePromptTemplate]): A Refinement Prompt
(see :ref:`Prompt-Templates`).
- text_qa_template (Optional[QuestionAnswerPrompt]): A Question Answering Prompt
+ text_qa_template (Optional[BasePromptTemplate]): A Question Answering Prompt
(see :ref:`Prompt-Templates`).
max_keywords_per_query (int): Maximum number of keywords to extract from query.
num_chunks_per_query (int): Maximum number of text chunks to query.
@@ -77,7 +76,7 @@ class KGTableRetriever(BaseRetriever):
def __init__(
self,
index: KnowledgeGraphIndex,
- query_keyword_extract_template: Optional[QueryKeywordExtractPrompt] = None,
+ query_keyword_extract_template: Optional[BasePromptTemplate] = None,
max_keywords_per_query: int = 10,
num_chunks_per_query: int = 10,
include_text: bool = True,
@@ -328,7 +327,7 @@ KEYWORDS: {question}
----
"""
-DEFAULT_SYNONYM_EXPAND_PROMPT = Prompt(
+DEFAULT_SYNONYM_EXPAND_PROMPT = PromptTemplate(
DEFAULT_SYNONYM_EXPAND_TEMPLATE,
prompt_type=PromptType.QUERY_KEYWORD_EXTRACT,
)
@@ -344,7 +343,7 @@ class KnowledgeGraphRAGRetriever(BaseRetriever):
service_context (Optional[ServiceContext]): A service context to use.
storage_context (Optional[StorageContext]): A storage context to use.
entity_extract_fn (Optional[Callable]): A function to extract entities.
- entity_extract_template Optional[QueryKeywordExtractPrompt]): A Query Key Entity
+ entity_extract_template Optional[BasePromptTemplate]): A Query Key Entity
Extraction Prompt (see :ref:`Prompt-Templates`).
entity_extract_policy (Optional[str]): The entity extraction policy to use.
default: "union"
@@ -376,10 +375,10 @@ class KnowledgeGraphRAGRetriever(BaseRetriever):
service_context: Optional[ServiceContext] = None,
storage_context: Optional[StorageContext] = None,
entity_extract_fn: Optional[Callable] = None,
- entity_extract_template: Optional[QueryKeywordExtractPrompt] = None,
+ entity_extract_template: Optional[BasePromptTemplate] = None,
entity_extract_policy: Optional[str] = "union",
synonym_expand_fn: Optional[Callable] = None,
- synonym_expand_template: Optional[Prompt] = None,
+ synonym_expand_template: Optional[BasePromptTemplate] = None,
synonym_expand_policy: Optional[str] = "union",
max_entities: int = 5,
max_synonyms: int = 5,
@@ -451,7 +450,7 @@ class KnowledgeGraphRAGRetriever(BaseRetriever):
self,
query_str: str,
handle_fn: Optional[Callable],
- handle_llm_prompt_template: Optional[Prompt],
+ handle_llm_prompt_template: Optional[BasePromptTemplate],
cross_handle_policy: Optional[str] = "union",
max_items: Optional[int] = 5,
result_start_token: str = "KEYWORDS:",
@@ -501,7 +500,7 @@ class KnowledgeGraphRAGRetriever(BaseRetriever):
self,
query_str: str,
handle_fn: Optional[Callable],
- handle_llm_prompt_template: Optional[Prompt],
+ handle_llm_prompt_template: Optional[BasePromptTemplate],
cross_handle_policy: Optional[str] = "union",
max_items: Optional[int] = 5,
result_start_token: str = "KEYWORDS:",
diff --git a/llama_index/indices/list/base.py b/llama_index/indices/list/base.py
index a55270308a..1eb213cb6d 100644
--- a/llama_index/indices/list/base.py
+++ b/llama_index/indices/list/base.py
@@ -7,7 +7,6 @@ in sequence in order to answer a given query.
from enum import Enum
from typing import Any, Dict, Optional, Sequence, Union
-from llama_index.utils import get_tqdm_iterable
from llama_index.data_structs.data_structs import IndexList
from llama_index.indices.base import BaseIndex
@@ -15,6 +14,7 @@ from llama_index.indices.base_retriever import BaseRetriever
from llama_index.indices.service_context import ServiceContext
from llama_index.schema import BaseNode
from llama_index.storage.docstore.types import RefDocInfo
+from llama_index.utils import get_tqdm_iterable
class ListRetrieverMode(str, Enum):
@@ -35,7 +35,7 @@ class ListIndex(BaseIndex[IndexList]):
answer from all the nodes.
Args:
- text_qa_template (Optional[QuestionAnswerPrompt]): A Question-Answer Prompt
+ text_qa_template (Optional[BasePromptTemplate]): A Question-Answer Prompt
(see :ref:`Prompt-Templates`).
NOTE: this is a deprecated field.
show_progress (bool): Whether to show tqdm progress bars. Defaults to False.
diff --git a/llama_index/indices/list/retrievers.py b/llama_index/indices/list/retrievers.py
index 6c05987096..7d1eefcc7e 100644
--- a/llama_index/indices/list/retrievers.py
+++ b/llama_index/indices/list/retrievers.py
@@ -11,9 +11,9 @@ from llama_index.indices.utils import (
default_format_node_batch_fn,
default_parse_choice_select_answer_fn,
)
-from llama_index.prompts.choice_select import (
+from llama_index.prompts import PromptTemplate
+from llama_index.prompts.default_prompts import (
DEFAULT_CHOICE_SELECT_PROMPT,
- ChoiceSelectPrompt,
)
from llama_index.schema import BaseNode, NodeWithScore, MetadataMode
@@ -123,7 +123,7 @@ class ListIndexLLMRetriever(BaseRetriever):
Args:
index (ListIndex): The index to retrieve from.
- choice_select_prompt (Optional[ChoiceSelectPrompt]): A Choice-Select Prompt
+ choice_select_prompt (Optional[PromptTemplate]): A Choice-Select Prompt
(see :ref:`Prompt-Templates`).)
choice_batch_size (int): The number of nodes to query at a time.
format_node_batch_fn (Optional[Callable]): A function that formats a
@@ -137,7 +137,7 @@ class ListIndexLLMRetriever(BaseRetriever):
def __init__(
self,
index: ListIndex,
- choice_select_prompt: Optional[ChoiceSelectPrompt] = None,
+ choice_select_prompt: Optional[PromptTemplate] = None,
choice_batch_size: int = 10,
format_node_batch_fn: Optional[Callable] = None,
parse_choice_select_answer_fn: Optional[Callable] = None,
diff --git a/llama_index/indices/postprocessor/llm_rerank.py b/llama_index/indices/postprocessor/llm_rerank.py
index 6343c455e5..d22b98aa0e 100644
--- a/llama_index/indices/postprocessor/llm_rerank.py
+++ b/llama_index/indices/postprocessor/llm_rerank.py
@@ -8,8 +8,8 @@ from llama_index.indices.utils import (
default_format_node_batch_fn,
default_parse_choice_select_answer_fn,
)
-from llama_index.prompts.choice_select import DEFAULT_CHOICE_SELECT_PROMPT
-from llama_index.prompts.prompts import QuestionAnswerPrompt
+from llama_index.prompts import BasePromptTemplate
+from llama_index.prompts.default_prompts import DEFAULT_CHOICE_SELECT_PROMPT
from llama_index.schema import NodeWithScore
@@ -18,7 +18,7 @@ class LLMRerank(BaseNodePostprocessor):
def __init__(
self,
- choice_select_prompt: Optional[QuestionAnswerPrompt] = None,
+ choice_select_prompt: Optional[BasePromptTemplate] = None,
choice_batch_size: int = 10,
format_node_batch_fn: Optional[Callable] = None,
parse_choice_select_answer_fn: Optional[Callable] = None,
diff --git a/llama_index/indices/postprocessor/node.py b/llama_index/indices/postprocessor/node.py
index 2a55fa73ac..683f6e5a00 100644
--- a/llama_index/indices/postprocessor/node.py
+++ b/llama_index/indices/postprocessor/node.py
@@ -10,7 +10,7 @@ from pydantic import BaseModel, Field, validator
from llama_index.indices.postprocessor.types import BaseNodePostprocessor
from llama_index.indices.query.schema import QueryBundle
from llama_index.indices.service_context import ServiceContext
-from llama_index.prompts.prompts import QuestionAnswerPrompt, RefinePrompt
+from llama_index.prompts.base import PromptTemplate
from llama_index.response_synthesizers import ResponseMode, get_response_synthesizer
from llama_index.schema import NodeRelationship, NodeWithScore
from llama_index.storage.docstore import BaseDocumentStore
@@ -310,10 +310,10 @@ class AutoPrevNextNodePostprocessor(BasePydanticNodePostprocessor):
if query_bundle is None:
raise ValueError("Missing query bundle.")
- infer_prev_next_prompt = QuestionAnswerPrompt(
+ infer_prev_next_prompt = PromptTemplate(
self.infer_prev_next_tmpl,
)
- refine_infer_prev_next_prompt = RefinePrompt(self.refine_prev_next_tmpl)
+ refine_infer_prev_next_prompt = PromptTemplate(self.refine_prev_next_tmpl)
all_nodes: Dict[str, NodeWithScore] = {}
for node in nodes:
diff --git a/llama_index/indices/postprocessor/node_recency.py b/llama_index/indices/postprocessor/node_recency.py
index 60db27f2cb..dd18fa23f3 100644
--- a/llama_index/indices/postprocessor/node_recency.py
+++ b/llama_index/indices/postprocessor/node_recency.py
@@ -1,15 +1,15 @@
"""Node recency post-processor."""
-from pydantic import Field
-from typing import Optional, List, Set
-import pandas as pd
-import numpy as np
from datetime import datetime
+from typing import List, Optional, Set
+
+import numpy as np
+import pandas as pd
+from pydantic import Field
from llama_index.indices.postprocessor.node import BasePydanticNodePostprocessor
from llama_index.indices.query.schema import QueryBundle
from llama_index.indices.service_context import ServiceContext
-from llama_index.schema import NodeWithScore, MetadataMode
-
+from llama_index.schema import MetadataMode, NodeWithScore
# NOTE: currently not being used
# DEFAULT_INFER_RECENCY_TMPL = (
@@ -67,17 +67,6 @@ class FixedRecencyPostprocessor(BasePydanticNodePostprocessor):
if query_bundle is None:
raise ValueError("Missing query bundle in extra info.")
- # query_bundle = cast(QueryBundle, metadata["query_bundle"])
- # infer_recency_prompt = SimpleInputPrompt(self.infer_recency_tmpl)
- # raw_pred = self.service_context.llm_predictor.predict(
- # prompt=infer_recency_prompt,
- # query_str=query_bundle.query_str,
- # )
- # pred = parse_recency_pred(raw_pred)
- # # if no need to use recency post-processor, return nodes as is
- # if not pred:
- # return nodes
-
# sort nodes by date
node_dates = pd.to_datetime(
[node.node.metadata[self.date_key] for node in nodes]
@@ -130,17 +119,6 @@ class EmbeddingRecencyPostprocessor(BasePydanticNodePostprocessor):
if query_bundle is None:
raise ValueError("Missing query bundle in extra info.")
- # query_bundle = cast(QueryBundle, metadata["query_bundle"])
- # infer_recency_prompt = SimpleInputPrompt(self.infer_recency_tmpl)
- # raw_pred = self.service_context.llm_predictor.predict(
- # prompt=infer_recency_prompt,
- # query_str=query_bundle.query_str,
- # )
- # pred = parse_recency_pred(raw_pred)
- # # if no need to use recency post-processor, return nodes as is
- # if not pred:
- # return nodes
-
# sort nodes by date
node_dates = pd.to_datetime(
[node.node.metadata[self.date_key] for node in nodes]
diff --git a/llama_index/indices/postprocessor/pii.py b/llama_index/indices/postprocessor/pii.py
index 152932af3b..adadfacfb4 100644
--- a/llama_index/indices/postprocessor/pii.py
+++ b/llama_index/indices/postprocessor/pii.py
@@ -6,7 +6,7 @@ from typing import List, Optional, Dict, Tuple, Callable
from llama_index.indices.postprocessor.node import BasePydanticNodePostprocessor
from llama_index.indices.query.schema import QueryBundle
from llama_index.indices.service_context import ServiceContext
-from llama_index.prompts.prompts import QuestionAnswerPrompt
+from llama_index.prompts.base import PromptTemplate
from llama_index.schema import NodeWithScore, MetadataMode
@@ -56,7 +56,7 @@ class PIINodePostprocessor(BasePydanticNodePostprocessor):
def mask_pii(self, text: str) -> Tuple[str, Dict]:
"""Mask PII in text."""
- pii_prompt = QuestionAnswerPrompt(self.pii_str_tmpl)
+ pii_prompt = PromptTemplate(self.pii_str_tmpl)
# TODO: allow customization
task_str = (
"Mask out the PII, replace each PII with a tag, and return the text. "
diff --git a/llama_index/indices/prompt_helper.py b/llama_index/indices/prompt_helper.py
index 057c18910e..4aee137e4b 100644
--- a/llama_index/indices/prompt_helper.py
+++ b/llama_index/indices/prompt_helper.py
@@ -12,11 +12,12 @@ import logging
from pydantic import BaseModel, Field, PrivateAttr
from typing import Callable, List, Optional, Sequence
+from llama_index.prompts import BasePromptTemplate
+
from llama_index.constants import DEFAULT_CONTEXT_WINDOW, DEFAULT_NUM_OUTPUTS
from llama_index.llms.openai_utils import is_chat_model
from llama_index.llm_predictor.base import LLMMetadata
-from llama_index.prompts.base import Prompt
-from llama_index.prompts.utils import get_empty_prompt_txt
+from llama_index.prompts.prompt_utils import get_empty_prompt_txt
from llama_index.text_splitter import TokenTextSplitter
from llama_index.text_splitter.utils import truncate_text
from llama_index.utils import globals_helper
@@ -127,7 +128,7 @@ class PromptHelper(BaseModel):
separator=separator,
)
- def _get_available_context_size(self, prompt: Prompt) -> int:
+ def _get_available_context_size(self, prompt: BasePromptTemplate) -> int:
"""Get available context size.
This is calculated as:
@@ -142,7 +143,7 @@ class PromptHelper(BaseModel):
return self.context_window - num_prompt_tokens - self.num_output
def _get_available_chunk_size(
- self, prompt: Prompt, num_chunks: int = 1, padding: int = 5
+ self, prompt: BasePromptTemplate, num_chunks: int = 1, padding: int = 5
) -> int:
"""Get available chunk size.
@@ -169,7 +170,10 @@ class PromptHelper(BaseModel):
return result
def get_text_splitter_given_prompt(
- self, prompt: Prompt, num_chunks: int = 1, padding: int = DEFAULT_PADDING
+ self,
+ prompt: BasePromptTemplate,
+ num_chunks: int = 1,
+ padding: int = DEFAULT_PADDING,
) -> TokenTextSplitter:
"""Get text splitter configured to maximally pack available context window,
taking into account of given prompt, and desired number of chunks.
@@ -187,7 +191,10 @@ class PromptHelper(BaseModel):
return text_splitter
def truncate(
- self, prompt: Prompt, text_chunks: Sequence[str], padding: int = DEFAULT_PADDING
+ self,
+ prompt: BasePromptTemplate,
+ text_chunks: Sequence[str],
+ padding: int = DEFAULT_PADDING,
) -> List[str]:
"""Truncate text chunks to fit available context window."""
text_splitter = self.get_text_splitter_given_prompt(
@@ -198,7 +205,10 @@ class PromptHelper(BaseModel):
return [truncate_text(chunk, text_splitter) for chunk in text_chunks]
def repack(
- self, prompt: Prompt, text_chunks: Sequence[str], padding: int = DEFAULT_PADDING
+ self,
+ prompt: BasePromptTemplate,
+ text_chunks: Sequence[str],
+ padding: int = DEFAULT_PADDING,
) -> List[str]:
"""Repack text chunks to fit available context window.
diff --git a/llama_index/indices/query/query_transform/base.py b/llama_index/indices/query/query_transform/base.py
index d643bf7909..2677991168 100644
--- a/llama_index/indices/query/query_transform/base.py
+++ b/llama_index/indices/query/query_transform/base.py
@@ -17,7 +17,7 @@ from llama_index.indices.query.query_transform.prompts import (
from llama_index.indices.query.schema import QueryBundle, QueryType
from llama_index.llm_predictor import LLMPredictor
from llama_index.llm_predictor.base import BaseLLMPredictor
-from llama_index.prompts.base import Prompt
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompts import DEFAULT_HYDE_PROMPT
from llama_index.response.schema import Response
@@ -87,7 +87,7 @@ class HyDEQueryTransform(BaseQueryTransform):
def __init__(
self,
llm_predictor: Optional[BaseLLMPredictor] = None,
- hyde_prompt: Optional[Prompt] = None,
+ hyde_prompt: Optional[BasePromptTemplate] = None,
include_original: bool = True,
) -> None:
"""Initialize HyDEQueryTransform.
@@ -95,7 +95,7 @@ class HyDEQueryTransform(BaseQueryTransform):
Args:
llm_predictor (Optional[LLMPredictor]): LLM for generating
hypothetical documents
- hyde_prompt (Optional[Prompt]): Custom prompt for HyDE
+ hyde_prompt (Optional[BasePromptTemplate]): Custom prompt for HyDE
include_original (bool): Whether to include original query
string as one of the embedding strings
"""
diff --git a/llama_index/indices/query/query_transform/feedback_transform.py b/llama_index/indices/query/query_transform/feedback_transform.py
index 4ba335c41d..0874ec9170 100644
--- a/llama_index/indices/query/query_transform/feedback_transform.py
+++ b/llama_index/indices/query/query_transform/feedback_transform.py
@@ -6,7 +6,7 @@ from llama_index.indices.query.query_transform.base import BaseQueryTransform
from llama_index.indices.query.schema import QueryBundle
from llama_index.llm_predictor import LLMPredictor
from llama_index.llm_predictor.base import BaseLLMPredictor
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import BasePromptTemplate, PromptTemplate
from llama_index.response.schema import Response
logger = logging.getLogger(__name__)
@@ -22,7 +22,7 @@ DEFAULT_RESYNTHESIS_PROMPT_TMPL = (
"Otherwise, please return the original query.\n"
)
-DEFAULT_RESYNTHESIS_PROMPT = Prompt(DEFAULT_RESYNTHESIS_PROMPT_TMPL)
+DEFAULT_RESYNTHESIS_PROMPT = PromptTemplate(DEFAULT_RESYNTHESIS_PROMPT_TMPL)
class FeedbackQueryTransformation(BaseQueryTransform):
@@ -32,7 +32,7 @@ class FeedbackQueryTransformation(BaseQueryTransform):
eval(Evaluation): An evaluation object.
llm_predictor(BaseLLMPredictor): An LLM predictor.
resynthesize_query(bool): Whether to resynthesize the query.
- resynthesis_prompt(Prompt): A prompt for resynthesizing the query.
+ resynthesis_prompt(BasePromptTemplate): A prompt for resynthesizing the query.
"""
@@ -40,7 +40,7 @@ class FeedbackQueryTransformation(BaseQueryTransform):
self,
llm_predictor: Optional[BaseLLMPredictor] = None,
resynthesize_query: bool = False,
- resynthesis_prompt: Optional[Prompt] = None,
+ resynthesis_prompt: Optional[BasePromptTemplate] = None,
) -> None:
super().__init__()
self.llm_predictor = llm_predictor or LLMPredictor()
diff --git a/llama_index/indices/query/query_transform/prompts.py b/llama_index/indices/query/query_transform/prompts.py
index b963222674..fd65cb49c7 100644
--- a/llama_index/indices/query/query_transform/prompts.py
+++ b/llama_index/indices/query/query_transform/prompts.py
@@ -1,35 +1,35 @@
"""Query transform prompts."""
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import PromptTemplate
from llama_index.prompts.prompt_type import PromptType
# deprecated, kept for backwards compatibility
"""Decompose prompt for query transformation.
-Prompt to "decompose" a query into another query
+PromptTemplate to "decompose" a query into another query
given the existing context.
Required template variables: `context_str`, `query_str`
"""
-DecomposeQueryTransformPrompt = Prompt
+DecomposeQueryTransformPrompt = PromptTemplate
"""Step Decompose prompt for query transformation.
-Prompt to "decompose" a query into another query
+PromptTemplate to "decompose" a query into another query
given the existing context + previous reasoning (the previous steps).
Required template variables: `context_str`, `query_str`, `prev_reasoning`
"""
-StepDecomposeQueryTransformPrompt = Prompt
+StepDecomposeQueryTransformPrompt = PromptTemplate
"""Image output prompt for query transformation.
-Prompt to add instructions for formatting image output.
+PromptTemplate to add instructions for formatting image output.
Required template variables: `query_str`, `image_width`
"""
-ImageOutputQueryTransformPrompt = Prompt
+ImageOutputQueryTransformPrompt = PromptTemplate
DEFAULT_DECOMPOSE_QUERY_TRANSFORM_TMPL = (
@@ -59,7 +59,7 @@ DEFAULT_DECOMPOSE_QUERY_TRANSFORM_TMPL = (
"New question: "
)
-DEFAULT_DECOMPOSE_QUERY_TRANSFORM_PROMPT = Prompt(
+DEFAULT_DECOMPOSE_QUERY_TRANSFORM_PROMPT = PromptTemplate(
DEFAULT_DECOMPOSE_QUERY_TRANSFORM_TMPL, prompt_type=PromptType.DECOMPOSE
)
@@ -70,7 +70,7 @@ DEFAULT_IMAGE_OUTPUT_TMPL = (
'e.g., <image src="data/img.jpg" width="{image_width}" />.'
)
-DEFAULT_IMAGE_OUTPUT_PROMPT = Prompt(DEFAULT_IMAGE_OUTPUT_TMPL)
+DEFAULT_IMAGE_OUTPUT_PROMPT = PromptTemplate(DEFAULT_IMAGE_OUTPUT_TMPL)
DEFAULT_STEP_DECOMPOSE_QUERY_TRANSFORM_TMPL = (
@@ -125,6 +125,6 @@ DEFAULT_STEP_DECOMPOSE_QUERY_TRANSFORM_TMPL = (
"New question: "
)
-DEFAULT_STEP_DECOMPOSE_QUERY_TRANSFORM_PROMPT = Prompt(
+DEFAULT_STEP_DECOMPOSE_QUERY_TRANSFORM_PROMPT = PromptTemplate(
DEFAULT_STEP_DECOMPOSE_QUERY_TRANSFORM_TMPL
)
diff --git a/llama_index/indices/service_context.py b/llama_index/indices/service_context.py
index 496a09077f..6ae333bf24 100644
--- a/llama_index/indices/service_context.py
+++ b/llama_index/indices/service_context.py
@@ -5,6 +5,7 @@ from typing import Optional
import llama_index
from llama_index.callbacks.base import CallbackManager
from llama_index.embeddings.base import BaseEmbedding
+from llama_index.embeddings.utils import EmbedType, resolve_embed_model
from llama_index.indices.prompt_helper import PromptHelper
from llama_index.llm_predictor import LLMPredictor
from llama_index.llm_predictor.base import BaseLLMPredictor, LLMMetadata
@@ -13,8 +14,7 @@ from llama_index.llms.utils import LLMType, resolve_llm
from llama_index.logger import LlamaLogger
from llama_index.node_parser.interface import NodeParser
from llama_index.node_parser.simple import SimpleNodeParser
-from llama_index.prompts.prompts import Prompt
-from llama_index.embeddings.utils import resolve_embed_model, EmbedType
+from llama_index.prompts.base import BasePromptTemplate
logger = logging.getLogger(__name__)
@@ -78,7 +78,7 @@ class ServiceContext:
llama_logger: Optional[LlamaLogger] = None,
callback_manager: Optional[CallbackManager] = None,
system_prompt: Optional[str] = None,
- query_wrapper_prompt: Optional[Prompt] = None,
+ query_wrapper_prompt: Optional[BasePromptTemplate] = None,
# node parser kwargs
chunk_size: Optional[int] = None,
chunk_overlap: Optional[int] = None,
@@ -106,7 +106,7 @@ class ServiceContext:
callback_manager (Optional[CallbackManager]): CallbackManager
system_prompt (Optional[str]): System-wide prompt to be prepended
to all input prompts, used to guide system "decision making"
- query_wrapper_prompt (Optional[SimpleInputPrompt]): A format to wrap
+ query_wrapper_prompt (Optional[BasePromptTemplate]): A format to wrap
passed-in input queries.
Deprecated Args:
@@ -184,7 +184,7 @@ class ServiceContext:
llama_logger: Optional[LlamaLogger] = None,
callback_manager: Optional[CallbackManager] = None,
system_prompt: Optional[str] = None,
- query_wrapper_prompt: Optional[Prompt] = None,
+ query_wrapper_prompt: Optional[BasePromptTemplate] = None,
# node parser kwargs
chunk_size: Optional[int] = None,
chunk_overlap: Optional[int] = None,
diff --git a/llama_index/indices/struct_store/base.py b/llama_index/indices/struct_store/base.py
index 7bda01f48e..1657c4197e 100644
--- a/llama_index/indices/struct_store/base.py
+++ b/llama_index/indices/struct_store/base.py
@@ -6,8 +6,8 @@ from typing import Any, Callable, Dict, Generic, Optional, Sequence, TypeVar
from llama_index.data_structs.table import BaseStructTable
from llama_index.indices.base import BaseIndex
from llama_index.indices.service_context import ServiceContext
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompts import DEFAULT_SCHEMA_EXTRACT_PROMPT
-from llama_index.prompts.prompts import SchemaExtractPrompt
from llama_index.schema import BaseNode
from llama_index.storage.docstore.types import RefDocInfo
@@ -44,7 +44,7 @@ class BaseStructStoreIndex(BaseIndex[BST], Generic[BST]):
nodes: Optional[Sequence[BaseNode]] = None,
index_struct: Optional[BST] = None,
service_context: Optional[ServiceContext] = None,
- schema_extract_prompt: Optional[SchemaExtractPrompt] = None,
+ schema_extract_prompt: Optional[BasePromptTemplate] = None,
output_parser: Optional[OUTPUT_PARSER_TYPE] = None,
**kwargs: Any,
) -> None:
diff --git a/llama_index/indices/struct_store/json_query.py b/llama_index/indices/struct_store/json_query.py
index 9d1486037c..0a8162e43d 100644
--- a/llama_index/indices/struct_store/json_query.py
+++ b/llama_index/indices/struct_store/json_query.py
@@ -7,7 +7,7 @@ from llama_index.bridge.langchain import print_text
from llama_index.indices.query.base import BaseQueryEngine
from llama_index.indices.query.schema import QueryBundle
from llama_index.indices.service_context import ServiceContext
-from llama_index.prompts.base import Prompt
+from llama_index.prompts import PromptTemplate, BasePromptTemplate
from llama_index.prompts.default_prompts import DEFAULT_JSON_PATH_PROMPT
from llama_index.prompts.prompt_type import PromptType
from llama_index.response.schema import Response
@@ -31,7 +31,7 @@ DEFAULT_RESPONSE_SYNTHESIS_PROMPT_TMPL = (
"Query: {query_str}\n"
"Response: "
)
-DEFAULT_RESPONSE_SYNTHESIS_PROMPT = Prompt(
+DEFAULT_RESPONSE_SYNTHESIS_PROMPT = PromptTemplate(
DEFAULT_RESPONSE_SYNTHESIS_PROMPT_TMPL,
prompt_type=PromptType.SQL_RESPONSE_SYNTHESIS,
)
@@ -58,7 +58,7 @@ class JSONQueryEngine(BaseQueryEngine):
json_value (JSONType): JSON value
json_schema (JSONType): JSON schema
service_context (ServiceContext): ServiceContext
- json_path_prompt (Prompt): The JSON Path prompt to use.
+ json_path_prompt (BasePromptTemplate): The JSON Path prompt to use.
output_processor (Callable): The output processor that executes the
JSON Path query.
output_kwargs (dict): Additional output processor kwargs for the
@@ -71,11 +71,11 @@ class JSONQueryEngine(BaseQueryEngine):
json_value: JSONType,
json_schema: JSONType,
service_context: ServiceContext,
- json_path_prompt: Optional[Prompt] = None,
+ json_path_prompt: Optional[BasePromptTemplate] = None,
output_processor: Optional[Callable] = None,
output_kwargs: Optional[dict] = None,
synthesize_response: bool = True,
- response_synthesis_prompt: Optional[Prompt] = None,
+ response_synthesis_prompt: Optional[BasePromptTemplate] = None,
verbose: bool = False,
**kwargs: Any,
) -> None:
diff --git a/llama_index/indices/struct_store/sql_query.py b/llama_index/indices/struct_store/sql_query.py
index 89e8acaf00..33aefc4e30 100644
--- a/llama_index/indices/struct_store/sql_query.py
+++ b/llama_index/indices/struct_store/sql_query.py
@@ -15,7 +15,7 @@ from llama_index.indices.struct_store.sql import SQLStructStoreIndex
from llama_index.langchain_helpers.sql_wrapper import SQLDatabase
from llama_index.objects.base import ObjectRetriever
from llama_index.objects.table_node_mapping import SQLTableSchema
-from llama_index.prompts.base import Prompt
+from llama_index.prompts import BasePromptTemplate, PromptTemplate
from llama_index.prompts.default_prompts import DEFAULT_TEXT_TO_SQL_PROMPT
from llama_index.prompts.prompt_type import PromptType
from llama_index.response.schema import Response
@@ -30,7 +30,7 @@ DEFAULT_RESPONSE_SYNTHESIS_PROMPT_TMPL = (
"SQL Response: {sql_response_str}\n"
"Response: "
)
-DEFAULT_RESPONSE_SYNTHESIS_PROMPT = Prompt(
+DEFAULT_RESPONSE_SYNTHESIS_PROMPT = PromptTemplate(
DEFAULT_RESPONSE_SYNTHESIS_PROMPT_TMPL,
prompt_type=PromptType.SQL_RESPONSE_SYNTHESIS,
)
@@ -85,24 +85,25 @@ class NLStructStoreQueryEngine(BaseQueryEngine):
Args:
index (SQLStructStoreIndex): A SQL Struct Store Index
- text_to_sql_prompt (Optional[Prompt]): A Text to SQL Prompt
- to use for the query. Defaults to DEFAULT_TEXT_TO_SQL_PROMPT.
+ text_to_sql_prompt (Optional[BasePromptTemplate]): A Text to SQL
+ BasePromptTemplate to use for the query.
+ Defaults to DEFAULT_TEXT_TO_SQL_PROMPT.
context_query_kwargs (Optional[dict]): Keyword arguments for the
context query. Defaults to {}.
synthesize_response (bool): Whether to synthesize a response from the
query results. Defaults to True.
- response_synthesis_prompt (Optional[Prompt]): A
- Response Synthesis Prompt to use for the query. Defaults to
+ response_synthesis_prompt (Optional[BasePromptTemplate]): A
+ Response Synthesis BasePromptTemplate to use for the query. Defaults to
DEFAULT_RESPONSE_SYNTHESIS_PROMPT.
"""
def __init__(
self,
index: SQLStructStoreIndex,
- text_to_sql_prompt: Optional[Prompt] = None,
+ text_to_sql_prompt: Optional[BasePromptTemplate] = None,
context_query_kwargs: Optional[dict] = None,
synthesize_response: bool = True,
- response_synthesis_prompt: Optional[Prompt] = None,
+ response_synthesis_prompt: Optional[BasePromptTemplate] = None,
**kwargs: Any,
) -> None:
"""Initialize params."""
@@ -215,10 +216,10 @@ class BaseSQLTableQueryEngine(BaseQueryEngine):
def __init__(
self,
sql_database: SQLDatabase,
- text_to_sql_prompt: Optional[Prompt] = None,
+ text_to_sql_prompt: Optional[BasePromptTemplate] = None,
context_query_kwargs: Optional[dict] = None,
synthesize_response: bool = True,
- response_synthesis_prompt: Optional[Prompt] = None,
+ response_synthesis_prompt: Optional[BasePromptTemplate] = None,
service_context: Optional[ServiceContext] = None,
**kwargs: Any,
) -> None:
@@ -321,10 +322,10 @@ class NLSQLTableQueryEngine(BaseSQLTableQueryEngine):
def __init__(
self,
sql_database: SQLDatabase,
- text_to_sql_prompt: Optional[Prompt] = None,
+ text_to_sql_prompt: Optional[BasePromptTemplate] = None,
context_query_kwargs: Optional[dict] = None,
synthesize_response: bool = True,
- response_synthesis_prompt: Optional[Prompt] = None,
+ response_synthesis_prompt: Optional[BasePromptTemplate] = None,
tables: Optional[Union[List[str], List[Table]]] = None,
service_context: Optional[ServiceContext] = None,
**kwargs: Any,
@@ -389,10 +390,10 @@ class SQLTableRetrieverQueryEngine(BaseSQLTableQueryEngine):
self,
sql_database: SQLDatabase,
table_retriever: ObjectRetriever[SQLTableSchema],
- text_to_sql_prompt: Optional[Prompt] = None,
+ text_to_sql_prompt: Optional[BasePromptTemplate] = None,
context_query_kwargs: Optional[dict] = None,
synthesize_response: bool = True,
- response_synthesis_prompt: Optional[Prompt] = None,
+ response_synthesis_prompt: Optional[BasePromptTemplate] = None,
service_context: Optional[ServiceContext] = None,
context_str_prefix: Optional[str] = None,
**kwargs: Any,
diff --git a/llama_index/indices/tree/all_leaf_retriever.py b/llama_index/indices/tree/all_leaf_retriever.py
index 85083b0d90..91d208c2d2 100644
--- a/llama_index/indices/tree/all_leaf_retriever.py
+++ b/llama_index/indices/tree/all_leaf_retriever.py
@@ -23,7 +23,7 @@ class TreeAllLeafRetriever(BaseRetriever):
when initialized, since we rebuild the tree for each query.
Args:
- text_qa_template (Optional[QuestionAnswerPrompt]): Question-Answer Prompt
+ text_qa_template (Optional[BasePromptTemplate]): Question-Answer Prompt
(see :ref:`Prompt-Templates`).
"""
diff --git a/llama_index/indices/tree/base.py b/llama_index/indices/tree/base.py
index 67ac0ffa6e..bfd8da7f4e 100644
--- a/llama_index/indices/tree/base.py
+++ b/llama_index/indices/tree/base.py
@@ -10,11 +10,11 @@ from llama_index.indices.base_retriever import BaseRetriever
from llama_index.indices.common_tree.base import GPTTreeIndexBuilder
from llama_index.indices.service_context import ServiceContext
from llama_index.indices.tree.inserter import TreeIndexInserter
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompts import (
DEFAULT_INSERT_PROMPT,
DEFAULT_SUMMARY_PROMPT,
)
-from llama_index.prompts.prompts import SummaryPrompt, TreeInsertPrompt
from llama_index.schema import BaseNode
from llama_index.storage.docstore.types import RefDocInfo
@@ -45,9 +45,9 @@ class TreeIndex(BaseIndex[IndexGraph]):
A secondary answer is to directly synthesize the answer from the root nodes.
Args:
- summary_template (Optional[SummaryPrompt]): A Summarization Prompt
+ summary_template (Optional[BasePromptTemplate]): A Summarization Prompt
(see :ref:`Prompt-Templates`).
- insert_prompt (Optional[TreeInsertPrompt]): An Tree Insertion Prompt
+ insert_prompt (Optional[BasePromptTemplate]): An Tree Insertion Prompt
(see :ref:`Prompt-Templates`).
num_children (int): The number of children each node should have.
build_tree (bool): Whether to build the tree during index construction.
@@ -62,8 +62,8 @@ class TreeIndex(BaseIndex[IndexGraph]):
nodes: Optional[Sequence[BaseNode]] = None,
index_struct: Optional[IndexGraph] = None,
service_context: Optional[ServiceContext] = None,
- summary_template: Optional[SummaryPrompt] = None,
- insert_prompt: Optional[TreeInsertPrompt] = None,
+ summary_template: Optional[BasePromptTemplate] = None,
+ insert_prompt: Optional[BasePromptTemplate] = None,
num_children: int = 10,
build_tree: bool = True,
use_async: bool = False,
@@ -74,7 +74,7 @@ class TreeIndex(BaseIndex[IndexGraph]):
# need to set parameters before building index in base class.
self.num_children = num_children
self.summary_template = summary_template or DEFAULT_SUMMARY_PROMPT
- self.insert_prompt: TreeInsertPrompt = insert_prompt or DEFAULT_INSERT_PROMPT
+ self.insert_prompt: BasePromptTemplate = insert_prompt or DEFAULT_INSERT_PROMPT
self.build_tree = build_tree
self._use_async = use_async
super().__init__(
diff --git a/llama_index/indices/tree/inserter.py b/llama_index/indices/tree/inserter.py
index 879061a7ef..0e28823955 100644
--- a/llama_index/indices/tree/inserter.py
+++ b/llama_index/indices/tree/inserter.py
@@ -11,7 +11,7 @@ from llama_index.indices.utils import (
extract_numbers_given_response,
get_sorted_node_list,
)
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import BasePromptTemplate
from llama_index.prompts.default_prompts import (
DEFAULT_INSERT_PROMPT,
DEFAULT_SUMMARY_PROMPT,
@@ -27,8 +27,8 @@ class TreeIndexInserter:
index_graph: IndexGraph,
service_context: ServiceContext,
num_children: int = 10,
- insert_prompt: Prompt = DEFAULT_INSERT_PROMPT,
- summary_prompt: Prompt = DEFAULT_SUMMARY_PROMPT,
+ insert_prompt: BasePromptTemplate = DEFAULT_INSERT_PROMPT,
+ summary_prompt: BasePromptTemplate = DEFAULT_SUMMARY_PROMPT,
docstore: Optional[BaseDocumentStore] = None,
) -> None:
"""Initialize with params."""
diff --git a/llama_index/indices/tree/select_leaf_embedding_retriever.py b/llama_index/indices/tree/select_leaf_embedding_retriever.py
index 009cc3285e..94031f7f94 100644
--- a/llama_index/indices/tree/select_leaf_embedding_retriever.py
+++ b/llama_index/indices/tree/select_leaf_embedding_retriever.py
@@ -3,11 +3,8 @@
import logging
from typing import Dict, List, Tuple, cast
-
from llama_index.indices.query.schema import QueryBundle
-from llama_index.indices.tree.select_leaf_retriever import (
- TreeSelectLeafRetriever,
-)
+from llama_index.indices.tree.select_leaf_retriever import TreeSelectLeafRetriever
from llama_index.indices.utils import get_sorted_node_list
from llama_index.schema import BaseNode, MetadataMode
@@ -21,14 +18,14 @@ class TreeSelectLeafEmbeddingRetriever(TreeSelectLeafRetriever):
query and the node text.
Args:
- query_template (Optional[TreeSelectPrompt]): Tree Select Query Prompt
+ query_template (Optional[BasePromptTemplate]): Tree Select Query Prompt
(see :ref:`Prompt-Templates`).
- query_template_multiple (Optional[TreeSelectMultiplePrompt]): Tree Select
+ query_template_multiple (Optional[BasePromptTemplate]): Tree Select
Query Prompt (Multiple)
(see :ref:`Prompt-Templates`).
- text_qa_template (Optional[QuestionAnswerPrompt]): Question-Answer Prompt
+ text_qa_template (Optional[BasePromptTemplate]): Question-Answer Prompt
(see :ref:`Prompt-Templates`).
- refine_template (Optional[RefinePrompt]): Refinement Prompt
+ refine_template (Optional[BasePromptTemplate]): Refinement Prompt
(see :ref:`Prompt-Templates`).
child_branch_factor (int): Number of child nodes to consider at each level.
If child_branch_factor is 1, then the query will only choose one child node
diff --git a/llama_index/indices/tree/select_leaf_retriever.py b/llama_index/indices/tree/select_leaf_retriever.py
index 6797728c9e..f921ec8d5c 100644
--- a/llama_index/indices/tree/select_leaf_retriever.py
+++ b/llama_index/indices/tree/select_leaf_retriever.py
@@ -4,7 +4,6 @@ import logging
from typing import Any, Dict, List, Optional, cast
from llama_index.bridge.langchain import print_text
-
from llama_index.indices.base_retriever import BaseRetriever
from llama_index.indices.query.schema import QueryBundle
from llama_index.indices.tree.base import TreeIndex
@@ -13,21 +12,16 @@ from llama_index.indices.utils import (
extract_numbers_given_response,
get_sorted_node_list,
)
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompt_selectors import DEFAULT_REFINE_PROMPT_SEL
from llama_index.prompts.default_prompts import (
DEFAULT_QUERY_PROMPT,
DEFAULT_QUERY_PROMPT_MULTIPLE,
DEFAULT_TEXT_QA_PROMPT,
)
-from llama_index.prompts.prompts import (
- QuestionAnswerPrompt,
- RefinePrompt,
- TreeSelectMultiplePrompt,
- TreeSelectPrompt,
-)
from llama_index.response.schema import Response
from llama_index.response_synthesizers import get_response_synthesizer
-from llama_index.schema import BaseNode, NodeWithScore, MetadataMode
+from llama_index.schema import BaseNode, MetadataMode, NodeWithScore
from llama_index.utils import truncate_text
logger = logging.getLogger(__name__)
@@ -57,9 +51,9 @@ class TreeSelectLeafRetriever(BaseRetriever):
answer the query.
Args:
- query_template (Optional[TreeSelectPrompt]): Tree Select Query Prompt
+ query_template (Optional[BasePromptTemplate]): Tree Select Query Prompt
(see :ref:`Prompt-Templates`).
- query_template_multiple (Optional[TreeSelectMultiplePrompt]): Tree Select
+ query_template_multiple (Optional[BasePromptTemplate]): Tree Select
Query Prompt (Multiple)
(see :ref:`Prompt-Templates`).
child_branch_factor (int): Number of child nodes to consider at each level.
@@ -72,10 +66,10 @@ class TreeSelectLeafRetriever(BaseRetriever):
def __init__(
self,
index: TreeIndex,
- query_template: Optional[TreeSelectPrompt] = None,
- text_qa_template: Optional[QuestionAnswerPrompt] = None,
- refine_template: Optional[RefinePrompt] = None,
- query_template_multiple: Optional[TreeSelectMultiplePrompt] = None,
+ query_template: Optional[BasePromptTemplate] = None,
+ text_qa_template: Optional[BasePromptTemplate] = None,
+ refine_template: Optional[BasePromptTemplate] = None,
+ query_template_multiple: Optional[BasePromptTemplate] = None,
child_branch_factor: int = 1,
verbose: bool = False,
**kwargs: Any,
diff --git a/llama_index/indices/vector_store/retrievers/auto_retriever/prompts.py b/llama_index/indices/vector_store/retrievers/auto_retriever/prompts.py
index 7f3c7e2aad..37f1ae00bb 100644
--- a/llama_index/indices/vector_store/retrievers/auto_retriever/prompts.py
+++ b/llama_index/indices/vector_store/retrievers/auto_retriever/prompts.py
@@ -1,7 +1,7 @@
"""Autoretriever prompts."""
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import PromptTemplate
from llama_index.prompts.prompt_type import PromptType
from llama_index.vector_stores.types import (
ExactMatchFilter,
@@ -103,9 +103,9 @@ DEFAULT_VECTOR_STORE_QUERY_PROMPT_TMPL = PREFIX + EXAMPLES + SUFFIX
# deprecated, kept for backwards compatibility
"""Vector store query prompt."""
-VectorStoreQueryPrompt = Prompt
+VectorStoreQueryPrompt = PromptTemplate
-DEFAULT_VECTOR_STORE_QUERY_PROMPT = Prompt(
+DEFAULT_VECTOR_STORE_QUERY_PROMPT = PromptTemplate(
template=DEFAULT_VECTOR_STORE_QUERY_PROMPT_TMPL,
prompt_type=PromptType.VECTOR_STORE_QUERY,
)
diff --git a/llama_index/llm_predictor/base.py b/llama_index/llm_predictor/base.py
index d65cbb7cb2..0949da7691 100644
--- a/llama_index/llm_predictor/base.py
+++ b/llama_index/llm_predictor/base.py
@@ -1,10 +1,11 @@
"""Wrapper functions around an LLM chain."""
import logging
-from abc import abstractmethod, ABC
-from pydantic import BaseModel, PrivateAttr
+from abc import ABC, abstractmethod
from typing import Any, List, Optional
+from pydantic import BaseModel, PrivateAttr
+
from llama_index.callbacks.base import CallbackManager
from llama_index.llm_predictor.utils import (
astream_chat_response_to_tokens,
@@ -15,8 +16,12 @@ from llama_index.llm_predictor.utils import (
from llama_index.llms.base import LLM, ChatMessage, LLMMetadata, MessageRole
from llama_index.llms.generic_utils import messages_to_prompt
from llama_index.llms.utils import LLMType, resolve_llm
-from llama_index.prompts.base import Prompt
-from llama_index.prompts.prompts import SimpleInputPrompt
+from llama_index.prompts.base import (
+ BasePromptTemplate,
+ ChatPromptTemplate,
+ PromptTemplate,
+ SelectorPromptTemplate,
+)
from llama_index.types import TokenAsyncGen, TokenGen
logger = logging.getLogger(__name__)
@@ -36,19 +41,21 @@ class BaseLLMPredictor(BaseModel, ABC):
"""Get LLM metadata."""
@abstractmethod
- def predict(self, prompt: Prompt, **prompt_args: Any) -> str:
+ def predict(self, prompt: BasePromptTemplate, **prompt_args: Any) -> str:
"""Predict the answer to a query."""
@abstractmethod
- def stream(self, prompt: Prompt, **prompt_args: Any) -> TokenGen:
+ def stream(self, prompt: BasePromptTemplate, **prompt_args: Any) -> TokenGen:
"""Stream the answer to a query."""
@abstractmethod
- async def apredict(self, prompt: Prompt, **prompt_args: Any) -> str:
+ async def apredict(self, prompt: BasePromptTemplate, **prompt_args: Any) -> str:
"""Async predict the answer to a query."""
@abstractmethod
- async def astream(self, prompt: Prompt, **prompt_args: Any) -> TokenAsyncGen:
+ async def astream(
+ self, prompt: BasePromptTemplate, **prompt_args: Any
+ ) -> TokenAsyncGen:
"""Async predict the answer to a query."""
@@ -67,7 +74,7 @@ class LLMPredictor(BaseLLMPredictor):
arbitrary_types_allowed = True
system_prompt: Optional[str]
- query_wrapper_prompt: Optional[Prompt]
+ query_wrapper_prompt: Optional[BasePromptTemplate]
_llm: LLM = PrivateAttr()
def __init__(
@@ -75,7 +82,7 @@ class LLMPredictor(BaseLLMPredictor):
llm: Optional[LLMType] = None,
callback_manager: Optional[CallbackManager] = None,
system_prompt: Optional[str] = None,
- query_wrapper_prompt: Optional[SimpleInputPrompt] = None,
+ query_wrapper_prompt: Optional[BasePromptTemplate] = None,
) -> None:
"""Initialize params."""
self._llm = resolve_llm(llm)
@@ -97,7 +104,7 @@ class LLMPredictor(BaseLLMPredictor):
"""Get LLM metadata."""
return self._llm.metadata
- def predict(self, prompt: Prompt, **prompt_args: Any) -> str:
+ def predict(self, prompt: BasePromptTemplate, **prompt_args: Any) -> str:
"""Predict."""
if self._llm.metadata.is_chat_model:
messages = prompt.format_messages(llm=self._llm, **prompt_args)
@@ -116,7 +123,7 @@ class LLMPredictor(BaseLLMPredictor):
return output
- def stream(self, prompt: Prompt, **prompt_args: Any) -> TokenGen:
+ def stream(self, prompt: BasePromptTemplate, **prompt_args: Any) -> TokenGen:
"""Stream."""
if self._llm.metadata.is_chat_model:
messages = prompt.format_messages(llm=self._llm, **prompt_args)
@@ -130,7 +137,7 @@ class LLMPredictor(BaseLLMPredictor):
stream_tokens = stream_completion_response_to_tokens(stream_response)
return stream_tokens
- async def apredict(self, prompt: Prompt, **prompt_args: Any) -> str:
+ async def apredict(self, prompt: BasePromptTemplate, **prompt_args: Any) -> str:
"""Async predict."""
if self._llm.metadata.is_chat_model:
messages = prompt.format_messages(llm=self._llm, **prompt_args)
@@ -149,7 +156,9 @@ class LLMPredictor(BaseLLMPredictor):
return output
- async def astream(self, prompt: Prompt, **prompt_args: Any) -> TokenAsyncGen:
+ async def astream(
+ self, prompt: BasePromptTemplate, **prompt_args: Any
+ ) -> TokenAsyncGen:
"""Async stream."""
if self._llm.metadata.is_chat_model:
messages = prompt.format_messages(llm=self._llm, **prompt_args)
@@ -163,18 +172,40 @@ class LLMPredictor(BaseLLMPredictor):
stream_tokens = await astream_completion_response_to_tokens(stream_response)
return stream_tokens
- def _extend_prompt(self, prompt: Prompt) -> Prompt:
+ def _extend_prompt(self, prompt: BasePromptTemplate) -> BasePromptTemplate:
"""Add system and query wrapper prompts to base prompt"""
+ # TODO: avoid mutating prompt attributes
if self.system_prompt:
- prompt.prompt_selector.default_prompt.template = (
- self.system_prompt
- + "\n\n"
- + prompt.prompt_selector.default_prompt.template
- )
+ if isinstance(prompt, SelectorPromptTemplate):
+ default_template = prompt.default_template
+ if isinstance(default_template, PromptTemplate):
+ default_template.template = (
+ self.system_prompt + "\n\n" + default_template.template
+ )
+ else:
+ raise ValueError("PromptTemplate expected as default_template")
+ elif isinstance(prompt, ChatPromptTemplate):
+ prompt.message_templates = [
+ ChatMessage(role=MessageRole.SYSTEM, content=self.system_prompt)
+ ] + prompt.message_templates
+ elif isinstance(prompt, PromptTemplate):
+ prompt.template = self.system_prompt + "\n\n" + prompt.template
+
if self.query_wrapper_prompt:
- prompt.partial_dict["query_str"] = self.query_wrapper_prompt.format(
- query_str=prompt.partial_dict["query_str"]
- )
+ if isinstance(prompt, (PromptTemplate, ChatPromptTemplate)):
+ prompt.kwargs["query_str"] = self.query_wrapper_prompt.format(
+ query_str=prompt.kwargs["query_str"]
+ )
+ elif isinstance(prompt, SelectorPromptTemplate):
+ if isinstance(default_template, PromptTemplate):
+ prompt.default_template.kwargs[
+ "query_str"
+ ] = self.query_wrapper_prompt.format(
+ query_str=prompt.default_template.kwargs["query_str"]
+ )
+ else:
+ raise ValueError("PromptTemplate expected as default_template")
+
return prompt
def _extend_messages(self, messages: List[ChatMessage]) -> List[ChatMessage]:
diff --git a/llama_index/llm_predictor/mock.py b/llama_index/llm_predictor/mock.py
index 37e442ce80..42ff863c30 100644
--- a/llama_index/llm_predictor/mock.py
+++ b/llama_index/llm_predictor/mock.py
@@ -1,11 +1,12 @@
"""Mock LLM Predictor."""
-from pydantic import Field
from typing import Any, Dict
+from pydantic import Field
+
from llama_index.constants import DEFAULT_NUM_OUTPUTS
from llama_index.llm_predictor.base import BaseLLMPredictor
-from llama_index.llms.base import LLMMetadata, LLM
-from llama_index.prompts.base import Prompt
+from llama_index.llms.base import LLM, LLMMetadata
+from llama_index.prompts.base import BasePromptTemplate
from llama_index.prompts.prompt_type import PromptType
from llama_index.token_counter.utils import (
mock_extract_keywords_response,
@@ -49,13 +50,13 @@ def _mock_answer(max_tokens: int, prompt_args: Dict) -> str:
return " ".join(["answer"] * token_limit)
-def _mock_refine(max_tokens: int, prompt: Prompt, prompt_args: Dict) -> str:
+def _mock_refine(max_tokens: int, prompt: BasePromptTemplate, prompt_args: Dict) -> str:
"""Mock refine."""
# tokens in response shouldn't be larger than tokens in
# `existing_answer` + `context_msg`
# NOTE: if existing_answer is not in prompt_args, we need to get it from the prompt
if "existing_answer" not in prompt_args:
- existing_answer = prompt.partial_dict["existing_answer"]
+ existing_answer = prompt.kwargs["existing_answer"]
else:
existing_answer = prompt_args["existing_answer"]
num_ctx_tokens = len(globals_helper.tokenizer(prompt_args["context_msg"]))
@@ -96,10 +97,10 @@ class MockLLMPredictor(BaseLLMPredictor):
def llm(self) -> LLM:
raise NotImplementedError("MockLLMPredictor does not have an LLM model.")
- def predict(self, prompt: Prompt, **prompt_args: Any) -> str:
+ def predict(self, prompt: BasePromptTemplate, **prompt_args: Any) -> str:
"""Mock predict."""
- prompt_str = prompt.prompt_type
+ prompt_str = prompt.metadata["prompt_type"]
if prompt_str == PromptType.SUMMARY:
output = _mock_summary_predict(self.max_tokens, prompt_args)
elif prompt_str == PromptType.TREE_INSERT:
@@ -118,7 +119,7 @@ class MockLLMPredictor(BaseLLMPredictor):
output = _mock_query_keyword_extract(prompt_args)
elif prompt_str == PromptType.KNOWLEDGE_TRIPLET_EXTRACT:
output = _mock_knowledge_graph_triplet_extract(
- prompt_args, prompt.partial_dict.get("max_knowledge_triplets", 2)
+ prompt_args, int(prompt.kwargs.get("max_knowledge_triplets", 2))
)
elif prompt_str == PromptType.CUSTOM:
# we don't know specific prompt type, return generic response
@@ -128,11 +129,13 @@ class MockLLMPredictor(BaseLLMPredictor):
return output
- def stream(self, prompt: Prompt, **prompt_args: Any) -> TokenGen:
+ def stream(self, prompt: BasePromptTemplate, **prompt_args: Any) -> TokenGen:
raise NotImplementedError
- async def apredict(self, prompt: Prompt, **prompt_args: Any) -> str:
+ async def apredict(self, prompt: BasePromptTemplate, **prompt_args: Any) -> str:
return self.predict(prompt, **prompt_args)
- async def astream(self, prompt: Prompt, **prompt_args: Any) -> TokenAsyncGen:
+ async def astream(
+ self, prompt: BasePromptTemplate, **prompt_args: Any
+ ) -> TokenAsyncGen:
raise NotImplementedError
diff --git a/llama_index/llm_predictor/structured.py b/llama_index/llm_predictor/structured.py
index 76eb862e91..9dfcdb2e36 100644
--- a/llama_index/llm_predictor/structured.py
+++ b/llama_index/llm_predictor/structured.py
@@ -5,7 +5,7 @@ import logging
from typing import Any
from llama_index.llm_predictor.base import LLMPredictor
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import BasePromptTemplate
from llama_index.types import TokenGen
logger = logging.getLogger(__name__)
@@ -19,11 +19,11 @@ class StructuredLLMPredictor(LLMPredictor):
"""
- def predict(self, prompt: Prompt, **prompt_args: Any) -> str:
+ def predict(self, prompt: BasePromptTemplate, **prompt_args: Any) -> str:
"""Predict the answer to a query.
Args:
- prompt (Prompt): Prompt to use for prediction.
+ prompt (BasePromptTemplate): BasePromptTemplate to use for prediction.
Returns:
Tuple[str, str]: Tuple of the predicted answer and the formatted prompt.
@@ -33,20 +33,21 @@ class StructuredLLMPredictor(LLMPredictor):
# run output parser
if prompt.output_parser is not None:
# TODO: return other formats
- parsed_llm_prediction = str(prompt.output_parser.parse(llm_prediction))
+ output_parser = prompt.output_parser
+ parsed_llm_prediction = str(output_parser.parse(llm_prediction))
else:
parsed_llm_prediction = llm_prediction
return parsed_llm_prediction
- def stream(self, prompt: Prompt, **prompt_args: Any) -> TokenGen:
+ def stream(self, prompt: BasePromptTemplate, **prompt_args: Any) -> TokenGen:
"""Stream the answer to a query.
NOTE: this is a beta feature. Will try to build or use
better abstractions about response handling.
Args:
- prompt (Prompt): Prompt to use for prediction.
+ prompt (BasePromptTemplate): BasePromptTemplate to use for prediction.
Returns:
str: The predicted answer.
@@ -56,11 +57,11 @@ class StructuredLLMPredictor(LLMPredictor):
"Streaming is not supported for structured LLM predictor."
)
- async def apredict(self, prompt: Prompt, **prompt_args: Any) -> str:
+ async def apredict(self, prompt: BasePromptTemplate, **prompt_args: Any) -> str:
"""Async predict the answer to a query.
Args:
- prompt (Prompt): Prompt to use for prediction.
+ prompt (BasePromptTemplate): BasePromptTemplate to use for prediction.
Returns:
Tuple[str, str]: Tuple of the predicted answer and the formatted prompt.
@@ -68,7 +69,8 @@ class StructuredLLMPredictor(LLMPredictor):
"""
llm_prediction = await super().apredict(prompt, **prompt_args)
if prompt.output_parser is not None:
- parsed_llm_prediction = str(prompt.output_parser.parse(llm_prediction))
+ output_parser = prompt.output_parser
+ parsed_llm_prediction = str(output_parser.parse(llm_prediction))
else:
parsed_llm_prediction = llm_prediction
return parsed_llm_prediction
diff --git a/llama_index/llm_predictor/vellum/predictor.py b/llama_index/llm_predictor/vellum/predictor.py
index 1cf6a05f54..b262c49606 100644
--- a/llama_index/llm_predictor/vellum/predictor.py
+++ b/llama_index/llm_predictor/vellum/predictor.py
@@ -3,7 +3,7 @@ from __future__ import annotations
from pydantic import Field, PrivateAttr
from typing import Any, Optional, Tuple, cast
-from llama_index import Prompt
+from llama_index.prompts import BasePromptTemplate
from llama_index.callbacks import CallbackManager
from llama_index.callbacks.schema import CBEventType, EventPayload
from llama_index.llm_predictor.base import BaseLLMPredictor, LLMMetadata, LLM
@@ -64,7 +64,7 @@ class VellumPredictor(BaseLLMPredictor):
"""Get the LLM."""
raise NotImplementedError("Vellum does not expose the LLM.")
- def predict(self, prompt: Prompt, **prompt_args: Any) -> str:
+ def predict(self, prompt: BasePromptTemplate, **prompt_args: Any) -> str:
"""Predict the answer to a query."""
from vellum import GenerateRequest
@@ -73,11 +73,13 @@ class VellumPredictor(BaseLLMPredictor):
prompt, **prompt_args
)
+ input_values = {
+ **prompt.kwargs,
+ **prompt_args,
+ }
result = self._vellum_client.generate(
deployment_id=registered_prompt.deployment_id,
- requests=[
- GenerateRequest(input_values=prompt.get_full_format_args(prompt_args))
- ],
+ requests=[GenerateRequest(input_values=input_values)],
)
completion_text = self._process_generate_response(
@@ -86,7 +88,7 @@ class VellumPredictor(BaseLLMPredictor):
return completion_text
- def stream(self, prompt: Prompt, **prompt_args: Any) -> TokenGen:
+ def stream(self, prompt: BasePromptTemplate, **prompt_args: Any) -> TokenGen:
"""Stream the answer to a query."""
from vellum import GenerateRequest, GenerateStreamResult
@@ -95,11 +97,13 @@ class VellumPredictor(BaseLLMPredictor):
prompt, **prompt_args
)
+ input_values = {
+ **prompt.kwargs,
+ **prompt_args,
+ }
responses = self._vellum_client.generate_stream(
deployment_id=registered_prompt.deployment_id,
- requests=[
- GenerateRequest(input_values=prompt.get_full_format_args(prompt_args))
- ],
+ requests=[GenerateRequest(input_values=input_values)],
)
def text_generator() -> TokenGen:
@@ -135,7 +139,7 @@ class VellumPredictor(BaseLLMPredictor):
return text_generator()
- async def apredict(self, prompt: Prompt, **prompt_args: Any) -> str:
+ async def apredict(self, prompt: BasePromptTemplate, **prompt_args: Any) -> str:
"""Asynchronously predict the answer to a query."""
from vellum import GenerateRequest
@@ -144,11 +148,13 @@ class VellumPredictor(BaseLLMPredictor):
prompt, **prompt_args
)
+ input_values = {
+ **prompt.kwargs,
+ **prompt_args,
+ }
result = await self._async_vellum_client.generate(
deployment_id=registered_prompt.deployment_id,
- requests=[
- GenerateRequest(input_values=prompt.get_full_format_args(prompt_args))
- ],
+ requests=[GenerateRequest(input_values=input_values)],
)
completion_text = self._process_generate_response(
@@ -157,7 +163,9 @@ class VellumPredictor(BaseLLMPredictor):
return completion_text
- async def astream(self, prompt: Prompt, **prompt_args: Any) -> TokenAsyncGen:
+ async def astream(
+ self, prompt: BasePromptTemplate, **prompt_args: Any
+ ) -> TokenAsyncGen:
async def gen() -> TokenAsyncGen:
for token in self.stream(prompt, **prompt_args):
yield token
@@ -166,7 +174,7 @@ class VellumPredictor(BaseLLMPredictor):
return gen()
def _prepare_generate_call(
- self, prompt: Prompt, **prompt_args: Any
+ self, prompt: BasePromptTemplate, **prompt_args: Any
) -> Tuple[VellumRegisteredPrompt, VellumCompiledPrompt, str]:
"""Prepare a generate call."""
diff --git a/llama_index/llm_predictor/vellum/prompt_registry.py b/llama_index/llm_predictor/vellum/prompt_registry.py
index 4421c75a92..ed7b9cb155 100644
--- a/llama_index/llm_predictor/vellum/prompt_registry.py
+++ b/llama_index/llm_predictor/vellum/prompt_registry.py
@@ -3,12 +3,13 @@ from __future__ import annotations
from typing import TYPE_CHECKING, Any, List, Tuple
from uuid import uuid4
-from llama_index import Prompt
from llama_index.llm_predictor.vellum.types import (
VellumCompiledPrompt,
VellumRegisteredPrompt,
)
from llama_index.llm_predictor.vellum.utils import convert_to_kebab_case
+from llama_index.prompts import BasePromptTemplate
+from llama_index.prompts.base import PromptTemplate
if TYPE_CHECKING:
import vellum # noqa: F401
@@ -33,7 +34,7 @@ class VellumPromptRegistry:
self._vellum_client = Vellum(api_key=vellum_api_key)
- def from_prompt(self, initial_prompt: Prompt) -> VellumRegisteredPrompt:
+ def from_prompt(self, initial_prompt: BasePromptTemplate) -> VellumRegisteredPrompt:
"""Accepts a LlamaIndex prompt and retrieves a corresponding registered prompt
from Vellum.
@@ -46,7 +47,7 @@ class VellumPromptRegistry:
You can reference a previously registered prompt by providing either
`vellum_deployment_id` or `vellum_deployment_name` as key/value pairs within
- `Prompt.metadata`.
+ `BasePromptTemplate.metadata`.
"""
from vellum.core import ApiError
@@ -112,7 +113,7 @@ class VellumPromptRegistry:
prompt_id=prompt_id,
)
- def _register_prompt(self, prompt: Prompt) -> VellumRegisteredPrompt:
+ def _register_prompt(self, prompt: BasePromptTemplate) -> VellumRegisteredPrompt:
"""Registers a prompt with Vellum.
By registering a prompt, Vellum will:
@@ -151,7 +152,7 @@ class VellumPromptRegistry:
parameters=params,
meta={
"source": "llamaindex",
- "prompt_type": prompt.prompt_type,
+ "prompt_type": prompt.metadata["prompt_type"],
},
)
@@ -164,22 +165,24 @@ class VellumPromptRegistry:
prompt_id=resp.prompt.id,
)
- def _generate_default_label(self, prompt: Prompt) -> str:
- return f"LlamaIndex Demo: {prompt.prompt_type}"
+ def _generate_default_label(self, prompt: BasePromptTemplate) -> str:
+ prompt_type = prompt.metadata["prompt_type"]
+ return f"LlamaIndex Demo: {prompt_type}'"
- def _generate_default_name(self, prompt: Prompt) -> str:
+ def _generate_default_name(self, prompt: BasePromptTemplate) -> str:
default_label = self._generate_default_label(prompt)
return convert_to_kebab_case(default_label)
def _construct_prompt_info(
- self, prompt: Prompt, for_chat_model: bool = True
+ self, prompt: BasePromptTemplate, for_chat_model: bool = True
) -> vellum.RegisterPromptPromptInfoRequest:
"""Converts a LlamaIndex prompt into Vellum's prompt representation."""
import vellum
- prompt_template = prompt.original_template
- for input_variable in prompt.get_langchain_prompt().input_variables:
+ assert isinstance(prompt, PromptTemplate)
+ prompt_template = prompt.template
+ for input_variable in prompt.template_vars:
prompt_template = prompt_template.replace(
input_variable, f"{{ {input_variable} }}"
)
@@ -190,8 +193,8 @@ class VellumPromptRegistry:
block_type=vellum.BlockTypeEnum.JINJA,
properties=vellum.PromptTemplateBlockPropertiesRequest(
template=self._prepare_prompt_jinja_template(
- prompt.original_template,
- prompt.get_langchain_prompt().input_variables,
+ prompt.template,
+ prompt.template_vars,
),
),
)
@@ -213,10 +216,7 @@ class VellumPromptRegistry:
version=1,
blocks=[block],
),
- input_variables=[
- {"key": input_var}
- for input_var in prompt.get_langchain_prompt().input_variables
- ],
+ input_variables=[{"key": input_var} for input_var in prompt.template_vars],
)
def _prepare_prompt_jinja_template(
diff --git a/llama_index/llms/generic_utils.py b/llama_index/llms/generic_utils.py
index 6ee3ab5086..0d09257a4b 100644
--- a/llama_index/llms/generic_utils.py
+++ b/llama_index/llms/generic_utils.py
@@ -1,4 +1,4 @@
-from typing import Any, Awaitable, Callable, Sequence
+from typing import Any, Awaitable, Callable, List, Sequence
from llama_index.llms.base import (
ChatMessage,
@@ -44,7 +44,7 @@ def messages_to_prompt(messages: Sequence[ChatMessage]) -> str:
return "\n".join(string_messages)
-def prompt_to_messages(prompt: str) -> Sequence[ChatMessage]:
+def prompt_to_messages(prompt: str) -> List[ChatMessage]:
"""Convert a string prompt to a sequence of messages."""
return [ChatMessage(role=MessageRole.USER, content=prompt)]
diff --git a/llama_index/llms/palm.py b/llama_index/llms/palm.py
index 567ca29ae1..68ba570355 100644
--- a/llama_index/llms/palm.py
+++ b/llama_index/llms/palm.py
@@ -80,7 +80,7 @@ class PaLM(CustomLLM):
"""Predict the answer to a query.
Args:
- prompt (Prompt): Prompt to use for prediction.
+ prompt (str): Prompt to use for prediction.
Returns:
Tuple[str, str]: Tuple of the predicted answer and the formatted prompt.
@@ -104,7 +104,7 @@ class PaLM(CustomLLM):
better abstractions about response handling.
Args:
- prompt (Prompt): Prompt to use for prediction.
+ prompt (str): Prompt to use for prediction.
Returns:
str: The predicted answer.
diff --git a/llama_index/node_parser/extractors/metadata_extractors.py b/llama_index/node_parser/extractors/metadata_extractors.py
index 06c8f40073..3b23744880 100644
--- a/llama_index/node_parser/extractors/metadata_extractors.py
+++ b/llama_index/node_parser/extractors/metadata_extractors.py
@@ -21,14 +21,15 @@ disambiguate the document or subsection from other similar documents or subsecti
"""
import json
from abc import abstractmethod
-from pydantic import Field, PrivateAttr
-from typing import Any, List, Optional, Sequence, cast, Dict, Callable
from functools import reduce
+from typing import Any, Callable, Dict, List, Optional, Sequence, cast
+
+from pydantic import Field, PrivateAttr
-from llama_index.llms.base import LLM
from llama_index.llm_predictor.base import BaseLLMPredictor, LLMPredictor
+from llama_index.llms.base import LLM
from llama_index.node_parser.interface import BaseExtractor
-from llama_index.prompts.base import Prompt
+from llama_index.prompts import PromptTemplate
from llama_index.schema import BaseNode, TextNode
@@ -199,7 +200,7 @@ class TitleExtractor(MetadataFeatureExtractor):
title_candidates = [
self.llm_predictor.predict(
- Prompt(template=self.node_template),
+ PromptTemplate(template=self.node_template),
context_str=cast(TextNode, node).text,
)
for node in nodes_to_extract_title
@@ -210,7 +211,7 @@ class TitleExtractor(MetadataFeatureExtractor):
)
title = self.llm_predictor.predict(
- Prompt(template=self.combine_template),
+ PromptTemplate(template=self.combine_template),
context_str=titles,
)
else:
@@ -257,7 +258,7 @@ class KeywordExtractor(MetadataFeatureExtractor):
# TODO: figure out a good way to allow users to customize keyword template
keywords = self.llm_predictor.predict(
- Prompt(
+ PromptTemplate(
template=f"""\
{{context_str}}. Give {self.keywords} unique keywords for this \
document. Format as comma separated. Keywords: """
@@ -322,7 +323,7 @@ class QuestionsAnsweredExtractor(MetadataFeatureExtractor):
# Extract the title from the first node
# TODO: figure out a good way to allow users to customize template
questions = self.llm_predictor.predict(
- Prompt(
+ PromptTemplate(
template=self.prompt_template
or f"""\
{{context_str}}. Given the contextual information, \
@@ -399,7 +400,7 @@ class SummaryExtractor(MetadataFeatureExtractor):
raise ValueError("Only `TextNode` is allowed for `Summary` extractor")
node_summaries = [
self.llm_predictor.predict(
- Prompt(template=self.prompt_template),
+ PromptTemplate(template=self.prompt_template),
context_str=cast(TextNode, node).text,
).strip()
for node in nodes
diff --git a/llama_index/program/llm_program.py b/llama_index/program/llm_program.py
index 96c71f8c10..ca19072dfd 100644
--- a/llama_index/program/llm_program.py
+++ b/llama_index/program/llm_program.py
@@ -1,12 +1,12 @@
from typing import Any, Dict, Optional, Type, Union
-
from pydantic import BaseModel
+
from llama_index.llms.base import LLM
from llama_index.llms.openai import OpenAI
-from llama_index.program.base_program import BasePydanticProgram
-from llama_index.prompts.base import Prompt
from llama_index.output_parsers.pydantic import PydanticOutputParser
+from llama_index.program.base_program import BasePydanticProgram
+from llama_index.prompts.base import PromptTemplate
class LLMTextCompletionProgram(BasePydanticProgram[BaseModel]):
@@ -20,7 +20,7 @@ class LLMTextCompletionProgram(BasePydanticProgram[BaseModel]):
def __init__(
self,
output_parser: PydanticOutputParser,
- prompt: Prompt,
+ prompt: PromptTemplate,
llm: LLM,
function_call: Union[str, Dict[str, Any]],
verbose: bool = False,
@@ -42,7 +42,7 @@ class LLMTextCompletionProgram(BasePydanticProgram[BaseModel]):
**kwargs: Any,
) -> "LLMTextCompletionProgram":
llm = llm or OpenAI(temperature=0, model="gpt-3.5-turbo-0613")
- prompt = Prompt(prompt_template_str)
+ prompt = PromptTemplate(prompt_template_str)
function_call = function_call or {
"name": output_parser.output_cls.schema()["title"]
}
@@ -64,9 +64,9 @@ class LLMTextCompletionProgram(BasePydanticProgram[BaseModel]):
**kwargs: Any,
) -> BaseModel:
prompt_with_parse_instrs_tmpl = self._output_parser.format(
- self._prompt.original_template
+ self._prompt.template
)
- prompt_with_parse_instrs = Prompt(prompt_with_parse_instrs_tmpl)
+ prompt_with_parse_instrs = PromptTemplate(prompt_with_parse_instrs_tmpl)
formatted_prompt = prompt_with_parse_instrs.format(**kwargs)
diff --git a/llama_index/program/openai_program.py b/llama_index/program/openai_program.py
index 3e06c55410..5dee0e8160 100644
--- a/llama_index/program/openai_program.py
+++ b/llama_index/program/openai_program.py
@@ -6,7 +6,7 @@ from llama_index.llms.base import LLM, ChatMessage, MessageRole
from llama_index.llms.openai import OpenAI
from llama_index.llms.openai_utils import to_openai_function
from llama_index.program.llm_prompt_program import BaseLLMFunctionProgram
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import BasePromptTemplate, PromptTemplate
from llama_index.types import Model
from llama_index.program.utils import create_list_model
from typing import Tuple
@@ -46,7 +46,7 @@ class OpenAIPydanticProgram(BaseLLMFunctionProgram[LLM]):
self,
output_cls: Type[Model],
llm: LLM,
- prompt: Prompt,
+ prompt: BasePromptTemplate,
function_call: Union[str, Dict[str, Any]],
verbose: bool = False,
) -> None:
@@ -75,7 +75,7 @@ class OpenAIPydanticProgram(BaseLLMFunctionProgram[LLM]):
"function calling API. "
)
- prompt = Prompt(prompt_template_str)
+ prompt = PromptTemplate(prompt_template_str)
function_call = function_call or _default_function_call(output_cls)
return cls(
output_cls=output_cls,
diff --git a/llama_index/program/predefined/df.py b/llama_index/program/predefined/df.py
index 9041828830..eb6f4a6402 100644
--- a/llama_index/program/predefined/df.py
+++ b/llama_index/program/predefined/df.py
@@ -3,7 +3,6 @@ from llama_index.program.base_program import BasePydanticProgram
from typing import Optional, List, Any, Type, cast
from pydantic import BaseModel, Field
from llama_index.program.openai_program import OpenAIPydanticProgram
-from llama_index.prompts.prompts import Prompt
from llama_index.program.llm_prompt_program import BaseLLMFunctionProgram
import pandas as pd
@@ -162,16 +161,12 @@ class DFRowsProgram(BasePydanticProgram[DataFrameRowsOnly]):
) -> None:
"""Init params."""
# partial format df parser template string with column schema
- # NOTE: hack where we use prompt class to partial format
- orig_prompt = Prompt(df_parser_template_str)
- new_prompt = Prompt.from_prompt(
- orig_prompt.partial_format(
- column_schema=column_schema,
- )
+ prompt_template_str = df_parser_template_str.replace(
+ "{column_schema}", column_schema or ""
)
pydantic_program = pydantic_program_cls.from_defaults(
- DataFrameRowsOnly, new_prompt.original_template, **program_kwargs
+ DataFrameRowsOnly, prompt_template_str, **program_kwargs
)
self._validate_program(pydantic_program)
self._pydantic_program = pydantic_program
diff --git a/llama_index/program/predefined/evaporate/extractor.py b/llama_index/program/predefined/evaporate/extractor.py
index 62702e173c..a42ae8d2e1 100644
--- a/llama_index/program/predefined/evaporate/extractor.py
+++ b/llama_index/program/predefined/evaporate/extractor.py
@@ -15,7 +15,6 @@ from llama_index.schema import BaseNode, MetadataMode, NodeWithScore
from llama_index.indices.query.schema import QueryBundle
-from llama_index.prompts.prompts import QuestionAnswerPrompt
from llama_index.program.predefined.evaporate.prompts import (
@@ -177,12 +176,11 @@ class EvaporateExtractor:
else:
expected_output_str = ""
- new_prompt = self._fn_generate_prompt.partial_format(
+ qa_prompt = self._fn_generate_prompt.partial_format(
attribute=field,
function_field=function_field,
expected_output_str=expected_output_str,
)
- qa_prompt = QuestionAnswerPrompt.from_prompt(new_prompt)
response_synthesizer = get_response_synthesizer(
service_context=self._service_context,
diff --git a/llama_index/program/predefined/evaporate/prompts.py b/llama_index/program/predefined/evaporate/prompts.py
index 8aa7bdb403..7e7351c8e5 100644
--- a/llama_index/program/predefined/evaporate/prompts.py
+++ b/llama_index/program/predefined/evaporate/prompts.py
@@ -6,32 +6,32 @@ Full credits go to: https://github.com/HazyResearch/evaporate
"""
-from llama_index.prompts.prompts import Prompt
+from llama_index.prompts import PromptTemplate
# deprecated, kept for backward compatibility
-"""Pandas prompt. Convert query to python code.
+"""Pandas PromptTemplate. Convert query to python code.
Required template variables: `chunk`, `topic`.
Args:
- template (str): Template for the prompt.
- **prompt_kwargs: Keyword arguments for the prompt.
+ template (str): Template for the PromptTemplate.
+ **prompt_kwargs: Keyword arguments for the PromptTemplate.
"""
-SchemaIDPrompt = Prompt
+SchemaIDPrompt = PromptTemplate
-"""Function generation prompt. Generate a function from existing text.
+"""Function generation PromptTemplate. Generate a function from existing text.
Required template variables: `context_str`, `query_str`,
`attribute`, `function_field`.
Args:
- template (str): Template for the prompt.
- **prompt_kwargs: Keyword arguments for the prompt.
+ template (str): Template for the PromptTemplate.
+ **prompt_kwargs: Keyword arguments for the PromptTemplate.
"""
-FnGeneratePrompt = Prompt
+FnGeneratePrompt = PromptTemplate
# used for schema identification
SCHEMA_ID_PROMPT_TMPL = f"""Sample text:
@@ -81,7 +81,7 @@ Sample text:
Question: List all relevant attributes about '{{topic:}}' that are exactly mentioned in this sample text if any.
Answer:""" # noqa: E501, F541
-SCHEMA_ID_PROMPT = Prompt(SCHEMA_ID_PROMPT_TMPL)
+SCHEMA_ID_PROMPT = PromptTemplate(SCHEMA_ID_PROMPT_TMPL)
# used for function generation
@@ -108,7 +108,7 @@ def get_{{function_field:}}_field(text: str):
\"""
""" # noqa: E501, F541
-FN_GENERATION_PROMPT = Prompt(FN_GENERATION_PROMPT_TMPL)
+FN_GENERATION_PROMPT = PromptTemplate(FN_GENERATION_PROMPT_TMPL)
FN_GENERATION_LIST_PROMPT_TMPL = f"""Here is a sample of text:
@@ -134,7 +134,7 @@ def get_{{function_field:}}_field(text: str) -> List:
\"""
""" # noqa: E501, F541
-FN_GENERATION_LIST_PROMPT = Prompt(FN_GENERATION_LIST_PROMPT_TMPL)
+FN_GENERATION_LIST_PROMPT = PromptTemplate(FN_GENERATION_LIST_PROMPT_TMPL)
DEFAULT_EXPECTED_OUTPUT_PREFIX_TMPL = (
"Here is the expected output on the text after running the function. "
diff --git a/llama_index/prompts/__init__.py b/llama_index/prompts/__init__.py
index 872453c72e..780c2b7cd9 100644
--- a/llama_index/prompts/__init__.py
+++ b/llama_index/prompts/__init__.py
@@ -1,5 +1,25 @@
"""Prompt class."""
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import (
+ BasePromptTemplate,
+ ChatPromptTemplate,
+ LangchainPromptTemplate,
+ Prompt,
+ PromptTemplate,
+ PromptType,
+ SelectorPromptTemplate,
+)
-__all__ = ["Prompt"]
+from llama_index.llms.base import ChatMessage, MessageRole
+
+__all__ = [
+ "Prompt",
+ "PromptTemplate",
+ "SelectorPromptTemplate",
+ "ChatPromptTemplate",
+ "LangchainPromptTemplate",
+ "BasePromptTemplate",
+ "PromptType",
+ "ChatMessage",
+ "MessageRole",
+]
diff --git a/llama_index/prompts/base.py b/llama_index/prompts/base.py
index 3911404ea5..3c4d5f1e46 100644
--- a/llama_index/prompts/base.py
+++ b/llama_index/prompts/base.py
@@ -1,173 +1,295 @@
-"""Base module for prompts."""
+"""Prompts."""
+
+
+from abc import ABC, abstractmethod
from copy import deepcopy
-from typing import Any, Dict, List, Optional
+from typing import Any, Callable, Dict, List, Optional, Tuple
+
+from pydantic import BaseModel
-from llama_index.bridge.langchain import BasePromptTemplate as BaseLangchainPrompt
-from llama_index.bridge.langchain import PromptTemplate as LangchainPrompt
+from llama_index.bridge.langchain import BasePromptTemplate as LangchainTemplate
+from llama_index.bridge.langchain import ConditionalPromptSelector as LangchainSelector
from llama_index.llms.base import LLM, ChatMessage
+from llama_index.llms.generic_utils import messages_to_prompt, prompt_to_messages
+from llama_index.llms.langchain import LangChainLLM
from llama_index.llms.langchain_utils import from_lc_messages
-from llama_index.prompts.prompt_selector import PromptSelector
from llama_index.prompts.prompt_type import PromptType
+from llama_index.prompts.utils import get_template_vars
from llama_index.types import BaseOutputParser
-class Prompt:
- """Prompt class for LlamaIndex.
+class BasePromptTemplate(BaseModel, ABC):
+ metadata: Dict[str, Any]
+ template_vars: List[str]
+ kwargs: Dict[str, str]
+ output_parser: Optional[BaseOutputParser]
+
+ class Config:
+ arbitrary_types_allowed = True
+
+ @abstractmethod
+ def partial_format(self, **kwargs: Any) -> "BasePromptTemplate":
+ ...
+
+ @abstractmethod
+ def format(self, llm: Optional[LLM] = None, **kwargs: Any) -> str:
+ ...
+
+ @abstractmethod
+ def format_messages(
+ self, llm: Optional[LLM] = None, **kwargs: Any
+ ) -> List[ChatMessage]:
+ ...
+
+
+class PromptTemplate(BasePromptTemplate):
+ template: str
+
+ def __init__(
+ self,
+ template: str,
+ prompt_type: str = PromptType.CUSTOM,
+ output_parser: Optional[BaseOutputParser] = None,
+ metadata: Optional[Dict[str, Any]] = None,
+ **kwargs: Any,
+ ) -> None:
+ if metadata is None:
+ metadata = {}
+ metadata["prompt_type"] = prompt_type
+
+ template_vars = get_template_vars(template)
+
+ super().__init__(
+ template=template,
+ template_vars=template_vars,
+ kwargs=kwargs,
+ metadata=metadata,
+ output_parser=output_parser,
+ )
+
+ def partial_format(self, **kwargs: Any) -> "PromptTemplate":
+ """Partially format the prompt."""
+ prompt = deepcopy(self)
+ prompt.kwargs.update(kwargs)
+ return prompt
+
+ def format(self, llm: Optional[LLM] = None, **kwargs: Any) -> str:
+ """Format the prompt into a string."""
+ del llm # unused
+ all_kwargs = {
+ **self.kwargs,
+ **kwargs,
+ }
+ return self.template.format(**all_kwargs)
+
+ def format_messages(
+ self, llm: Optional[LLM] = None, **kwargs: Any
+ ) -> List[ChatMessage]:
+ """Format the prompt into a list of chat messages."""
+ del llm # unused
+ prompt = self.format(**kwargs)
+ return prompt_to_messages(prompt)
+
+
+class ChatPromptTemplate(BasePromptTemplate):
+ message_templates: List[ChatMessage]
+
+ def __init__(
+ self,
+ message_templates: List[ChatMessage],
+ prompt_type: str = PromptType.CUSTOM,
+ output_parser: Optional[BaseOutputParser] = None,
+ metadata: Optional[Dict[str, Any]] = None,
+ **kwargs: Any,
+ ):
+ if metadata is None:
+ metadata = {}
+ metadata["prompt_type"] = prompt_type
+
+ template_vars = []
+ for message_template in message_templates:
+ template_vars.extend(get_template_vars(message_template.content or ""))
+
+ super().__init__(
+ message_templates=message_templates,
+ kwargs=kwargs,
+ metadata=metadata,
+ output_parser=output_parser,
+ template_vars=template_vars,
+ )
+
+ def partial_format(self, **kwargs: Any) -> "ChatPromptTemplate":
+ prompt = deepcopy(self)
+ prompt.kwargs.update(kwargs)
+ return prompt
+
+ def format(self, llm: Optional[LLM] = None, **kwargs: Any) -> str:
+ del llm # unused
+ messages = self.format_messages(**kwargs)
+ prompt = messages_to_prompt(messages)
+ return prompt
+
+ def format_messages(
+ self, llm: Optional[LLM] = None, **kwargs: Any
+ ) -> List[ChatMessage]:
+ del llm # unused
+ """Format the prompt into a list of chat messages."""
+ all_kwargs = {
+ **self.kwargs,
+ **kwargs,
+ }
+
+ messages = []
+ for message_template in self.message_templates:
+ template_vars = get_template_vars(message_template.content or "")
+ relevant_kwargs = {
+ k: v for k, v in all_kwargs.items() if k in template_vars
+ }
+ content_template = message_template.content or ""
+ content = content_template.format(**relevant_kwargs)
- Wrapper around langchain's prompt class. Adds ability to:
- - enforce certain prompt types
- - partially fill values
- """
+ message = message_template.copy()
+ message.content = content
+ messages.append(message)
+
+ return messages
+
+
+class SelectorPromptTemplate(BasePromptTemplate):
+ default_template: BasePromptTemplate
+ conditionals: Optional[
+ List[Tuple[Callable[[LLM], bool], BasePromptTemplate]]
+ ] = None
def __init__(
self,
- template: Optional[str] = None,
- langchain_prompt: Optional[BaseLangchainPrompt] = None,
- langchain_prompt_selector: Optional[PromptSelector] = None,
+ default_template: BasePromptTemplate,
+ conditionals: Optional[
+ List[Tuple[Callable[[LLM], bool], BasePromptTemplate]]
+ ] = None,
+ ):
+ metadata = default_template.metadata
+ kwargs = default_template.kwargs
+ template_vars = default_template.template_vars
+ output_parser = default_template.output_parser
+ super().__init__(
+ default_template=default_template,
+ conditionals=conditionals,
+ metadata=metadata,
+ kwargs=kwargs,
+ template_vars=template_vars,
+ output_parser=output_parser,
+ )
+
+ def _select(self, llm: Optional[LLM] = None) -> BasePromptTemplate:
+ if llm is None:
+ return self.default_template
+
+ if self.conditionals is not None:
+ for condition, prompt in self.conditionals:
+ if condition(llm):
+ return prompt
+ return self.default_template
+
+ def partial_format(self, **kwargs: Any) -> "SelectorPromptTemplate":
+ default_template = self.default_template.partial_format(**kwargs)
+ if self.conditionals is None:
+ conditionals = None
+ else:
+ conditionals = [
+ (condition, prompt.partial_format(**kwargs))
+ for condition, prompt in self.conditionals
+ ]
+ return SelectorPromptTemplate(
+ default_template=default_template, conditionals=conditionals
+ )
+
+ def format(self, llm: Optional[LLM] = None, **kwargs: Any) -> str:
+ """Format the prompt into a string."""
+ prompt = self._select(llm=llm)
+ return prompt.format(**kwargs)
+
+ def format_messages(
+ self, llm: Optional[LLM] = None, **kwargs: Any
+ ) -> List[ChatMessage]:
+ """Format the prompt into a list of chat messages."""
+ prompt = self._select(llm=llm)
+ return prompt.format_messages(**kwargs)
+
+
+class LangchainPromptTemplate(BasePromptTemplate):
+ selector: LangchainSelector
+
+ def __init__(
+ self,
+ template: Optional[LangchainTemplate] = None,
+ selector: Optional[LangchainSelector] = None,
output_parser: Optional[BaseOutputParser] = None,
prompt_type: str = PromptType.CUSTOM,
metadata: Optional[Dict[str, Any]] = None,
- **prompt_kwargs: Any,
) -> None:
- """Init params."""
- # first check if langchain_prompt_selector is provided
- # TODO: self.prompt is deprecated, switch to prompt_selector under the hood
- if langchain_prompt_selector is not None:
- self.prompt_selector = langchain_prompt_selector
- self.prompt: BaseLangchainPrompt = self.prompt_selector.default_prompt
- # then check if template is provided
- elif langchain_prompt is None:
+ if selector is None:
if template is None:
- raise ValueError(
- "`template` must be specified if `langchain_prompt` is None"
- )
-
- self.prompt = LangchainPrompt.from_template(
- template=template, **prompt_kwargs
- )
- self.prompt_selector = PromptSelector(default_prompt=self.prompt)
- # finally, check if langchain_prompt is provided
+ raise ValueError("Must provide either template or selector.")
+ selector = LangchainSelector(default_prompt=template)
else:
- if template:
- raise ValueError(
- f"Both template ({template}) and langchain_prompt "
- f"({langchain_prompt}) are provided, only one should be."
- )
- self.prompt = langchain_prompt
- self.prompt_selector = PromptSelector(default_prompt=self.prompt)
-
- self.partial_dict: Dict[str, Any] = {}
- self.prompt_kwargs = prompt_kwargs
- # NOTE: this is only used for token counting and testing
- self.prompt_type = prompt_type
-
- self.output_parser = output_parser
-
- self._original_template = template
-
- # Metadata is used to pass arbitrary information to other consumers of the
- # prompt. For example, VellumPromptRegistry uses this to access vellum-specific
- # identifiers that users can pass along with the prompt.
- self.metadata = metadata or {}
-
- @property
- def original_template(self) -> str:
- """Return the originally specified template, if supplied."""
-
- if not self._original_template:
- raise ValueError("No original template specified.")
-
- return self._original_template
-
- @classmethod
- def from_langchain_prompt(
- cls, prompt: BaseLangchainPrompt, **kwargs: Any
- ) -> "Prompt":
- """Load prompt from LangChain prompt."""
- return cls(langchain_prompt=prompt, **kwargs)
-
- @classmethod
- def from_langchain_prompt_selector(
- cls, prompt_selector: PromptSelector, **kwargs: Any
- ) -> "Prompt":
- """Load prompt from LangChain prompt."""
- return cls(langchain_prompt_selector=prompt_selector, **kwargs)
-
- def partial_format(self, **kwargs: Any) -> "Prompt":
- """Format the prompt partially.
-
- Return an instance of itself.
-
- """
- try:
- # NOTE: this is a hack to get around deepcopy failing on output parser
- output_parser = self.output_parser
- self.output_parser = None
-
- copy_obj = deepcopy(self)
- copy_obj.output_parser = output_parser
- copy_obj.partial_dict.update(kwargs)
- self.output_parser = output_parser
- except Exception as e:
- raise e
-
- return copy_obj
-
- @classmethod
- def from_prompt(
- cls,
- prompt: "Prompt",
- llm: Optional[LLM] = None,
- prompt_type: Optional[PromptType] = None,
- ) -> "Prompt":
- """Create a prompt from an existing prompt.
-
- Use case: If the existing prompt is already partially filled,
- and the remaining fields satisfy the requirements of the
- prompt class, then we can create a new prompt from the existing
- partially filled prompt.
-
- """
- lc_prompt = prompt.get_langchain_prompt(llm=llm)
- tmpl_vars = lc_prompt.input_variables
- format_dict = {}
- for var in tmpl_vars:
- if var not in prompt.partial_dict:
- format_dict[var] = f"{{{var}}}"
-
- template_str = prompt.format(llm=llm, **format_dict)
- cls_obj = cls(
- template_str,
- prompt_type=prompt_type or PromptType.CUSTOM,
- **prompt.prompt_kwargs,
+ if template is not None:
+ raise ValueError("Must provide either template or selector.")
+ selector = selector
+
+ kwargs = selector.default_prompt.partial_variables
+ template_vars = selector.default_prompt.input_variables
+
+ if metadata is None:
+ metadata = {}
+ metadata["prompt_type"] = prompt_type
+
+ super().__init__(
+ selector=selector,
+ metadata=metadata,
+ kwargs=kwargs,
+ template_vars=template_vars,
+ output_parser=output_parser,
)
- return cls_obj
- def get_langchain_prompt(self, llm: Optional[LLM] = None) -> BaseLangchainPrompt:
- """Get langchain prompt."""
- return self.prompt_selector.select(llm=llm)
+ def partial_format(self, **kwargs: Any) -> "BasePromptTemplate":
+ """Partially format the prompt."""
+ default_prompt = self.selector.default_prompt.partial(**kwargs)
+ conditionals = [
+ (condition, prompt.partial(**kwargs))
+ for condition, prompt in self.selector.conditionals
+ ]
+ lc_selector = LangchainSelector(
+ default_prompt=default_prompt, conditionals=conditionals
+ )
+ return LangchainPromptTemplate(selector=lc_selector)
def format(self, llm: Optional[LLM] = None, **kwargs: Any) -> str:
"""Format the prompt into a string."""
- kwargs.update(self.partial_dict)
- lc_prompt = self.get_langchain_prompt(llm=llm)
- return lc_prompt.format(**kwargs)
+ if llm is not None:
+ if not isinstance(llm, LangChainLLM):
+ raise ValueError("Must provide a LangChainLLM.")
+ lc_template = self.selector.get_prompt(llm=llm.llm)
+ else:
+ lc_template = self.selector.default_prompt
+
+ return lc_template.format(**kwargs)
def format_messages(
self, llm: Optional[LLM] = None, **kwargs: Any
) -> List[ChatMessage]:
"""Format the prompt into a list of chat messages."""
- kwargs.update(self.partial_dict)
- lc_template = self.get_langchain_prompt(llm=llm)
- lc_value = lc_template.format_prompt(**kwargs)
- lc_messages = lc_value.to_messages()
- return from_lc_messages(lc_messages)
-
- def get_full_format_args(self, kwargs: Dict) -> Dict[str, Any]:
- """Get dict of all format args.
+ if llm is not None:
+ if not isinstance(llm, LangChainLLM):
+ raise ValueError("Must provide a LangChainLLM.")
+ lc_template = self.selector.get_prompt(llm=llm.llm)
+ else:
+ lc_template = self.selector.default_prompt
+ lc_prompt_value = lc_template.format_prompt(**kwargs)
+ lc_messages = lc_prompt_value.to_messages()
+ messages = from_lc_messages(lc_messages)
+ return messages
- Hack to pass into Langchain to pass validation.
- """
- kwargs.update(self.partial_dict)
- return kwargs
+# NOTE: only for backwards compatibility
+Prompt = PromptTemplate
diff --git a/llama_index/prompts/chat_prompts.py b/llama_index/prompts/chat_prompts.py
index 51a3c0ca56..97d6732941 100644
--- a/llama_index/prompts/chat_prompts.py
+++ b/llama_index/prompts/chat_prompts.py
@@ -1,114 +1,109 @@
"""Prompts for ChatGPT."""
-from llama_index.bridge.langchain import (
- AIMessagePromptTemplate,
- ChatPromptTemplate,
- HumanMessagePromptTemplate,
- SystemMessagePromptTemplate,
-)
-
-from llama_index.prompts.prompts import (
- QuestionAnswerPrompt,
- SummaryPrompt,
- RefinePrompt,
- RefineTableContextPrompt,
-)
+from llama_index.llms.base import ChatMessage, MessageRole
+from llama_index.prompts.base import ChatPromptTemplate
# text qa prompt
-TEXT_QA_SYSTEM_PROMPT = SystemMessagePromptTemplate.from_template(
- "You are an expert Q&A system that is trusted around the world.\n"
- "Always answer the query using the provided context information, "
- "and not prior knowledge.\n"
- "Some rules to follow:\n"
- "1. Never directly reference the given context in your answer.\n"
- "2. Avoid statements like 'Based on the context, ...' or "
- "'The context information ...' or anything along "
- "those lines."
+TEXT_QA_SYSTEM_PROMPT = ChatMessage(
+ content=(
+ "You are an expert Q&A system that is trusted around the world.\n"
+ "Always answer the query using the provided context information, "
+ "and not prior knowledge.\n"
+ "Some rules to follow:\n"
+ "1. Never directly reference the given context in your answer.\n"
+ "2. Avoid statements like 'Based on the context, ...' or "
+ "'The context information ...' or anything along "
+ "those lines."
+ ),
+ role=MessageRole.SYSTEM,
)
TEXT_QA_PROMPT_TMPL_MSGS = [
TEXT_QA_SYSTEM_PROMPT,
- HumanMessagePromptTemplate.from_template(
- "Context information is below.\n"
- "---------------------\n"
- "{context_str}\n"
- "---------------------\n"
- "Given the context information and not prior knowledge, "
- "answer the query.\n"
- "Query: {query_str}\n"
- "Answer: "
+ ChatMessage(
+ content=(
+ "Context information is below.\n"
+ "---------------------\n"
+ "{context_str}\n"
+ "---------------------\n"
+ "Given the context information and not prior knowledge, "
+ "answer the query.\n"
+ "Query: {query_str}\n"
+ "Answer: "
+ ),
+ role=MessageRole.USER,
),
]
-CHAT_TEXT_QA_PROMPT_LC = ChatPromptTemplate.from_messages(TEXT_QA_PROMPT_TMPL_MSGS)
-CHAT_TEXT_QA_PROMPT = QuestionAnswerPrompt.from_langchain_prompt(CHAT_TEXT_QA_PROMPT_LC)
-
+CHAT_TEXT_QA_PROMPT = ChatPromptTemplate(message_templates=TEXT_QA_PROMPT_TMPL_MSGS)
# Tree Summarize
TREE_SUMMARIZE_PROMPT_TMPL_MSGS = [
TEXT_QA_SYSTEM_PROMPT,
- HumanMessagePromptTemplate.from_template(
- "Context information from multiple sources is below.\n"
- "---------------------\n"
- "{context_str}\n"
- "---------------------\n"
- "Given the information from multiple sources and not prior knowledge, "
- "answer the query.\n"
- "Query: {query_str}\n"
- "Answer: "
+ ChatMessage(
+ content=(
+ "Context information from multiple sources is below.\n"
+ "---------------------\n"
+ "{context_str}\n"
+ "---------------------\n"
+ "Given the information from multiple sources and not prior knowledge, "
+ "answer the query.\n"
+ "Query: {query_str}\n"
+ "Answer: "
+ ),
+ role=MessageRole.USER,
),
]
-CHAT_TREE_SUMMARIZE_PROMPT_LC = ChatPromptTemplate.from_messages(
- TREE_SUMMARIZE_PROMPT_TMPL_MSGS
-)
-CHAT_TREE_SUMMARIZE_PROMPT = SummaryPrompt.from_langchain_prompt(
- CHAT_TREE_SUMMARIZE_PROMPT_LC
+CHAT_TREE_SUMMARIZE_PROMPT = ChatPromptTemplate(
+ message_templates=TREE_SUMMARIZE_PROMPT_TMPL_MSGS
)
# Refine Prompt
CHAT_REFINE_PROMPT_TMPL_MSGS = [
- HumanMessagePromptTemplate.from_template(
- "You are an expert Q&A system that stricly operates in two modes"
- "when refining existing answers:\n"
- "1. **Rewrite** an original answer using the new context.\n"
- "2. **Repeat** the original answer if the new context isn't useful.\n"
- "Never reference the original answer or context directly in your answer.\n"
- "When in doubt, just repeat the original answer."
- "New Context: {context_msg}\n"
- "Query: {query_str}\n"
- "Original Answer: {existing_answer}\n"
- "New Answer: "
- ),
+ ChatMessage(
+ content=(
+ "You are an expert Q&A system that stricly operates in two modes"
+ "when refining existing answers:\n"
+ "1. **Rewrite** an original answer using the new context.\n"
+ "2. **Repeat** the original answer if the new context isn't useful.\n"
+ "Never reference the original answer or context directly in your answer.\n"
+ "When in doubt, just repeat the original answer."
+ "New Context: {context_msg}\n"
+ "Query: {query_str}\n"
+ "Original Answer: {existing_answer}\n"
+ "New Answer: "
+ ),
+ role=MessageRole.USER,
+ )
]
-CHAT_REFINE_PROMPT_LC = ChatPromptTemplate.from_messages(CHAT_REFINE_PROMPT_TMPL_MSGS)
-CHAT_REFINE_PROMPT = RefinePrompt.from_langchain_prompt(CHAT_REFINE_PROMPT_LC)
+CHAT_REFINE_PROMPT = ChatPromptTemplate(message_templates=CHAT_REFINE_PROMPT_TMPL_MSGS)
# Table Context Refine Prompt
CHAT_REFINE_TABLE_CONTEXT_TMPL_MSGS = [
- HumanMessagePromptTemplate.from_template("{query_str}"),
- AIMessagePromptTemplate.from_template("{existing_answer}"),
- HumanMessagePromptTemplate.from_template(
- "We have provided a table schema below. "
- "---------------------\n"
- "{schema}\n"
- "---------------------\n"
- "We have also provided some context information below. "
- "{context_msg}\n"
- "---------------------\n"
- "Given the context information and the table schema, "
- "refine the original answer to better "
- "answer the question. "
- "If the context isn't useful, return the original answer."
+ ChatMessage(content="{query_str}", role=MessageRole.USER),
+ ChatMessage(content="{existing_answer}", role=MessageRole.ASSISTANT),
+ ChatMessage(
+ content=(
+ "We have provided a table schema below. "
+ "---------------------\n"
+ "{schema}\n"
+ "---------------------\n"
+ "We have also provided some context information below. "
+ "{context_msg}\n"
+ "---------------------\n"
+ "Given the context information and the table schema, "
+ "refine the original answer to better "
+ "answer the question. "
+ "If the context isn't useful, return the original answer."
+ ),
+ role=MessageRole.USER,
),
]
-CHAT_REFINE_TABLE_CONTEXT_PROMPT_LC = ChatPromptTemplate.from_messages(
- CHAT_REFINE_TABLE_CONTEXT_TMPL_MSGS
-)
-CHAT_REFINE_TABLE_CONTEXT_PROMPT = RefineTableContextPrompt.from_langchain_prompt(
- CHAT_REFINE_TABLE_CONTEXT_PROMPT_LC
+CHAT_REFINE_TABLE_CONTEXT_PROMPT = ChatPromptTemplate(
+ message_templates=CHAT_REFINE_TABLE_CONTEXT_TMPL_MSGS
)
diff --git a/llama_index/prompts/choice_select.py b/llama_index/prompts/choice_select.py
deleted file mode 100644
index b03af6c021..0000000000
--- a/llama_index/prompts/choice_select.py
+++ /dev/null
@@ -1,34 +0,0 @@
-"""Default choice select prompt."""
-
-from llama_index.prompts.base import Prompt
-from llama_index.prompts.prompt_type import PromptType
-
-# deprecated, kept for backward compatibility
-ChoiceSelectPrompt = Prompt
-
-DEFAULT_CHOICE_SELECT_PROMPT_TMPL = (
- "A list of documents is shown below. Each document has a number next to it along "
- "with a summary of the document. A question is also provided. \n"
- "Respond with the numbers of the documents "
- "you should consult to answer the question, in order of relevance, as well \n"
- "as the relevance score. The relevance score is a number from 1-10 based on "
- "how relevant you think the document is to the question.\n"
- "Do not include any documents that are not relevant to the question. \n"
- "Example format: \n"
- "Document 1:\n<summary of document 1>\n\n"
- "Document 2:\n<summary of document 2>\n\n"
- "...\n\n"
- "Document 10:\n<summary of document 10>\n\n"
- "Question: <question>\n"
- "Answer:\n"
- "Doc: 9, Relevance: 7\n"
- "Doc: 3, Relevance: 4\n"
- "Doc: 7, Relevance: 3\n\n"
- "Let's try this now: \n\n"
- "{context_str}\n"
- "Question: {query_str}\n"
- "Answer:\n"
-)
-DEFAULT_CHOICE_SELECT_PROMPT = Prompt(
- DEFAULT_CHOICE_SELECT_PROMPT_TMPL, prompt_type=PromptType.CHOICE_SELECT
-)
diff --git a/llama_index/prompts/default_prompt_selectors.py b/llama_index/prompts/default_prompt_selectors.py
index 17c0307f24..d11b026d3d 100644
--- a/llama_index/prompts/default_prompt_selectors.py
+++ b/llama_index/prompts/default_prompt_selectors.py
@@ -1,59 +1,35 @@
-"""Prompt selectors."""
+"""Default prompt selectors."""
+from llama_index.prompts import SelectorPromptTemplate
from llama_index.prompts.chat_prompts import (
- CHAT_TEXT_QA_PROMPT,
- CHAT_TREE_SUMMARIZE_PROMPT,
CHAT_REFINE_PROMPT,
CHAT_REFINE_TABLE_CONTEXT_PROMPT,
+ CHAT_TEXT_QA_PROMPT,
+ CHAT_TREE_SUMMARIZE_PROMPT,
)
from llama_index.prompts.default_prompts import (
- DEFAULT_TEXT_QA_PROMPT,
- DEFAULT_TREE_SUMMARIZE_PROMPT,
DEFAULT_REFINE_PROMPT,
DEFAULT_REFINE_TABLE_CONTEXT_PROMPT,
+ DEFAULT_TEXT_QA_PROMPT,
+ DEFAULT_TREE_SUMMARIZE_PROMPT,
)
-from llama_index.prompts.prompt_selector import PromptSelector, is_chat_model
-from llama_index.prompts.prompt_type import PromptType
-from llama_index.prompts.prompts import (
- QuestionAnswerPrompt,
- RefinePrompt,
- RefineTableContextPrompt,
-)
-
-DEFAULT_TEXT_QA_PROMPT_SEL_LC = PromptSelector(
- default_prompt=DEFAULT_TEXT_QA_PROMPT.get_langchain_prompt(),
- conditionals=[(is_chat_model, CHAT_TEXT_QA_PROMPT.get_langchain_prompt())],
-)
-DEFAULT_TEXT_QA_PROMPT_SEL = QuestionAnswerPrompt(
- langchain_prompt_selector=DEFAULT_TEXT_QA_PROMPT_SEL_LC,
- prompt_type=PromptType.QUESTION_ANSWER,
-)
+from llama_index.prompts.utils import is_chat_model
-DEFAULT_TREE_SUMMARIZE_PROMPT_SEL_LC = PromptSelector(
- default_prompt=DEFAULT_TREE_SUMMARIZE_PROMPT.get_langchain_prompt(),
- conditionals=[(is_chat_model, CHAT_TREE_SUMMARIZE_PROMPT.get_langchain_prompt())],
-)
-DEFAULT_TREE_SUMMARIZE_PROMPT_SEL = QuestionAnswerPrompt(
- langchain_prompt_selector=DEFAULT_TREE_SUMMARIZE_PROMPT_SEL_LC,
- prompt_type=PromptType.SUMMARY,
+DEFAULT_TEXT_QA_PROMPT_SEL = SelectorPromptTemplate(
+ default_template=DEFAULT_TEXT_QA_PROMPT,
+ conditionals=[(is_chat_model, CHAT_TEXT_QA_PROMPT)],
)
-DEFAULT_REFINE_PROMPT_SEL_LC = PromptSelector(
- default_prompt=DEFAULT_REFINE_PROMPT.get_langchain_prompt(),
- conditionals=[(is_chat_model, CHAT_REFINE_PROMPT.get_langchain_prompt())],
-)
-DEFAULT_REFINE_PROMPT_SEL = RefinePrompt(
- langchain_prompt_selector=DEFAULT_REFINE_PROMPT_SEL_LC,
- prompt_type=PromptType.REFINE,
+DEFAULT_TREE_SUMMARIZE_PROMPT_SEL = SelectorPromptTemplate(
+ default_template=DEFAULT_TREE_SUMMARIZE_PROMPT,
+ conditionals=[(is_chat_model, CHAT_TREE_SUMMARIZE_PROMPT)],
)
-DEFAULT_REFINE_TABLE_CONTEXT_PROMPT_SEL_LC = PromptSelector(
- default_prompt=DEFAULT_REFINE_TABLE_CONTEXT_PROMPT.get_langchain_prompt(),
- conditionals=[
- (is_chat_model, CHAT_REFINE_TABLE_CONTEXT_PROMPT.get_langchain_prompt())
- ],
+DEFAULT_REFINE_PROMPT_SEL = SelectorPromptTemplate(
+ default_template=DEFAULT_REFINE_PROMPT,
+ conditionals=[(is_chat_model, CHAT_REFINE_PROMPT)],
)
-DEFAULT_REFINE_TABLE_CONTEXT_PROMPT_SEL = RefineTableContextPrompt(
- langchain_prompt_selector=DEFAULT_REFINE_TABLE_CONTEXT_PROMPT_SEL_LC,
- prompt_type=PromptType.TABLE_CONTEXT,
+DEFAULT_REFINE_TABLE_CONTEXT_PROMPT_SEL = SelectorPromptTemplate(
+ default_template=DEFAULT_REFINE_TABLE_CONTEXT_PROMPT,
+ conditionals=[(is_chat_model, CHAT_REFINE_TABLE_CONTEXT_PROMPT)],
)
diff --git a/llama_index/prompts/default_prompts.py b/llama_index/prompts/default_prompts.py
index 56ea34a48f..41e583f230 100644
--- a/llama_index/prompts/default_prompts.py
+++ b/llama_index/prompts/default_prompts.py
@@ -1,6 +1,6 @@
"""Set of default prompts."""
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import PromptTemplate
from llama_index.prompts.prompt_type import PromptType
############################################
@@ -19,7 +19,7 @@ DEFAULT_SUMMARY_PROMPT_TMPL = (
'SUMMARY:"""\n'
)
-DEFAULT_SUMMARY_PROMPT = Prompt(
+DEFAULT_SUMMARY_PROMPT = PromptTemplate(
DEFAULT_SUMMARY_PROMPT_TMPL, prompt_type=PromptType.SUMMARY
)
@@ -37,7 +37,7 @@ DEFAULT_INSERT_PROMPT_TMPL = (
"The answer should be the number corresponding to the "
"summary that is most relevant to the question.\n"
)
-DEFAULT_INSERT_PROMPT = Prompt(
+DEFAULT_INSERT_PROMPT = PromptTemplate(
DEFAULT_INSERT_PROMPT_TMPL, prompt_type=PromptType.TREE_INSERT
)
@@ -55,7 +55,7 @@ DEFAULT_QUERY_PROMPT_TMPL = (
"Provide choice in the following format: 'ANSWER: <number>' and explain why "
"this summary was selected in relation to the question.\n"
)
-DEFAULT_QUERY_PROMPT = Prompt(
+DEFAULT_QUERY_PROMPT = PromptTemplate(
DEFAULT_QUERY_PROMPT_TMPL, prompt_type=PromptType.TREE_SELECT
)
@@ -73,7 +73,7 @@ DEFAULT_QUERY_PROMPT_MULTIPLE_TMPL = (
"Provide choices in the following format: 'ANSWER: <numbers>' and explain why "
"these summaries were selected in relation to the question.\n"
)
-DEFAULT_QUERY_PROMPT_MULTIPLE = Prompt(
+DEFAULT_QUERY_PROMPT_MULTIPLE = PromptTemplate(
DEFAULT_QUERY_PROMPT_MULTIPLE_TMPL, prompt_type=PromptType.TREE_SELECT_MULTIPLE
)
@@ -91,7 +91,7 @@ DEFAULT_REFINE_PROMPT_TMPL = (
"If the context isn't useful, return the original answer.\n"
"Refined Answer: "
)
-DEFAULT_REFINE_PROMPT = Prompt(
+DEFAULT_REFINE_PROMPT = PromptTemplate(
DEFAULT_REFINE_PROMPT_TMPL, prompt_type=PromptType.REFINE
)
@@ -106,7 +106,7 @@ DEFAULT_TEXT_QA_PROMPT_TMPL = (
"Query: {query_str}\n"
"Answer: "
)
-DEFAULT_TEXT_QA_PROMPT = Prompt(
+DEFAULT_TEXT_QA_PROMPT = PromptTemplate(
DEFAULT_TEXT_QA_PROMPT_TMPL, prompt_type=PromptType.QUESTION_ANSWER
)
@@ -120,7 +120,7 @@ DEFAULT_TREE_SUMMARIZE_TMPL = (
"Query: {query_str}\n"
"Answer: "
)
-DEFAULT_TREE_SUMMARIZE_PROMPT = Prompt(
+DEFAULT_TREE_SUMMARIZE_PROMPT = PromptTemplate(
DEFAULT_TREE_SUMMARIZE_TMPL, prompt_type=PromptType.SUMMARY
)
@@ -137,7 +137,7 @@ DEFAULT_KEYWORD_EXTRACT_TEMPLATE_TMPL = (
"---------------------\n"
"Provide keywords in the following comma-separated format: 'KEYWORDS: <keywords>'\n"
)
-DEFAULT_KEYWORD_EXTRACT_TEMPLATE = Prompt(
+DEFAULT_KEYWORD_EXTRACT_TEMPLATE = PromptTemplate(
DEFAULT_KEYWORD_EXTRACT_TEMPLATE_TMPL, prompt_type=PromptType.KEYWORD_EXTRACT
)
@@ -153,7 +153,7 @@ DEFAULT_QUERY_KEYWORD_EXTRACT_TEMPLATE_TMPL = (
"---------------------\n"
"Provide keywords in the following comma-separated format: 'KEYWORDS: <keywords>'\n"
)
-DEFAULT_QUERY_KEYWORD_EXTRACT_TEMPLATE = Prompt(
+DEFAULT_QUERY_KEYWORD_EXTRACT_TEMPLATE = PromptTemplate(
DEFAULT_QUERY_KEYWORD_EXTRACT_TEMPLATE_TMPL,
prompt_type=PromptType.QUERY_KEYWORD_EXTRACT,
)
@@ -179,7 +179,7 @@ DEFAULT_SCHEMA_EXTRACT_TMPL = (
"If no fields are present in the text, return a blank string.\n"
"Fields: "
)
-DEFAULT_SCHEMA_EXTRACT_PROMPT = Prompt(
+DEFAULT_SCHEMA_EXTRACT_PROMPT = PromptTemplate(
DEFAULT_SCHEMA_EXTRACT_TMPL, prompt_type=PromptType.SCHEMA_EXTRACT
)
@@ -208,7 +208,7 @@ DEFAULT_TEXT_TO_SQL_TMPL = (
"SQLQuery: "
)
-DEFAULT_TEXT_TO_SQL_PROMPT = Prompt(
+DEFAULT_TEXT_TO_SQL_PROMPT = PromptTemplate(
DEFAULT_TEXT_TO_SQL_TMPL,
prompt_type=PromptType.TEXT_TO_SQL,
)
@@ -238,11 +238,11 @@ DEFAULT_TABLE_CONTEXT_QUERY = (
"...\n\n"
)
-DEFAULT_TABLE_CONTEXT_PROMPT = Prompt(
+DEFAULT_TABLE_CONTEXT_PROMPT = PromptTemplate(
DEFAULT_TABLE_CONTEXT_TMPL, prompt_type=PromptType.TABLE_CONTEXT
)
-# NOTE: by partially filling schema, we can reduce to a RefinePrompt
+# NOTE: by partially filling schema, we can reduce to a refine prompt
# that we can feed to ur table
DEFAULT_REFINE_TABLE_CONTEXT_TMPL = (
"We have provided a table schema below. "
@@ -259,7 +259,7 @@ DEFAULT_REFINE_TABLE_CONTEXT_TMPL = (
"answer the question. "
"If the context isn't useful, return the original answer."
)
-DEFAULT_REFINE_TABLE_CONTEXT_PROMPT = Prompt(
+DEFAULT_REFINE_TABLE_CONTEXT_PROMPT = PromptTemplate(
DEFAULT_REFINE_TABLE_CONTEXT_TMPL, prompt_type=PromptType.TABLE_CONTEXT
)
@@ -285,7 +285,7 @@ DEFAULT_KG_TRIPLET_EXTRACT_TMPL = (
"Text: {text}\n"
"Triplets:\n"
)
-DEFAULT_KG_TRIPLET_EXTRACT_PROMPT = Prompt(
+DEFAULT_KG_TRIPLET_EXTRACT_PROMPT = PromptTemplate(
DEFAULT_KG_TRIPLET_EXTRACT_TMPL, prompt_type=PromptType.KNOWLEDGE_TRIPLET_EXTRACT
)
@@ -304,7 +304,7 @@ HYDE_TMPL = (
'Passage:"""\n'
)
-DEFAULT_HYDE_PROMPT = Prompt(HYDE_TMPL, prompt_type=PromptType.SUMMARY)
+DEFAULT_HYDE_PROMPT = PromptTemplate(HYDE_TMPL, prompt_type=PromptType.SUMMARY)
############################################
@@ -312,7 +312,7 @@ DEFAULT_HYDE_PROMPT = Prompt(HYDE_TMPL, prompt_type=PromptType.SUMMARY)
############################################
DEFAULT_SIMPLE_INPUT_TMPL = "{query_str}"
-DEFAULT_SIMPLE_INPUT_PROMPT = Prompt(
+DEFAULT_SIMPLE_INPUT_PROMPT = PromptTemplate(
DEFAULT_SIMPLE_INPUT_TMPL, prompt_type=PromptType.SIMPLE_INPUT
)
@@ -333,7 +333,9 @@ DEFAULT_PANDAS_TMPL = (
"Output:\n"
)
-DEFAULT_PANDAS_PROMPT = Prompt(DEFAULT_PANDAS_TMPL, prompt_type=PromptType.PANDAS)
+DEFAULT_PANDAS_PROMPT = PromptTemplate(
+ DEFAULT_PANDAS_TMPL, prompt_type=PromptType.PANDAS
+)
############################################
@@ -349,6 +351,38 @@ DEFAULT_JSON_PATH_TMPL = (
"JSONPath: "
)
-DEFAULT_JSON_PATH_PROMPT = Prompt(
+DEFAULT_JSON_PATH_PROMPT = PromptTemplate(
DEFAULT_JSON_PATH_TMPL, prompt_type=PromptType.JSON_PATH
)
+
+
+############################################
+# Choice Select
+############################################
+
+DEFAULT_CHOICE_SELECT_PROMPT_TMPL = (
+ "A list of documents is shown below. Each document has a number next to it along "
+ "with a summary of the document. A question is also provided. \n"
+ "Respond with the numbers of the documents "
+ "you should consult to answer the question, in order of relevance, as well \n"
+ "as the relevance score. The relevance score is a number from 1-10 based on "
+ "how relevant you think the document is to the question.\n"
+ "Do not include any documents that are not relevant to the question. \n"
+ "Example format: \n"
+ "Document 1:\n<summary of document 1>\n\n"
+ "Document 2:\n<summary of document 2>\n\n"
+ "...\n\n"
+ "Document 10:\n<summary of document 10>\n\n"
+ "Question: <question>\n"
+ "Answer:\n"
+ "Doc: 9, Relevance: 7\n"
+ "Doc: 3, Relevance: 4\n"
+ "Doc: 7, Relevance: 3\n\n"
+ "Let's try this now: \n\n"
+ "{context_str}\n"
+ "Question: {query_str}\n"
+ "Answer:\n"
+)
+DEFAULT_CHOICE_SELECT_PROMPT = PromptTemplate(
+ DEFAULT_CHOICE_SELECT_PROMPT_TMPL, prompt_type=PromptType.CHOICE_SELECT
+)
diff --git a/llama_index/prompts/prompt_selector.py b/llama_index/prompts/prompt_selector.py
deleted file mode 100644
index 36b02fe9d2..0000000000
--- a/llama_index/prompts/prompt_selector.py
+++ /dev/null
@@ -1,26 +0,0 @@
-from typing import Callable, List, Optional, Tuple
-
-from pydantic import BaseModel, Field
-
-from llama_index.bridge.langchain import BasePromptTemplate
-from llama_index.llms.base import LLM
-
-
-class PromptSelector(BaseModel):
- default_prompt: BasePromptTemplate
- conditionals: List[Tuple[Callable[[LLM], bool], BasePromptTemplate]] = Field(
- default_factory=list
- )
-
- def select(self, llm: Optional[LLM] = None) -> BasePromptTemplate:
- if llm is None:
- return self.default_prompt
-
- for condition, prompt in self.conditionals:
- if condition(llm):
- return prompt
- return self.default_prompt
-
-
-def is_chat_model(llm: LLM) -> bool:
- return llm.metadata.is_chat_model
diff --git a/llama_index/prompts/prompt_utils.py b/llama_index/prompts/prompt_utils.py
new file mode 100644
index 0000000000..ca121ed1ba
--- /dev/null
+++ b/llama_index/prompts/prompt_utils.py
@@ -0,0 +1,31 @@
+from typing import List
+from llama_index.prompts.base import BasePromptTemplate
+
+
+def get_empty_prompt_txt(prompt: BasePromptTemplate) -> str:
+ """Get empty prompt text.
+
+ Substitute empty strings in parts of the prompt that have
+ not yet been filled out. Skip variables that have already
+ been partially formatted. This is used to compute the initial tokens.
+
+ """
+ partial_kargs = prompt.kwargs
+ empty_kwargs = {v: "" for v in prompt.template_vars if v not in partial_kargs}
+ all_kwargs = {**partial_kargs, **empty_kwargs}
+ empty_prompt_txt = prompt.format(llm=None, **all_kwargs)
+ return empty_prompt_txt
+
+
+def get_biggest_prompt(prompts: List[BasePromptTemplate]) -> BasePromptTemplate:
+ """Get biggest prompt.
+
+ Oftentimes we need to fetch the biggest prompt, in order to
+ be the most conservative about chunking text. This
+ is a helper utility for that.
+
+ """
+ empty_prompt_txts = [get_empty_prompt_txt(prompt) for prompt in prompts]
+ empty_prompt_txt_lens = [len(txt) for txt in empty_prompt_txts]
+ biggest_prompt = prompts[empty_prompt_txt_lens.index(max(empty_prompt_txt_lens))]
+ return biggest_prompt
diff --git a/llama_index/prompts/prompts.py b/llama_index/prompts/prompts.py
index 8cfab47b6c..fc1d542276 100644
--- a/llama_index/prompts/prompts.py
+++ b/llama_index/prompts/prompts.py
@@ -1,41 +1,41 @@
"""Subclasses from base prompt."""
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import PromptTemplate
# deprecated, kept for backward compatibility
"""Summary prompt.
-Prompt to summarize the provided `context_str`.
+PromptTemplate to summarize the provided `context_str`.
Required template variables: `context_str`
"""
-SummaryPrompt = Prompt
+SummaryPrompt = PromptTemplate
"""Tree Insert prompt.
-Prompt to insert a new chunk of text `new_chunk_text` into the tree index.
+PromptTemplate to insert a new chunk of text `new_chunk_text` into the tree index.
More specifically, this prompt has the LLM select the relevant candidate
child node to continue tree traversal.
Required template variables: `num_chunks`, `context_list`, `new_chunk_text`
"""
-TreeInsertPrompt = Prompt
+TreeInsertPrompt = PromptTemplate
"""Tree select prompt.
-Prompt to select a candidate child node out of all child nodes
+PromptTemplate to select a candidate child node out of all child nodes
provided in `context_list`, given a query `query_str`. `num_chunks` is
the number of child nodes in `context_list`.
Required template variables: `num_chunks`, `context_list`, `query_str`
"""
-TreeSelectPrompt = Prompt
+TreeSelectPrompt = PromptTemplate
"""Tree select multiple prompt.
-Prompt to select multiple candidate child nodes out of all
+PromptTemplate to select multiple candidate child nodes out of all
child nodes provided in `context_list`, given a query `query_str`.
`branching_factor` refers to the number of child nodes to select, and
`num_chunks` is the number of child nodes in `context_list`.
@@ -43,91 +43,98 @@ child nodes provided in `context_list`, given a query `query_str`.
Required template variables: `num_chunks`, `context_list`, `query_str`,
`branching_factor`
"""
-TreeSelectMultiplePrompt = Prompt
+TreeSelectMultiplePrompt = PromptTemplate
"""Refine prompt.
-Prompt to refine an existing answer `existing_answer` given a context `context_msg`,
-and a query `query_str`.
+PromptTemplate to refine an existing answer `existing_answer`
+given a context `context_msg`, and a query `query_str`.
Required template variables: `query_str`, `existing_answer`, `context_msg`
"""
-RefinePrompt = Prompt
+RefinePrompt = PromptTemplate
"""Question Answer prompt.
-Prompt to answer a question `query_str` given a context `context_str`.
+PromptTemplate to answer a question `query_str` given a context `context_str`.
Required template variables: `context_str`, `query_str`
"""
-QuestionAnswerPrompt = Prompt
+QuestionAnswerPrompt = PromptTemplate
"""Keyword extract prompt.
-Prompt to extract keywords from a text `text` with a maximum of
+PromptTemplate to extract keywords from a text `text` with a maximum of
`max_keywords` keywords.
Required template variables: `text`, `max_keywords`
"""
-KeywordExtractPrompt = Prompt
+KeywordExtractPrompt = PromptTemplate
"""Query keyword extract prompt.
-Prompt to extract keywords from a query `query_str` with a maximum
+PromptTemplate to extract keywords from a query `query_str` with a maximum
of `max_keywords` keywords.
Required template variables: `query_str`, `max_keywords`
"""
-QueryKeywordExtractPrompt = Prompt
+QueryKeywordExtractPrompt = PromptTemplate
"""Schema extract prompt.
-Prompt to extract schema from unstructured text `text`.
+PromptTemplate to extract schema from unstructured text `text`.
Required template variables: `text`, `schema`
"""
-SchemaExtractPrompt = Prompt
+SchemaExtractPrompt = PromptTemplate
"""Text to SQL prompt.
-Prompt to translate a natural language query into SQL in the dialect
+PromptTemplate to translate a natural language query into SQL in the dialect
`dialect` given a schema `schema`.
Required template variables: `query_str`, `schema`, `dialect`
"""
-TextToSQLPrompt = Prompt
+TextToSQLPrompt = PromptTemplate
"""Table context prompt.
-Prompt to generate a table context given a table schema `schema`,
+PromptTemplate to generate a table context given a table schema `schema`,
as well as unstructured text context `context_str`, and
a task `query_str`.
This includes both a high-level description of the table
as well as a description of each column in the table.
"""
-TableContextPrompt = Prompt
+TableContextPrompt = PromptTemplate
"""Refine Table context prompt.
-Prompt to refine a table context given a table schema `schema`,
+PromptTemplate to refine a table context given a table schema `schema`,
as well as unstructured text context `context_msg`, and
a task `query_str`.
This includes both a high-level description of the table
as well as a description of each column in the table.
"""
-RefineTableContextPrompt = Prompt
+RefineTableContextPrompt = PromptTemplate
"""Define the knowledge graph triplet extraction prompt."""
-KnowledgeGraphPrompt = Prompt
+KnowledgeGraphPrompt = PromptTemplate
"""Simple Input prompt.
Required template variables: `query_str`.
"""
-SimpleInputPrompt = Prompt
+SimpleInputPrompt = PromptTemplate
"""Pandas prompt. Convert query to python code.
Required template variables: `query_str`, `df_str`, `instruction_str`.
"""
-PandasPrompt = Prompt
+PandasPrompt = PromptTemplate
+
+
+"""Choice select prompt. Select from a list of choices.
+
+Required template variables: `context_str`, `query_str`.
+"""
+ChoiceSelectPrompt = PromptTemplate
diff --git a/llama_index/prompts/utils.py b/llama_index/prompts/utils.py
index 90d4a8ae91..b81c8c6496 100644
--- a/llama_index/prompts/utils.py
+++ b/llama_index/prompts/utils.py
@@ -1,35 +1,20 @@
+from string import Formatter
from typing import List
-from llama_index.prompts.base import Prompt
+from llama_index.llms.base import LLM
-def get_empty_prompt_txt(prompt: Prompt) -> str:
- """Get empty prompt text.
+def get_template_vars(template_str: str) -> List[str]:
+ """Get template variables from a template string."""
+ variables = []
+ formatter = Formatter()
- Substitute empty strings in parts of the prompt that have
- not yet been filled out. Skip variables that have already
- been partially formatted. This is used to compute the initial tokens.
+ for _, variable_name, _, _ in formatter.parse(template_str):
+ if variable_name:
+ variables.append(variable_name)
- """
- fmt_dict = {
- v: ""
- for v in prompt.get_langchain_prompt().input_variables
- if v not in prompt.partial_dict
- }
- # TODO: change later from llm=None
- empty_prompt_txt = prompt.format(llm=None, **fmt_dict)
- return empty_prompt_txt
+ return variables
-def get_biggest_prompt(prompts: List[Prompt]) -> Prompt:
- """Get biggest prompt.
-
- Oftentimes we need to fetch the biggest prompt, in order to
- be the most conservative about chunking text. This
- is a helper utility for that.
-
- """
- empty_prompt_txts = [get_empty_prompt_txt(prompt) for prompt in prompts]
- empty_prompt_txt_lens = [len(txt) for txt in empty_prompt_txts]
- biggest_prompt = prompts[empty_prompt_txt_lens.index(max(empty_prompt_txt_lens))]
- return biggest_prompt
+def is_chat_model(llm: LLM) -> bool:
+ return llm.metadata.is_chat_model
diff --git a/llama_index/query_engine/citation_query_engine.py b/llama_index/query_engine/citation_query_engine.py
index 7cc3e7a427..508582ed26 100644
--- a/llama_index/query_engine/citation_query_engine.py
+++ b/llama_index/query_engine/citation_query_engine.py
@@ -7,7 +7,8 @@ from llama_index.indices.base_retriever import BaseRetriever
from llama_index.indices.postprocessor.types import BaseNodePostprocessor
from llama_index.indices.query.base import BaseQueryEngine
from llama_index.indices.query.schema import QueryBundle
-from llama_index.prompts.base import Prompt
+from llama_index.prompts import PromptTemplate
+from llama_index.prompts.base import BasePromptTemplate
from llama_index.response.schema import RESPONSE_TYPE
from llama_index.response_synthesizers import (
BaseSynthesizer,
@@ -18,7 +19,7 @@ from llama_index.schema import NodeWithScore, TextNode
from llama_index.text_splitter import get_default_text_splitter
from llama_index.text_splitter.types import TextSplitter
-CITATION_QA_TEMPLATE = Prompt(
+CITATION_QA_TEMPLATE = PromptTemplate(
"Please provide an answer based solely on the provided sources. "
"When referencing information from a source, "
"cite the appropriate source(s) using their corresponding numbers. "
@@ -41,7 +42,7 @@ CITATION_QA_TEMPLATE = Prompt(
"Answer: "
)
-CITATION_REFINE_TEMPLATE = Prompt(
+CITATION_REFINE_TEMPLATE = PromptTemplate(
"Please provide an answer based solely on the provided sources. "
"When referencing information from a source, "
"cite the appropriate source(s) using their corresponding numbers. "
@@ -120,8 +121,8 @@ class CitationQueryEngine(BaseQueryEngine):
citation_chunk_size: int = DEFAULT_CITATION_CHUNK_SIZE,
citation_chunk_overlap: int = DEFAULT_CITATION_CHUNK_OVERLAP,
text_splitter: Optional[TextSplitter] = None,
- citation_qa_template: Prompt = CITATION_QA_TEMPLATE,
- citation_refine_template: Prompt = CITATION_REFINE_TEMPLATE,
+ citation_qa_template: BasePromptTemplate = CITATION_QA_TEMPLATE,
+ citation_refine_template: BasePromptTemplate = CITATION_REFINE_TEMPLATE,
retriever: Optional[BaseRetriever] = None,
node_postprocessors: Optional[List[BaseNodePostprocessor]] = None,
# response synthesizer args
@@ -142,8 +143,9 @@ class CitationQueryEngine(BaseQueryEngine):
text_splitter (Optional[TextSplitter]):
A text splitter for creating citation source nodes. Default is
a SentenceSplitter.
- citation_qa_template (Prompt): Template for initial citation QA
- citation_refine_template (Prompt): Template for citation refinement.
+ citation_qa_template (BasePromptTemplate): Template for initial citation QA
+ citation_refine_template (BasePromptTemplate):
+ Template for citation refinement.
retriever (BaseRetriever): A retriever object.
service_context (Optional[ServiceContext]): A ServiceContext object.
node_postprocessors (Optional[List[BaseNodePostprocessor]]): A list of
diff --git a/llama_index/query_engine/flare/answer_inserter.py b/llama_index/query_engine/flare/answer_inserter.py
index f0fb8ba032..abcc5fd6ca 100644
--- a/llama_index/query_engine/flare/answer_inserter.py
+++ b/llama_index/query_engine/flare/answer_inserter.py
@@ -3,7 +3,7 @@
from abc import ABC, abstractmethod
from typing import List, Optional
from llama_index.query_engine.flare.schema import QueryTask
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import BasePromptTemplate, PromptTemplate
from llama_index.indices.service_context import ServiceContext
@@ -117,7 +117,7 @@ Query-Answer Pairs:
Synthesized Response:
"""
-DEFAULT_ANSWER_INSERT_PROMPT = Prompt(DEFAULT_ANSWER_INSERT_PROMPT_TMPL)
+DEFAULT_ANSWER_INSERT_PROMPT = PromptTemplate(DEFAULT_ANSWER_INSERT_PROMPT_TMPL)
class LLMLookaheadAnswerInserter(BaseLookaheadAnswerInserter):
@@ -134,7 +134,7 @@ class LLMLookaheadAnswerInserter(BaseLookaheadAnswerInserter):
def __init__(
self,
service_context: Optional[ServiceContext] = None,
- answer_insert_prompt: Optional[Prompt] = None,
+ answer_insert_prompt: Optional[BasePromptTemplate] = None,
) -> None:
"""Init params."""
self._service_context = service_context or ServiceContext.from_defaults()
diff --git a/llama_index/query_engine/flare/base.py b/llama_index/query_engine/flare/base.py
index a9b008c562..9d214ef927 100644
--- a/llama_index/query_engine/flare/base.py
+++ b/llama_index/query_engine/flare/base.py
@@ -10,7 +10,7 @@ from llama_index.indices.query.base import BaseQueryEngine
from llama_index.indices.service_context import ServiceContext
from llama_index.indices.query.schema import QueryBundle
from llama_index.response.schema import RESPONSE_TYPE, Response
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import PromptTemplate, BasePromptTemplate
from llama_index.callbacks.base import CallbackManager
from llama_index.query_engine.flare.output_parser import (
IsDoneOutputParser,
@@ -87,7 +87,7 @@ Answer: """
)
)
-DEFAULT_INSTRUCT_PROMPT = Prompt(DEFAULT_INSTRUCT_PROMPT_TMPL)
+DEFAULT_INSTRUCT_PROMPT = PromptTemplate(DEFAULT_INSTRUCT_PROMPT_TMPL)
class FLAREInstructQueryEngine(BaseQueryEngine):
@@ -102,7 +102,7 @@ class FLAREInstructQueryEngine(BaseQueryEngine):
query_engine (BaseQueryEngine): query engine to use
service_context (Optional[ServiceContext]): service context.
Defaults to None.
- instruct_prompt (Optional[Prompt]): instruct prompt. Defaults to None.
+ instruct_prompt (Optional[PromptTemplate]): instruct prompt. Defaults to None.
lookahead_answer_inserter (Optional[BaseLookaheadAnswerInserter]):
lookahead answer inserter. Defaults to None.
done_output_parser (Optional[IsDoneOutputParser]): done output parser.
@@ -121,7 +121,7 @@ class FLAREInstructQueryEngine(BaseQueryEngine):
self,
query_engine: BaseQueryEngine,
service_context: Optional[ServiceContext] = None,
- instruct_prompt: Optional[Prompt] = None,
+ instruct_prompt: Optional[BasePromptTemplate] = None,
lookahead_answer_inserter: Optional[BaseLookaheadAnswerInserter] = None,
done_output_parser: Optional[IsDoneOutputParser] = None,
query_task_output_parser: Optional[QueryTaskOutputParser] = None,
diff --git a/llama_index/query_engine/knowledge_graph_query_engine.py b/llama_index/query_engine/knowledge_graph_query_engine.py
index f8bef189ff..36a18f5f19 100644
--- a/llama_index/query_engine/knowledge_graph_query_engine.py
+++ b/llama_index/query_engine/knowledge_graph_query_engine.py
@@ -12,7 +12,7 @@ from llama_index.graph_stores.registery import (
from llama_index.indices.query.base import BaseQueryEngine
from llama_index.indices.query.schema import QueryBundle
from llama_index.indices.service_context import ServiceContext
-from llama_index.prompts.base import Prompt, PromptType
+from llama_index.prompts.base import BasePromptTemplate, PromptTemplate, PromptType
from llama_index.response.schema import RESPONSE_TYPE
from llama_index.response_synthesizers import BaseSynthesizer, get_response_synthesizer
from llama_index.schema import NodeWithScore, TextNode
@@ -49,7 +49,7 @@ Question: {query_str}
NebulaGraph Cypher dialect query:
"""
-DEFAULT_NEBULAGRAPH_NL2CYPHER_PROMPT = Prompt(
+DEFAULT_NEBULAGRAPH_NL2CYPHER_PROMPT = PromptTemplate(
DEFAULT_NEBULAGRAPH_NL2CYPHER_PROMPT_TMPL,
prompt_type=PromptType.TEXT_TO_GRAPH_QUERY,
)
@@ -71,7 +71,7 @@ DEFAULT_NEO4J_NL2CYPHER_PROMPT_TMPL = (
"{query_str}\n"
)
-DEFAULT_NEO4J_NL2CYPHER_PROMPT = Prompt(
+DEFAULT_NEO4J_NL2CYPHER_PROMPT = PromptTemplate(
DEFAULT_NEO4J_NL2CYPHER_PROMPT_TMPL,
prompt_type=PromptType.TEXT_TO_GRAPH_QUERY,
)
@@ -93,7 +93,7 @@ Graph response: {kg_response_str}
Response:
"""
-DEFAULT_KG_RESPONSE_ANSWER_PROMPT = Prompt(
+DEFAULT_KG_RESPONSE_ANSWER_PROMPT = PromptTemplate(
DEFAULT_KG_RESPONSE_ANSWER_PROMPT_TMPL,
prompt_type=PromptType.QUESTION_ANSWER,
)
@@ -119,8 +119,8 @@ class KnowledgeGraphQueryEngine(BaseQueryEngine):
self,
service_context: Optional[ServiceContext] = None,
storage_context: Optional[StorageContext] = None,
- graph_query_synthesis_prompt: Optional[Prompt] = None,
- graph_response_answer_prompt: Optional[Prompt] = None,
+ graph_query_synthesis_prompt: Optional[BasePromptTemplate] = None,
+ graph_response_answer_prompt: Optional[BasePromptTemplate] = None,
refresh_schema: bool = False,
verbose: bool = False,
response_synthesizer: Optional[BaseSynthesizer] = None,
diff --git a/llama_index/query_engine/pandas_query_engine.py b/llama_index/query_engine/pandas_query_engine.py
index 95749eec1f..63533409dc 100644
--- a/llama_index/query_engine/pandas_query_engine.py
+++ b/llama_index/query_engine/pandas_query_engine.py
@@ -10,16 +10,16 @@ require heavy sandboxing or virtual machines
import logging
from typing import Any, Callable, Optional
-import pandas as pd
import numpy as np
-from llama_index.bridge.langchain import print_text
+import pandas as pd
+from llama_index.bridge.langchain import print_text
from llama_index.indices.query.base import BaseQueryEngine
from llama_index.indices.query.schema import QueryBundle
from llama_index.indices.service_context import ServiceContext
from llama_index.indices.struct_store.pandas import PandasIndex
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompts import DEFAULT_PANDAS_PROMPT
-from llama_index.prompts.prompts import PandasPrompt
from llama_index.response.schema import Response
logger = logging.getLogger(__name__)
@@ -91,7 +91,7 @@ class PandasQueryEngine(BaseQueryEngine):
output_processor (Optional[Callable[[str], str]]): Output processor.
A callable that takes in the output string, pandas DataFrame,
and any output kwargs and returns a string.
- pandas_prompt (Optional[PandasPrompt]): Pandas prompt to use.
+ pandas_prompt (Optional[BasePromptTemplate]): Pandas prompt to use.
head (int): Number of rows to show in the table context.
"""
@@ -101,7 +101,7 @@ class PandasQueryEngine(BaseQueryEngine):
df: pd.DataFrame,
instruction_str: Optional[str] = None,
output_processor: Optional[Callable] = None,
- pandas_prompt: Optional[PandasPrompt] = None,
+ pandas_prompt: Optional[BasePromptTemplate] = None,
output_kwargs: Optional[dict] = None,
head: int = 5,
verbose: bool = False,
diff --git a/llama_index/query_engine/retriever_query_engine.py b/llama_index/query_engine/retriever_query_engine.py
index ca8dc0f86d..7fcc0ada4a 100644
--- a/llama_index/query_engine/retriever_query_engine.py
+++ b/llama_index/query_engine/retriever_query_engine.py
@@ -7,17 +7,13 @@ from llama_index.indices.postprocessor.types import BaseNodePostprocessor
from llama_index.indices.query.base import BaseQueryEngine
from llama_index.indices.query.schema import QueryBundle
from llama_index.indices.service_context import ServiceContext
+from llama_index.prompts import BasePromptTemplate
+from llama_index.response.schema import RESPONSE_TYPE
from llama_index.response_synthesizers import (
BaseSynthesizer,
ResponseMode,
get_response_synthesizer,
)
-from llama_index.prompts.prompts import (
- QuestionAnswerPrompt,
- RefinePrompt,
- SimpleInputPrompt,
-)
-from llama_index.response.schema import RESPONSE_TYPE
from llama_index.schema import NodeWithScore
@@ -55,9 +51,9 @@ class RetrieverQueryEngine(BaseQueryEngine):
node_postprocessors: Optional[List[BaseNodePostprocessor]] = None,
# response synthesizer args
response_mode: ResponseMode = ResponseMode.COMPACT,
- text_qa_template: Optional[QuestionAnswerPrompt] = None,
- refine_template: Optional[RefinePrompt] = None,
- simple_template: Optional[SimpleInputPrompt] = None,
+ text_qa_template: Optional[BasePromptTemplate] = None,
+ refine_template: Optional[BasePromptTemplate] = None,
+ simple_template: Optional[BasePromptTemplate] = None,
use_async: bool = False,
streaming: bool = False,
# class-specific args
@@ -72,10 +68,10 @@ class RetrieverQueryEngine(BaseQueryEngine):
node postprocessors.
verbose (bool): Whether to print out debug info.
response_mode (ResponseMode): A ResponseMode object.
- text_qa_template (Optional[QuestionAnswerPrompt]): A QuestionAnswerPrompt
+ text_qa_template (Optional[BasePromptTemplate]): A BasePromptTemplate
object.
- refine_template (Optional[RefinePrompt]): A RefinePrompt object.
- simple_template (Optional[SimpleInputPrompt]): A SimpleInputPrompt object.
+ refine_template (Optional[BasePromptTemplate]): A BasePromptTemplate object.
+ simple_template (Optional[BasePromptTemplate]): A BasePromptTemplate object.
use_async (bool): Whether to use async.
streaming (bool): Whether to use streaming.
diff --git a/llama_index/query_engine/sql_join_query_engine.py b/llama_index/query_engine/sql_join_query_engine.py
index 69d8d22883..592ce061e3 100644
--- a/llama_index/query_engine/sql_join_query_engine.py
+++ b/llama_index/query_engine/sql_join_query_engine.py
@@ -1,36 +1,40 @@
"""SQL Join query engine."""
+import logging
+from typing import Callable, Dict, Optional, Union
+
from llama_index.bridge.langchain import print_text
-from typing import Optional, Dict, Callable, Union
+from llama_index.callbacks.base import CallbackManager
from llama_index.indices.query.base import BaseQueryEngine
+from llama_index.indices.query.query_transform.base import BaseQueryTransform
+from llama_index.indices.query.schema import QueryBundle
+from llama_index.indices.service_context import ServiceContext
from llama_index.indices.struct_store.sql_query import (
BaseSQLTableQueryEngine,
NLSQLTableQueryEngine,
)
-from llama_index.indices.query.schema import QueryBundle
+from llama_index.llm_predictor import LLMPredictor
+from llama_index.llm_predictor.base import BaseLLMPredictor
+from llama_index.prompts.base import BasePromptTemplate, PromptTemplate
from llama_index.response.schema import RESPONSE_TYPE, Response
-from llama_index.tools.query_engine import QueryEngineTool
-from llama_index.indices.service_context import ServiceContext
-from llama_index.selectors.utils import get_selector_from_context
from llama_index.selectors.llm_selectors import LLMSingleSelector
from llama_index.selectors.pydantic_selectors import PydanticSingleSelector
-from llama_index.prompts.base import Prompt
-from llama_index.indices.query.query_transform.base import BaseQueryTransform
-import logging
-from llama_index.llm_predictor import LLMPredictor
-from llama_index.llm_predictor.base import BaseLLMPredictor
-from llama_index.callbacks.base import CallbackManager
+from llama_index.selectors.utils import get_selector_from_context
+from llama_index.tools.query_engine import QueryEngineTool
logger = logging.getLogger(__name__)
DEFAULT_SQL_JOIN_SYNTHESIS_PROMPT_TMPL = """
The original question is given below.
-This question has been translated into a SQL query. Both the SQL query and the response are given below.
-Given the SQL response, the question has also been transformed into a more detailed query,
+This question has been translated into a SQL query. Both the SQL query and \
+the response are given below.
+Given the SQL response, the question has also been transformed into a more \
+detailed query,
and executed against another query engine.
The transformed query and query engine response are also given below.
-Given SQL query, SQL response, transformed query, and query engine response, please synthesize a response to the original question.
+Given SQL query, SQL response, transformed query, and query engine response, \
+please synthesize a response to the original question.
Original question: {query_str}
SQL query: {sql_query_str}
@@ -39,18 +43,26 @@ Transformed query: {query_engine_query_str}
Query engine response: {query_engine_response_str}
Response:
""" # noqa
-DEFAULT_SQL_JOIN_SYNTHESIS_PROMPT = Prompt(DEFAULT_SQL_JOIN_SYNTHESIS_PROMPT_TMPL)
+DEFAULT_SQL_JOIN_SYNTHESIS_PROMPT = PromptTemplate(
+ DEFAULT_SQL_JOIN_SYNTHESIS_PROMPT_TMPL
+)
DEFAULT_SQL_AUGMENT_TRANSFORM_PROMPT_TMPL = """
"The original question is given below.
-This question has been translated into a SQL query. Both the SQL query and the response are given below.
-The SQL response either answers the question, or should provide additional context that can be used to make the question more specific.
-Your job is to come up with a more specific question that needs to be answered to fully answer the original question, or 'None' if the original question has already been fully answered from the SQL response. Do not create a new question that is irrelevant to the original question; in that case return None instead.
+This question has been translated into a SQL query. Both the SQL query and the \
+response are given below.
+The SQL response either answers the question, or should provide additional context \
+that can be used to make the question more specific.
+Your job is to come up with a more specific question that needs to be answered to \
+fully answer the original question, or 'None' if the original question has already \
+been fully answered from the SQL response. Do not create a new question that is \
+irrelevant to the original question; in that case return None instead.
Examples:
-Original question: Please give more details about the demographics of the city with the highest population.
+Original question: Please give more details about the demographics of the city with \
+the highest population.
SQL query: SELECT city, population FROM cities ORDER BY population DESC LIMIT 1
SQL response: The city with the highest population is New York City.
New question: Can you tell me more about the demographics of New York City?
@@ -75,7 +87,9 @@ SQL query: {sql_query_str}
SQL response: {sql_response_str}
New question: "
""" # noqa
-DEFAULT_SQL_AUGMENT_TRANSFORM_PROMPT = Prompt(DEFAULT_SQL_AUGMENT_TRANSFORM_PROMPT_TMPL)
+DEFAULT_SQL_AUGMENT_TRANSFORM_PROMPT = PromptTemplate(
+ DEFAULT_SQL_AUGMENT_TRANSFORM_PROMPT_TMPL
+)
def _default_check_stop(query_bundle: QueryBundle) -> bool:
@@ -96,7 +110,8 @@ class SQLAugmentQueryTransform(BaseQueryTransform):
Args:
llm_predictor (LLMPredictor): LLM predictor to use for query transformation.
- sql_augment_transform_prompt (Prompt): Prompt to use for query transformation.
+ sql_augment_transform_prompt (BasePromptTemplate): PromptTemplate to use
+ for query transformation.
check_stop_parser (Optional[Callable[[str], bool]]): Check stop function.
"""
@@ -104,7 +119,7 @@ class SQLAugmentQueryTransform(BaseQueryTransform):
def __init__(
self,
llm_predictor: Optional[BaseLLMPredictor] = None,
- sql_augment_transform_prompt: Optional[Prompt] = None,
+ sql_augment_transform_prompt: Optional[BasePromptTemplate] = None,
check_stop_parser: Optional[Callable[[QueryBundle], bool]] = None,
) -> None:
"""Initialize params."""
@@ -150,8 +165,8 @@ class SQLJoinQueryEngine(BaseQueryEngine):
selector (Optional[Union[LLMSingleSelector, PydanticSingleSelector]]):
Selector to use.
service_context (Optional[ServiceContext]): Service context to use.
- sql_join_synthesis_prompt (Optional[Prompt]): Prompt to use for SQL join
- synthesis.
+ sql_join_synthesis_prompt (Optional[BasePromptTemplate]):
+ PromptTemplate to use for SQL join synthesis.
sql_augment_query_transform (Optional[SQLAugmentQueryTransform]): Query
transform to use for SQL augmentation.
use_sql_join_synthesis (bool): Whether to use SQL join synthesis.
@@ -166,7 +181,7 @@ class SQLJoinQueryEngine(BaseQueryEngine):
other_query_tool: QueryEngineTool,
selector: Optional[Union[LLMSingleSelector, PydanticSingleSelector]] = None,
service_context: Optional[ServiceContext] = None,
- sql_join_synthesis_prompt: Optional[Prompt] = None,
+ sql_join_synthesis_prompt: Optional[BasePromptTemplate] = None,
sql_augment_query_transform: Optional[SQLAugmentQueryTransform] = None,
use_sql_join_synthesis: bool = True,
callback_manager: Optional[CallbackManager] = None,
diff --git a/llama_index/query_engine/sql_vector_query_engine.py b/llama_index/query_engine/sql_vector_query_engine.py
index 91120375b2..1fd7c81bff 100644
--- a/llama_index/query_engine/sql_vector_query_engine.py
+++ b/llama_index/query_engine/sql_vector_query_engine.py
@@ -1,6 +1,10 @@
"""SQL Vector query engine."""
-from typing import Optional, Any, Union
+import logging
+from typing import Any, Optional, Union
+
+from llama_index.callbacks.base import CallbackManager
+from llama_index.indices.service_context import ServiceContext
from llama_index.indices.struct_store.sql_query import (
BaseSQLTableQueryEngine,
NLSQLTableQueryEngine,
@@ -8,28 +12,27 @@ from llama_index.indices.struct_store.sql_query import (
from llama_index.indices.vector_store.retrievers.auto_retriever import (
VectorIndexAutoRetriever,
)
-from llama_index.tools.query_engine import QueryEngineTool
+from llama_index.prompts.base import BasePromptTemplate, PromptTemplate
from llama_index.query_engine.retriever_query_engine import RetrieverQueryEngine
-from llama_index.indices.service_context import ServiceContext
-from llama_index.selectors.llm_selectors import LLMSingleSelector
-from llama_index.selectors.pydantic_selectors import PydanticSingleSelector
-from llama_index.prompts.base import Prompt
-import logging
-from llama_index.callbacks.base import CallbackManager
from llama_index.query_engine.sql_join_query_engine import (
- SQLJoinQueryEngine,
SQLAugmentQueryTransform,
+ SQLJoinQueryEngine,
)
+from llama_index.selectors.llm_selectors import LLMSingleSelector
+from llama_index.selectors.pydantic_selectors import PydanticSingleSelector
+from llama_index.tools.query_engine import QueryEngineTool
logger = logging.getLogger(__name__)
DEFAULT_SQL_VECTOR_SYNTHESIS_PROMPT_TMPL = """
The original question is given below.
-This question has been translated into a SQL query. Both the SQL query and the response are given below.
+This question has been translated into a SQL query. \
+Both the SQL query and the response are given below.
Given the SQL response, the question has also been translated into a vector store query.
The vector store query and response is given below.
-Given SQL query, SQL response, transformed vector store query, and vector store response, please synthesize a response to the original question.
+Given SQL query, SQL response, transformed vector store query, and vector store \
+response, please synthesize a response to the original question.
Original question: {query_str}
SQL query: {sql_query_str}
@@ -38,7 +41,9 @@ Transformed vector store query: {query_engine_query_str}
Vector store response: {query_engine_response_str}
Response:
""" # noqa
-DEFAULT_SQL_VECTOR_SYNTHESIS_PROMPT = Prompt(DEFAULT_SQL_VECTOR_SYNTHESIS_PROMPT_TMPL)
+DEFAULT_SQL_VECTOR_SYNTHESIS_PROMPT = PromptTemplate(
+ DEFAULT_SQL_VECTOR_SYNTHESIS_PROMPT_TMPL
+)
# NOTE: maintain for backwards compatibility
@@ -58,8 +63,8 @@ class SQLAutoVectorQueryEngine(SQLJoinQueryEngine):
selector (Optional[Union[LLMSingleSelector, PydanticSingleSelector]]):
Selector to use.
service_context (Optional[ServiceContext]): Service context to use.
- sql_vector_synthesis_prompt (Optional[Prompt]): Prompt to use for SQL vector
- synthesis.
+ sql_vector_synthesis_prompt (Optional[BasePromptTemplate]):
+ Prompt to use for SQL vector synthesis.
sql_augment_query_transform (Optional[SQLAugmentQueryTransform]): Query
transform to use for SQL augmentation.
use_sql_vector_synthesis (bool): Whether to use SQL vector synthesis.
@@ -74,7 +79,7 @@ class SQLAutoVectorQueryEngine(SQLJoinQueryEngine):
vector_query_tool: QueryEngineTool,
selector: Optional[Union[LLMSingleSelector, PydanticSingleSelector]] = None,
service_context: Optional[ServiceContext] = None,
- sql_vector_synthesis_prompt: Optional[Prompt] = None,
+ sql_vector_synthesis_prompt: Optional[BasePromptTemplate] = None,
sql_augment_query_transform: Optional[SQLAugmentQueryTransform] = None,
use_sql_vector_synthesis: bool = True,
callback_manager: Optional[CallbackManager] = None,
diff --git a/llama_index/question_gen/llm_generators.py b/llama_index/question_gen/llm_generators.py
index f61e027db9..ce8836732b 100644
--- a/llama_index/question_gen/llm_generators.py
+++ b/llama_index/question_gen/llm_generators.py
@@ -5,7 +5,7 @@ from llama_index.indices.service_context import ServiceContext
from llama_index.llm_predictor.base import BaseLLMPredictor
from llama_index.output_parsers.base import StructuredOutput
from llama_index.types import BaseOutputParser
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import BasePromptTemplate, PromptTemplate
from llama_index.prompts.prompt_type import PromptType
from llama_index.question_gen.output_parser import SubQuestionOutputParser
from llama_index.question_gen.prompts import (
@@ -20,7 +20,7 @@ class LLMQuestionGenerator(BaseQuestionGenerator):
def __init__(
self,
llm_predictor: BaseLLMPredictor,
- prompt: Prompt,
+ prompt: BasePromptTemplate,
) -> None:
self._llm_predictor = llm_predictor
self._prompt = prompt
@@ -41,7 +41,7 @@ class LLMQuestionGenerator(BaseQuestionGenerator):
output_parser = output_parser or SubQuestionOutputParser()
# construct prompt
- prompt = Prompt(
+ prompt = PromptTemplate(
template=prompt_template_str,
output_parser=output_parser,
prompt_type=PromptType.SUB_QUESTION,
diff --git a/llama_index/question_gen/prompts.py b/llama_index/question_gen/prompts.py
index daa7738fa6..aed8f382af 100644
--- a/llama_index/question_gen/prompts.py
+++ b/llama_index/question_gen/prompts.py
@@ -1,12 +1,12 @@
import json
from typing import Sequence
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import PromptTemplate
from llama_index.question_gen.types import SubQuestion
from llama_index.tools.types import ToolMetadata
# deprecated, kept for backward compatibility
-SubQuestionPrompt = Prompt
+SubQuestionPrompt = PromptTemplate
def build_tools_text(tools: Sequence[ToolMetadata]) -> str:
diff --git a/llama_index/response_synthesizers/accumulate.py b/llama_index/response_synthesizers/accumulate.py
index ebce6f9c21..5f00979e01 100644
--- a/llama_index/response_synthesizers/accumulate.py
+++ b/llama_index/response_synthesizers/accumulate.py
@@ -1,12 +1,10 @@
import asyncio
-from typing import Any, List, Sequence, Optional
+from typing import Any, List, Optional, Sequence
from llama_index.async_utils import run_async_tasks
from llama_index.indices.service_context import ServiceContext
-from llama_index.prompts.default_prompts import (
- DEFAULT_TEXT_QA_PROMPT,
-)
-from llama_index.prompts.prompts import QuestionAnswerPrompt
+from llama_index.prompts import BasePromptTemplate
+from llama_index.prompts.default_prompts import DEFAULT_TEXT_QA_PROMPT
from llama_index.response_synthesizers.base import BaseSynthesizer
from llama_index.types import RESPONSE_TEXT_TYPE
@@ -16,7 +14,7 @@ class Accumulate(BaseSynthesizer):
def __init__(
self,
- text_qa_template: Optional[QuestionAnswerPrompt] = None,
+ text_qa_template: Optional[BasePromptTemplate] = None,
service_context: Optional[ServiceContext] = None,
streaming: bool = False,
use_async: bool = False,
diff --git a/llama_index/response_synthesizers/compact_and_refine.py b/llama_index/response_synthesizers/compact_and_refine.py
index c99617f930..ee49ff6d7f 100644
--- a/llama_index/response_synthesizers/compact_and_refine.py
+++ b/llama_index/response_synthesizers/compact_and_refine.py
@@ -1,6 +1,6 @@
from typing import Any, List, Sequence
-from llama_index.prompts.utils import get_biggest_prompt
+from llama_index.prompts.prompt_utils import get_biggest_prompt
from llama_index.response_synthesizers.refine import Refine
from llama_index.types import RESPONSE_TEXT_TYPE
diff --git a/llama_index/response_synthesizers/factory.py b/llama_index/response_synthesizers/factory.py
index 2ef1f015f6..0405da137f 100644
--- a/llama_index/response_synthesizers/factory.py
+++ b/llama_index/response_synthesizers/factory.py
@@ -2,18 +2,13 @@ from typing import Optional
from llama_index.callbacks.base import CallbackManager
from llama_index.indices.service_context import ServiceContext
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompt_selectors import (
+ DEFAULT_REFINE_PROMPT_SEL,
DEFAULT_TEXT_QA_PROMPT_SEL,
DEFAULT_TREE_SUMMARIZE_PROMPT_SEL,
- DEFAULT_REFINE_PROMPT_SEL,
)
from llama_index.prompts.default_prompts import DEFAULT_SIMPLE_INPUT_PROMPT
-from llama_index.prompts.prompts import (
- QuestionAnswerPrompt,
- RefinePrompt,
- SimpleInputPrompt,
- SummaryPrompt,
-)
from llama_index.response_synthesizers.accumulate import Accumulate
from llama_index.response_synthesizers.base import BaseSynthesizer
from llama_index.response_synthesizers.compact_and_accumulate import (
@@ -21,19 +16,19 @@ from llama_index.response_synthesizers.compact_and_accumulate import (
)
from llama_index.response_synthesizers.compact_and_refine import CompactAndRefine
from llama_index.response_synthesizers.generation import Generation
+from llama_index.response_synthesizers.no_text import NoText
from llama_index.response_synthesizers.refine import Refine
from llama_index.response_synthesizers.simple_summarize import SimpleSummarize
from llama_index.response_synthesizers.tree_summarize import TreeSummarize
from llama_index.response_synthesizers.type import ResponseMode
-from llama_index.response_synthesizers.no_text import NoText
def get_response_synthesizer(
service_context: Optional[ServiceContext] = None,
- text_qa_template: Optional[QuestionAnswerPrompt] = None,
- refine_template: Optional[RefinePrompt] = None,
- summary_template: Optional[SummaryPrompt] = None,
- simple_template: Optional[SimpleInputPrompt] = None,
+ text_qa_template: Optional[BasePromptTemplate] = None,
+ refine_template: Optional[BasePromptTemplate] = None,
+ summary_template: Optional[BasePromptTemplate] = None,
+ simple_template: Optional[BasePromptTemplate] = None,
response_mode: ResponseMode = ResponseMode.COMPACT,
callback_manager: Optional[CallbackManager] = None,
use_async: bool = False,
diff --git a/llama_index/response_synthesizers/generation.py b/llama_index/response_synthesizers/generation.py
index 3524e85d22..6bd4617034 100644
--- a/llama_index/response_synthesizers/generation.py
+++ b/llama_index/response_synthesizers/generation.py
@@ -1,8 +1,8 @@
from typing import Any, Optional, Sequence
from llama_index.indices.service_context import ServiceContext
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompts import DEFAULT_SIMPLE_INPUT_PROMPT
-from llama_index.prompts.prompts import SimpleInputPrompt
from llama_index.response_synthesizers.base import BaseSynthesizer
from llama_index.types import RESPONSE_TEXT_TYPE
@@ -10,7 +10,7 @@ from llama_index.types import RESPONSE_TEXT_TYPE
class Generation(BaseSynthesizer):
def __init__(
self,
- simple_template: Optional[SimpleInputPrompt] = None,
+ simple_template: Optional[BasePromptTemplate] = None,
service_context: Optional[ServiceContext] = None,
streaming: bool = False,
) -> None:
diff --git a/llama_index/response_synthesizers/refine.py b/llama_index/response_synthesizers/refine.py
index 0a9ea51b5c..a686908aab 100644
--- a/llama_index/response_synthesizers/refine.py
+++ b/llama_index/response_synthesizers/refine.py
@@ -3,11 +3,11 @@ from typing import Any, Generator, Optional, Sequence, cast
from llama_index.indices.service_context import ServiceContext
from llama_index.indices.utils import truncate_text
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompt_selectors import (
- DEFAULT_TEXT_QA_PROMPT_SEL,
DEFAULT_REFINE_PROMPT_SEL,
+ DEFAULT_TEXT_QA_PROMPT_SEL,
)
-from llama_index.prompts.prompts import QuestionAnswerPrompt, RefinePrompt
from llama_index.response.utils import get_response_text
from llama_index.response_synthesizers.base import BaseSynthesizer
from llama_index.types import RESPONSE_TEXT_TYPE
@@ -21,8 +21,8 @@ class Refine(BaseSynthesizer):
def __init__(
self,
service_context: Optional[ServiceContext] = None,
- text_qa_template: Optional[QuestionAnswerPrompt] = None,
- refine_template: Optional[RefinePrompt] = None,
+ text_qa_template: Optional[BasePromptTemplate] = None,
+ refine_template: Optional[BasePromptTemplate] = None,
streaming: bool = False,
verbose: bool = False,
) -> None:
diff --git a/llama_index/response_synthesizers/simple_summarize.py b/llama_index/response_synthesizers/simple_summarize.py
index 47352a554c..3a983d5b38 100644
--- a/llama_index/response_synthesizers/simple_summarize.py
+++ b/llama_index/response_synthesizers/simple_summarize.py
@@ -1,8 +1,8 @@
from typing import Any, Generator, Optional, Sequence, cast
from llama_index.indices.service_context import ServiceContext
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompt_selectors import DEFAULT_TEXT_QA_PROMPT_SEL
-from llama_index.prompts.prompts import QuestionAnswerPrompt
from llama_index.response_synthesizers.base import BaseSynthesizer
from llama_index.types import RESPONSE_TEXT_TYPE
@@ -10,7 +10,7 @@ from llama_index.types import RESPONSE_TEXT_TYPE
class SimpleSummarize(BaseSynthesizer):
def __init__(
self,
- text_qa_template: Optional[QuestionAnswerPrompt] = None,
+ text_qa_template: Optional[BasePromptTemplate] = None,
service_context: Optional[ServiceContext] = None,
streaming: bool = False,
) -> None:
diff --git a/llama_index/response_synthesizers/tree_summarize.py b/llama_index/response_synthesizers/tree_summarize.py
index 3b037317f7..c45fd7b501 100644
--- a/llama_index/response_synthesizers/tree_summarize.py
+++ b/llama_index/response_synthesizers/tree_summarize.py
@@ -3,10 +3,10 @@ from typing import Any, List, Optional, Sequence
from llama_index.async_utils import run_async_tasks
from llama_index.indices.service_context import ServiceContext
+from llama_index.prompts import BasePromptTemplate
from llama_index.prompts.default_prompt_selectors import (
DEFAULT_TREE_SUMMARIZE_PROMPT_SEL,
)
-from llama_index.prompts.prompts import SummaryPrompt
from llama_index.response_synthesizers.base import BaseSynthesizer
from llama_index.types import RESPONSE_TEXT_TYPE
@@ -26,7 +26,7 @@ class TreeSummarize(BaseSynthesizer):
def __init__(
self,
- summary_template: Optional[SummaryPrompt] = None,
+ summary_template: Optional[BasePromptTemplate] = None,
service_context: Optional[ServiceContext] = None,
streaming: bool = False,
use_async: bool = False,
diff --git a/llama_index/selectors/prompts.py b/llama_index/selectors/prompts.py
index 88dd0186a5..b6331babb0 100644
--- a/llama_index/selectors/prompts.py
+++ b/llama_index/selectors/prompts.py
@@ -1,25 +1,25 @@
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import PromptTemplate
from llama_index.prompts.prompt_type import PromptType
"""Single select prompt.
-Prompt to select one out of `num_choices` options provided in `context_list`,
+PromptTemplate to select one out of `num_choices` options provided in `context_list`,
given a query `query_str`.
Required template variables: `num_chunks`, `context_list`, `query_str`
"""
-SingleSelectPrompt = Prompt
+SingleSelectPrompt = PromptTemplate
"""Multiple select prompt.
-Prompt to select multiple candidates (up to `max_outputs`) out of `num_choices`
+PromptTemplate to select multiple candidates (up to `max_outputs`) out of `num_choices`
options provided in `context_list`, given a query `query_str`.
Required template variables: `num_chunks`, `context_list`, `query_str`,
`max_outputs`
"""
-MultiSelectPrompt = Prompt
+MultiSelectPrompt = PromptTemplate
# single select
@@ -35,7 +35,7 @@ DEFAULT_SINGLE_SELECT_PROMPT_TMPL = (
)
-DEFAULT_SINGLE_SELECT_PROMPT = Prompt(
+DEFAULT_SINGLE_SELECT_PROMPT = PromptTemplate(
template=DEFAULT_SINGLE_SELECT_PROMPT_TMPL, prompt_type=PromptType.SINGLE_SELECT
)
@@ -54,7 +54,7 @@ DEFAULT_MULTI_SELECT_PROMPT_TMPL = (
)
-DEFAULT_MULTIPLE_SELECT_PROMPT = Prompt(
+DEFAULT_MULTIPLE_SELECT_PROMPT = PromptTemplate(
template=DEFAULT_MULTI_SELECT_PROMPT_TMPL, prompt_type=PromptType.MULTI_SELECT
)
diff --git a/llama_index/types.py b/llama_index/types.py
index b4cf82bb75..2d6d2ba720 100644
--- a/llama_index/types.py
+++ b/llama_index/types.py
@@ -1,5 +1,14 @@
-from typing import AsyncGenerator, Generator, Union, Protocol, Any, TypeVar
from abc import abstractmethod
+from typing import (
+ Any,
+ AsyncGenerator,
+ Generator,
+ Protocol,
+ TypeVar,
+ Union,
+ runtime_checkable,
+)
+
from pydantic import BaseModel
Model = TypeVar("Model", bound=BaseModel)
@@ -10,6 +19,8 @@ RESPONSE_TEXT_TYPE = Union[str, TokenGen]
# TODO: move into a `core` folder
+# NOTE: this is necessary to make it compatible with pydantic
+@runtime_checkable
class BaseOutputParser(Protocol):
"""Output parser class."""
diff --git a/tests/indices/list/test_retrievers.py b/tests/indices/list/test_retrievers.py
index f65378fae6..c830dbc4a1 100644
--- a/tests/indices/list/test_retrievers.py
+++ b/tests/indices/list/test_retrievers.py
@@ -5,8 +5,7 @@ from llama_index.indices.list.base import ListIndex
from llama_index.indices.list.retrievers import ListIndexEmbeddingRetriever
from llama_index.indices.service_context import ServiceContext
from llama_index.llm_predictor.base import LLMPredictor
-from llama_index.prompts.choice_select import ChoiceSelectPrompt
-from llama_index.prompts.prompts import Prompt
+from llama_index.prompts import BasePromptTemplate
from llama_index.schema import Document
from tests.indices.list.test_index import _get_embeddings
@@ -47,9 +46,10 @@ def test_embedding_query(
assert nodes[0].node.get_content() == "Hello world."
-def mock_llmpredictor_predict(self: Any, prompt: Prompt, **prompt_args: Any) -> str:
+def mock_llmpredictor_predict(
+ self: Any, prompt: BasePromptTemplate, **prompt_args: Any
+) -> str:
"""Patch llm predictor predict."""
- assert isinstance(prompt, ChoiceSelectPrompt)
return "Doc: 2, Relevance: 5"
diff --git a/tests/indices/postprocessor/test_llm_rerank.py b/tests/indices/postprocessor/test_llm_rerank.py
index 3e06fcfc49..8a26cb86d2 100644
--- a/tests/indices/postprocessor/test_llm_rerank.py
+++ b/tests/indices/postprocessor/test_llm_rerank.py
@@ -1,19 +1,20 @@
"""Test LLM reranker."""
-from llama_index.indices.query.schema import QueryBundle
-from llama_index.prompts.prompts import Prompt
-from llama_index.llm_predictor import LLMPredictor
+from typing import Any, List
from unittest.mock import patch
-from typing import List, Any
-from llama_index.prompts.prompts import QuestionAnswerPrompt
+
from llama_index.indices.postprocessor.llm_rerank import LLMRerank
+from llama_index.indices.query.schema import QueryBundle
from llama_index.indices.service_context import ServiceContext
-from llama_index.schema import BaseNode, TextNode, NodeWithScore
+from llama_index.llm_predictor import LLMPredictor
+from llama_index.prompts import BasePromptTemplate
+from llama_index.schema import BaseNode, NodeWithScore, TextNode
-def mock_llmpredictor_predict(self: Any, prompt: Prompt, **prompt_args: Any) -> str:
+def mock_llmpredictor_predict(
+ self: Any, prompt: BasePromptTemplate, **prompt_args: Any
+) -> str:
"""Patch llm predictor predict."""
- assert isinstance(prompt, QuestionAnswerPrompt)
context_str = prompt_args["context_str"]
node_strs = context_str.split("\n")
node_to_choice_and_score = {
diff --git a/tests/indices/response/test_response_builder.py b/tests/indices/response/test_response_builder.py
index bb18561dcc..7f5223a5f3 100644
--- a/tests/indices/response/test_response_builder.py
+++ b/tests/indices/response/test_response_builder.py
@@ -6,7 +6,7 @@ from typing import List
from llama_index.constants import DEFAULT_CONTEXT_WINDOW, DEFAULT_NUM_OUTPUTS
from llama_index.indices.prompt_helper import PromptHelper
from llama_index.indices.service_context import ServiceContext
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import PromptTemplate
from llama_index.prompts.prompt_type import PromptType
from llama_index.response_synthesizers import ResponseMode, get_response_synthesizer
from llama_index.schema import Document
@@ -58,10 +58,14 @@ def test_compact_response(mock_service_context: ServiceContext) -> None:
# test response with ResponseMode.COMPACT
# NOTE: here we want to guarante that prompts have 0 extra tokens
mock_refine_prompt_tmpl = "{query_str}{existing_answer}{context_msg}"
- mock_refine_prompt = Prompt(mock_refine_prompt_tmpl, prompt_type=PromptType.REFINE)
+ mock_refine_prompt = PromptTemplate(
+ mock_refine_prompt_tmpl, prompt_type=PromptType.REFINE
+ )
mock_qa_prompt_tmpl = "{context_str}{query_str}"
- mock_qa_prompt = Prompt(mock_qa_prompt_tmpl, prompt_type=PromptType.QUESTION_ANSWER)
+ mock_qa_prompt = PromptTemplate(
+ mock_qa_prompt_tmpl, prompt_type=PromptType.QUESTION_ANSWER
+ )
# max input size is 11, prompt is two tokens (the query) --> 9 tokens
# --> padding is 1 --> 8 tokens
@@ -106,7 +110,9 @@ def test_accumulate_response(
# test response with ResponseMode.ACCUMULATE
# NOTE: here we want to guarante that prompts have 0 extra tokens
mock_qa_prompt_tmpl = "{context_str}{query_str}"
- mock_qa_prompt = Prompt(mock_qa_prompt_tmpl, prompt_type=PromptType.QUESTION_ANSWER)
+ mock_qa_prompt = PromptTemplate(
+ mock_qa_prompt_tmpl, prompt_type=PromptType.QUESTION_ANSWER
+ )
# max input size is 11, prompt is two tokens (the query) --> 9 tokens
# --> padding is 1 --> 8 tokens
@@ -163,7 +169,9 @@ def test_accumulate_response_async(
# test response with ResponseMode.ACCUMULATE
# NOTE: here we want to guarante that prompts have 0 extra tokens
mock_qa_prompt_tmpl = "{context_str}{query_str}"
- mock_qa_prompt = Prompt(mock_qa_prompt_tmpl, prompt_type=PromptType.QUESTION_ANSWER)
+ mock_qa_prompt = PromptTemplate(
+ mock_qa_prompt_tmpl, prompt_type=PromptType.QUESTION_ANSWER
+ )
# max input size is 11, prompt is two tokens (the query) --> 9 tokens
# --> padding is 1 --> 8 tokens
@@ -221,7 +229,9 @@ def test_accumulate_response_aget(
# test response with ResponseMode.ACCUMULATE
# NOTE: here we want to guarante that prompts have 0 extra tokens
mock_qa_prompt_tmpl = "{context_str}{query_str}"
- mock_qa_prompt = Prompt(mock_qa_prompt_tmpl, prompt_type=PromptType.QUESTION_ANSWER)
+ mock_qa_prompt = PromptTemplate(
+ mock_qa_prompt_tmpl, prompt_type=PromptType.QUESTION_ANSWER
+ )
# max input size is 11, prompt is two tokens (the query) --> 9 tokens
# --> padding is 1 --> 8 tokens
@@ -281,7 +291,9 @@ def test_accumulate_compact_response(patch_llm_predictor: None) -> None:
# test response with ResponseMode.ACCUMULATE
# NOTE: here we want to guarante that prompts have 0 extra tokens
mock_qa_prompt_tmpl = "{context_str}{query_str}"
- mock_qa_prompt = Prompt(mock_qa_prompt_tmpl, prompt_type=PromptType.QUESTION_ANSWER)
+ mock_qa_prompt = PromptTemplate(
+ mock_qa_prompt_tmpl, prompt_type=PromptType.QUESTION_ANSWER
+ )
# max input size is 11, prompt is two tokens (the query) --> 9 tokens
# --> padding is 1 --> 8 tokens
diff --git a/tests/indices/response/test_tree_summarize.py b/tests/indices/response/test_tree_summarize.py
index 5f451dcd79..017f4e4d72 100644
--- a/tests/indices/response/test_tree_summarize.py
+++ b/tests/indices/response/test_tree_summarize.py
@@ -8,7 +8,7 @@ import pytest
from llama_index.indices.prompt_helper import PromptHelper
from llama_index.response_synthesizers import TreeSummarize
from llama_index.indices.service_context import ServiceContext
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import PromptTemplate
from llama_index.prompts.prompt_type import PromptType
@@ -16,7 +16,9 @@ from llama_index.prompts.prompt_type import PromptType
def mock_service_context_merge_chunks(
mock_service_context: ServiceContext,
) -> ServiceContext:
- def mock_repack(prompt_template: Prompt, text_chunks: Sequence[str]) -> List[str]:
+ def mock_repack(
+ prompt_template: PromptTemplate, text_chunks: Sequence[str]
+ ) -> List[str]:
merged_chunks = []
for chunks in zip(*[iter(text_chunks)] * 2):
merged_chunks.append("\n".join(chunks))
@@ -30,7 +32,7 @@ def mock_service_context_merge_chunks(
def test_tree_summarize(mock_service_context_merge_chunks: ServiceContext) -> None:
mock_summary_prompt_tmpl = "{context_str}{query_str}"
- mock_summary_prompt = Prompt(
+ mock_summary_prompt = PromptTemplate(
mock_summary_prompt_tmpl, prompt_type=PromptType.SUMMARY
)
@@ -55,7 +57,7 @@ def test_tree_summarize_use_async(
mock_service_context_merge_chunks: ServiceContext,
) -> None:
mock_summary_prompt_tmpl = "{context_str}{query_str}"
- mock_summary_prompt = Prompt(
+ mock_summary_prompt = PromptTemplate(
mock_summary_prompt_tmpl, prompt_type=PromptType.SUMMARY
)
@@ -82,7 +84,7 @@ async def test_tree_summarize_async(
mock_service_context_merge_chunks: ServiceContext,
) -> None:
mock_summary_prompt_tmpl = "{context_str}{query_str}"
- mock_summary_prompt = Prompt(
+ mock_summary_prompt = PromptTemplate(
mock_summary_prompt_tmpl, prompt_type=PromptType.SUMMARY
)
diff --git a/tests/indices/test_prompt_helper.py b/tests/indices/test_prompt_helper.py
index 2946ee2fb0..6ec3da1410 100644
--- a/tests/indices/test_prompt_helper.py
+++ b/tests/indices/test_prompt_helper.py
@@ -1,11 +1,9 @@
"""Test PromptHelper."""
-from typing import cast
-from llama_index.bridge.langchain import PromptTemplate as LangchainPrompt
from llama_index.indices.prompt_helper import PromptHelper
from llama_index.indices.tree.utils import get_numbered_text_from_nodes
-from llama_index.prompts.base import Prompt
-from llama_index.prompts.utils import get_biggest_prompt, get_empty_prompt_txt
+from llama_index.prompts.base import PromptTemplate
+from llama_index.prompts.prompt_utils import get_biggest_prompt, get_empty_prompt_txt
from llama_index.schema import TextNode
from llama_index.text_splitter.utils import truncate_text
from tests.mock_utils.mock_utils import mock_tokenizer
@@ -14,7 +12,7 @@ from tests.mock_utils.mock_utils import mock_tokenizer
def test_get_chunk_size() -> None:
"""Test get chunk size given prompt."""
# test with 1 chunk
- prompt = Prompt("This is the prompt")
+ prompt = PromptTemplate("This is the prompt")
prompt_helper = PromptHelper(
context_window=11, num_output=1, chunk_overlap_ratio=0, tokenizer=mock_tokenizer
)
@@ -50,7 +48,7 @@ def test_get_chunk_size() -> None:
def test_get_text_splitter() -> None:
"""Test get text splitter."""
test_prompt_text = "This is the prompt{text}"
- test_prompt = Prompt(test_prompt_text)
+ test_prompt = PromptTemplate(test_prompt_text)
prompt_helper = PromptHelper(
context_window=11, num_output=1, chunk_overlap_ratio=0, tokenizer=mock_tokenizer
)
@@ -84,7 +82,7 @@ def test_get_text_splitter_partial() -> None:
# test without partially formatting
test_prompt_text = "This is the {foo} prompt{text}"
- test_prompt = Prompt(test_prompt_text)
+ test_prompt = PromptTemplate(test_prompt_text)
prompt_helper = PromptHelper(
context_window=11, num_output=1, chunk_overlap_ratio=0, tokenizer=mock_tokenizer
)
@@ -98,7 +96,7 @@ def test_get_text_splitter_partial() -> None:
assert truncated_text == "Hello world"
# test with partially formatting
- test_prompt = Prompt(test_prompt_text)
+ test_prompt = PromptTemplate(test_prompt_text)
test_prompt = test_prompt.partial_format(foo="bar")
prompt_helper = PromptHelper(
context_window=12, num_output=1, chunk_overlap_ratio=0, tokenizer=mock_tokenizer
@@ -118,7 +116,7 @@ def test_truncate() -> None:
"""Test truncate."""
# test prompt uses up one token
test_prompt_txt = "test{text}"
- test_prompt = Prompt(test_prompt_txt)
+ test_prompt = PromptTemplate(test_prompt_txt)
# set context_window=19
# For each text chunk, there's 4 tokens for text + 5 for the padding
prompt_helper = PromptHelper(
@@ -139,7 +137,7 @@ def test_get_numbered_text_from_nodes() -> None:
"""Test get_text_from_nodes."""
# test prompt uses up one token
test_prompt_txt = "test{text}"
- test_prompt = Prompt(test_prompt_txt)
+ test_prompt = PromptTemplate(test_prompt_txt)
# set context_window=17
# For each text chunk, there's 3 for text, 5 for padding (including number)
prompt_helper = PromptHelper(
@@ -159,7 +157,7 @@ def test_get_numbered_text_from_nodes() -> None:
def test_repack() -> None:
"""Test repack."""
test_prompt_text = "This is the prompt{text}"
- test_prompt = Prompt(test_prompt_text)
+ test_prompt = PromptTemplate(test_prompt_text)
prompt_helper = PromptHelper(
context_window=13,
num_output=1,
@@ -174,11 +172,9 @@ def test_repack() -> None:
def test_get_biggest_prompt() -> None:
"""Test get_biggest_prompt from PromptHelper."""
- prompt1 = Prompt("This is the prompt{text}")
- prompt2 = Prompt("This is the longer prompt{text}")
- prompt3 = Prompt("This is the {text}")
+ prompt1 = PromptTemplate("This is the prompt{text}")
+ prompt2 = PromptTemplate("This is the longer prompt{text}")
+ prompt3 = PromptTemplate("This is the {text}")
biggest_prompt = get_biggest_prompt([prompt1, prompt2, prompt3])
- lc_biggest_template = cast(LangchainPrompt, biggest_prompt.prompt).template
- prompt2_template = cast(LangchainPrompt, prompt2.prompt).template
- assert lc_biggest_template == prompt2_template
+ assert biggest_prompt == prompt2
diff --git a/tests/llm_predictor/test_base.py b/tests/llm_predictor/test_base.py
index 01ed7dca00..c1a07a33a6 100644
--- a/tests/llm_predictor/test_base.py
+++ b/tests/llm_predictor/test_base.py
@@ -2,10 +2,10 @@
from typing import Any
from unittest.mock import patch
-
from llama_index.llm_predictor.structured import LLMPredictor, StructuredLLMPredictor
+from llama_index.prompts import BasePromptTemplate
+from llama_index.prompts.base import PromptTemplate
from llama_index.types import BaseOutputParser
-from llama_index.prompts.prompts import Prompt, SimpleInputPrompt
try:
gptcache_installed = True
@@ -25,7 +25,7 @@ class MockOutputParser(BaseOutputParser):
return output
-def mock_llmpredictor_predict(prompt: Prompt, **prompt_args: Any) -> str:
+def mock_llmpredictor_predict(prompt: BasePromptTemplate, **prompt_args: Any) -> str:
"""Mock LLMPredictor predict."""
return prompt_args["query_str"]
@@ -36,12 +36,12 @@ def test_struct_llm_predictor(mock_init: Any, mock_predict: Any) -> None:
"""Test LLM predictor."""
llm_predictor = StructuredLLMPredictor()
output_parser = MockOutputParser()
- prompt = SimpleInputPrompt("{query_str}", output_parser=output_parser)
+ prompt = PromptTemplate("{query_str}", output_parser=output_parser)
llm_prediction = llm_predictor.predict(prompt, query_str="hello world")
assert llm_prediction == "hello world\nhello world"
# no change
- prompt = SimpleInputPrompt("{query_str}")
+ prompt = PromptTemplate("{query_str}")
llm_prediction = llm_predictor.predict(prompt, query_str="hello world")
assert llm_prediction == "hello world"
diff --git a/tests/llm_predictor/vellum/conftest.py b/tests/llm_predictor/vellum/conftest.py
index 29cfb14d7e..94faebbed5 100644
--- a/tests/llm_predictor/vellum/conftest.py
+++ b/tests/llm_predictor/vellum/conftest.py
@@ -1,26 +1,16 @@
-from typing import Optional, Type, Callable
+from typing import Optional, Callable
from unittest import mock
import pytest
-from llama_index import Prompt
from llama_index.callbacks import CallbackManager
from llama_index.llm_predictor.vellum import VellumPredictor, VellumPromptRegistry
-from llama_index.prompts.prompt_type import PromptType
+from llama_index.prompts.base import PromptTemplate
@pytest.fixture
-def dummy_prompt_class() -> Type[Prompt]:
- class DummyPrompt(Prompt):
- prompt_type = PromptType.CUSTOM
- input_variables = ["thing"]
-
- return DummyPrompt
-
-
-@pytest.fixture
-def dummy_prompt(dummy_prompt_class: Type[Prompt]) -> Prompt:
- return dummy_prompt_class(template="What's your favorite {thing}?")
+def dummy_prompt() -> PromptTemplate:
+ return PromptTemplate(template="What's your favorite {thing}?")
@pytest.fixture
diff --git a/tests/llm_predictor/vellum/test_predictor.py b/tests/llm_predictor/vellum/test_predictor.py
index 785c4518a7..0f2ff4042f 100644
--- a/tests/llm_predictor/vellum/test_predictor.py
+++ b/tests/llm_predictor/vellum/test_predictor.py
@@ -3,16 +3,14 @@ from unittest import mock
import pytest
-from llama_index import Prompt
-from llama_index.llm_predictor.vellum import (
- VellumPredictor,
-)
+from llama_index.llm_predictor.vellum import VellumPredictor
+from llama_index.prompts import BasePromptTemplate
def test_predict__basic(
mock_vellum_client_factory: Callable[..., mock.MagicMock],
vellum_predictor_factory: Callable[..., VellumPredictor],
- dummy_prompt: Prompt,
+ dummy_prompt: BasePromptTemplate,
) -> None:
"""When the Vellum API returns expected values, so should our predictor"""
@@ -31,7 +29,7 @@ def test_predict__basic(
def test_stream__basic(
mock_vellum_client_factory: Callable[..., mock.MagicMock],
vellum_predictor_factory: Callable[..., VellumPredictor],
- dummy_prompt: Prompt,
+ dummy_prompt: BasePromptTemplate,
) -> None:
"""When the Vellum API streams expected values, so should our predictor"""
diff --git a/tests/llm_predictor/vellum/test_prompt_registry.py b/tests/llm_predictor/vellum/test_prompt_registry.py
index 2f75715e5b..7350d8f500 100644
--- a/tests/llm_predictor/vellum/test_prompt_registry.py
+++ b/tests/llm_predictor/vellum/test_prompt_registry.py
@@ -1,16 +1,15 @@
-from typing import Callable, Type
+from typing import Callable
from unittest import mock
-from llama_index import Prompt
from llama_index.llm_predictor.vellum import (
VellumRegisteredPrompt,
VellumCompiledPrompt,
VellumPromptRegistry,
)
+from llama_index.prompts.base import PromptTemplate
def test_from_prompt__new(
- dummy_prompt_class: Type[Prompt],
mock_vellum_client_factory: Callable[..., mock.MagicMock],
vellum_prompt_registry_factory: Callable[..., VellumPromptRegistry],
) -> None:
@@ -18,7 +17,7 @@ def test_from_prompt__new(
from vellum.core import ApiError
- dummy_prompt = dummy_prompt_class(template="What's your favorite {thing}?")
+ dummy_prompt = PromptTemplate(template="What's your favorite {thing}?")
vellum_client = mock_vellum_client_factory()
@@ -31,13 +30,12 @@ def test_from_prompt__new(
def test_from_prompt__existing(
- dummy_prompt_class: Type[Prompt],
mock_vellum_client_factory: Callable[..., mock.MagicMock],
vellum_prompt_registry_factory: Callable[..., VellumPromptRegistry],
) -> None:
"""We shouldn't register a new prompt if a deployment id or name is provided"""
- dummy_prompt = dummy_prompt_class(
+ dummy_prompt = PromptTemplate(
template="What's your favorite {thing}?",
metadata={"vellum_deployment_id": "abc"},
)
diff --git a/tests/mock_utils/mock_predict.py b/tests/mock_utils/mock_predict.py
index 32cc806b2f..f3b5ac82eb 100644
--- a/tests/mock_utils/mock_predict.py
+++ b/tests/mock_utils/mock_predict.py
@@ -3,7 +3,9 @@
import json
from typing import Any, Dict
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import (
+ BasePromptTemplate,
+)
from llama_index.prompts.prompt_type import PromptType
from llama_index.token_counter.utils import mock_extract_keywords_response
@@ -150,50 +152,54 @@ def _mock_conversation(prompt_args: Dict) -> str:
return prompt_args["history"] + ":" + prompt_args["message"]
-def mock_llmpredictor_predict(prompt: Prompt, **prompt_args: Any) -> str:
+def mock_llmpredictor_predict(prompt: BasePromptTemplate, **prompt_args: Any) -> str:
"""Mock predict method of LLMPredictor.
Depending on the prompt, return response.
"""
- full_prompt_args = prompt.get_full_format_args(prompt_args)
- if prompt.prompt_type == PromptType.SUMMARY:
+ full_prompt_args = {
+ **prompt.kwargs,
+ **prompt_args,
+ }
+ prompt_type = prompt.metadata["prompt_type"]
+ if prompt_type == PromptType.SUMMARY:
response = _mock_summary_predict(full_prompt_args)
- elif prompt.prompt_type == PromptType.TREE_INSERT:
+ elif prompt_type == PromptType.TREE_INSERT:
response = _mock_insert_predict()
- elif prompt.prompt_type == PromptType.TREE_SELECT:
+ elif prompt_type == PromptType.TREE_SELECT:
response = _mock_query_select()
- elif prompt.prompt_type == PromptType.REFINE:
+ elif prompt_type == PromptType.REFINE:
response = _mock_refine(full_prompt_args)
- elif prompt.prompt_type == PromptType.QUESTION_ANSWER:
+ elif prompt_type == PromptType.QUESTION_ANSWER:
response = _mock_answer(full_prompt_args)
- elif prompt.prompt_type == PromptType.KEYWORD_EXTRACT:
+ elif prompt_type == PromptType.KEYWORD_EXTRACT:
response = _mock_keyword_extract(full_prompt_args)
- elif prompt.prompt_type == PromptType.QUERY_KEYWORD_EXTRACT:
+ elif prompt_type == PromptType.QUERY_KEYWORD_EXTRACT:
response = _mock_query_keyword_extract(full_prompt_args)
- elif prompt.prompt_type == PromptType.SCHEMA_EXTRACT:
+ elif prompt_type == PromptType.SCHEMA_EXTRACT:
response = _mock_schema_extract(full_prompt_args)
- elif prompt.prompt_type == PromptType.TEXT_TO_SQL:
+ elif prompt_type == PromptType.TEXT_TO_SQL:
response = _mock_text_to_sql(full_prompt_args)
- elif prompt.prompt_type == PromptType.KNOWLEDGE_TRIPLET_EXTRACT:
+ elif prompt_type == PromptType.KNOWLEDGE_TRIPLET_EXTRACT:
response = _mock_kg_triplet_extract(full_prompt_args)
- elif prompt.prompt_type == PromptType.SIMPLE_INPUT:
+ elif prompt_type == PromptType.SIMPLE_INPUT:
response = _mock_input(full_prompt_args)
- elif prompt.prompt_type == PromptType.SINGLE_SELECT:
+ elif prompt_type == PromptType.SINGLE_SELECT:
response = _mock_single_select()
- elif prompt.prompt_type == PromptType.MULTI_SELECT:
+ elif prompt_type == PromptType.MULTI_SELECT:
response = _mock_multi_select(full_prompt_args)
- elif prompt.prompt_type == PromptType.SUB_QUESTION:
+ elif prompt_type == PromptType.SUB_QUESTION:
response = _mock_sub_questions()
- elif prompt.prompt_type == PromptType.PANDAS:
+ elif prompt_type == PromptType.PANDAS:
response = _mock_pandas(full_prompt_args)
- elif prompt.prompt_type == PromptType.SQL_RESPONSE_SYNTHESIS:
+ elif prompt_type == PromptType.SQL_RESPONSE_SYNTHESIS:
response = _mock_sql_response_synthesis(full_prompt_args)
- elif prompt.prompt_type == PromptType.DECOMPOSE:
+ elif prompt_type == PromptType.DECOMPOSE:
response = _mock_decompose_query(full_prompt_args)
- elif prompt.prompt_type == PromptType.CHOICE_SELECT:
+ elif prompt_type == PromptType.CHOICE_SELECT:
response = _mock_choice_select(full_prompt_args)
- elif prompt.prompt_type == PromptType.CONVERSATION:
+ elif prompt_type == PromptType.CONVERSATION:
response = _mock_conversation(full_prompt_args)
else:
response = str(full_prompt_args)
@@ -201,7 +207,9 @@ def mock_llmpredictor_predict(prompt: Prompt, **prompt_args: Any) -> str:
return response
-def patch_llmpredictor_predict(self: Any, prompt: Prompt, **prompt_args: Any) -> str:
+def patch_llmpredictor_predict(
+ self: Any, prompt: BasePromptTemplate, **prompt_args: Any
+) -> str:
"""Mock predict method of LLMPredictor.
Depending on the prompt, return response.
@@ -211,12 +219,14 @@ def patch_llmpredictor_predict(self: Any, prompt: Prompt, **prompt_args: Any) ->
async def patch_llmpredictor_apredict(
- self: Any, prompt: Prompt, **prompt_args: Any
+ self: Any, prompt: BasePromptTemplate, **prompt_args: Any
) -> str:
"""Mock apredict method of LLMPredictor."""
return patch_llmpredictor_predict(self, prompt, **prompt_args)
-async def mock_llmpredictor_apredict(prompt: Prompt, **prompt_args: Any) -> str:
+async def mock_llmpredictor_apredict(
+ prompt: BasePromptTemplate, **prompt_args: Any
+) -> str:
"""Mock apredict method of LLMPredictor."""
return mock_llmpredictor_predict(prompt, **prompt_args)
diff --git a/tests/mock_utils/mock_prompts.py b/tests/mock_utils/mock_prompts.py
index 5835c90293..390359c9e7 100644
--- a/tests/mock_utils/mock_prompts.py
+++ b/tests/mock_utils/mock_prompts.py
@@ -1,65 +1,77 @@
"""Mock prompt utils."""
-from llama_index.prompts.base import Prompt
+from llama_index.prompts.base import PromptTemplate
from llama_index.prompts.prompt_type import PromptType
MOCK_SUMMARY_PROMPT_TMPL = "{context_str}\n"
-MOCK_SUMMARY_PROMPT = Prompt(MOCK_SUMMARY_PROMPT_TMPL, prompt_type=PromptType.SUMMARY)
+MOCK_SUMMARY_PROMPT = PromptTemplate(
+ MOCK_SUMMARY_PROMPT_TMPL, prompt_type=PromptType.SUMMARY
+)
MOCK_INSERT_PROMPT_TMPL = "{num_chunks}\n{context_list}{new_chunk_text}\n"
-MOCK_INSERT_PROMPT = Prompt(MOCK_INSERT_PROMPT_TMPL, prompt_type=PromptType.TREE_INSERT)
+MOCK_INSERT_PROMPT = PromptTemplate(
+ MOCK_INSERT_PROMPT_TMPL, prompt_type=PromptType.TREE_INSERT
+)
# # single choice
MOCK_QUERY_PROMPT_TMPL = "{num_chunks}\n" "{context_list}\n" "{query_str}'\n"
-MOCK_QUERY_PROMPT = Prompt(MOCK_QUERY_PROMPT_TMPL, prompt_type=PromptType.TREE_SELECT)
+MOCK_QUERY_PROMPT = PromptTemplate(
+ MOCK_QUERY_PROMPT_TMPL, prompt_type=PromptType.TREE_SELECT
+)
MOCK_REFINE_PROMPT_TMPL = "{query_str}\n" "{existing_answer}\n" "{context_msg}\n"
-MOCK_REFINE_PROMPT = Prompt(MOCK_REFINE_PROMPT_TMPL, prompt_type=PromptType.REFINE)
+MOCK_REFINE_PROMPT = PromptTemplate(
+ MOCK_REFINE_PROMPT_TMPL, prompt_type=PromptType.REFINE
+)
MOCK_TEXT_QA_PROMPT_TMPL = "{context_str}\n" "{query_str}\n"
-MOCK_TEXT_QA_PROMPT = Prompt(
+MOCK_TEXT_QA_PROMPT = PromptTemplate(
MOCK_TEXT_QA_PROMPT_TMPL, prompt_type=PromptType.QUESTION_ANSWER
)
MOCK_KEYWORD_EXTRACT_PROMPT_TMPL = "{max_keywords}\n{text}\n"
-MOCK_KEYWORD_EXTRACT_PROMPT = Prompt(
+MOCK_KEYWORD_EXTRACT_PROMPT = PromptTemplate(
MOCK_KEYWORD_EXTRACT_PROMPT_TMPL, prompt_type=PromptType.KEYWORD_EXTRACT
)
# TODO: consolidate with keyword extract
MOCK_QUERY_KEYWORD_EXTRACT_PROMPT_TMPL = "{max_keywords}\n{question}\n"
-MOCK_QUERY_KEYWORD_EXTRACT_PROMPT = Prompt(
+MOCK_QUERY_KEYWORD_EXTRACT_PROMPT = PromptTemplate(
MOCK_QUERY_KEYWORD_EXTRACT_PROMPT_TMPL, prompt_type=PromptType.QUERY_KEYWORD_EXTRACT
)
MOCK_SCHEMA_EXTRACT_PROMPT_TMPL = "{text}\n{schema}"
-MOCK_SCHEMA_EXTRACT_PROMPT = Prompt(
+MOCK_SCHEMA_EXTRACT_PROMPT = PromptTemplate(
MOCK_SCHEMA_EXTRACT_PROMPT_TMPL, prompt_type=PromptType.SCHEMA_EXTRACT
)
MOCK_TEXT_TO_SQL_PROMPT_TMPL = "{dialect}\n{schema}\n{query_str}"
-MOCK_TEXT_TO_SQL_PROMPT = Prompt(
+MOCK_TEXT_TO_SQL_PROMPT = PromptTemplate(
MOCK_TEXT_TO_SQL_PROMPT_TMPL, prompt_type=PromptType.TEXT_TO_SQL
)
MOCK_TABLE_CONTEXT_PROMPT_TMPL = "{schema}\n{context_str}\n{query_str}"
-MOCK_TABLE_CONTEXT_PROMPT = Prompt(
+MOCK_TABLE_CONTEXT_PROMPT = PromptTemplate(
MOCK_TABLE_CONTEXT_PROMPT_TMPL, prompt_type=PromptType.TABLE_CONTEXT
)
MOCK_KG_TRIPLET_EXTRACT_PROMPT_TMPL = "{max_knowledge_triplets}\n{text}"
-MOCK_KG_TRIPLET_EXTRACT_PROMPT = Prompt(
+MOCK_KG_TRIPLET_EXTRACT_PROMPT = PromptTemplate(
MOCK_KG_TRIPLET_EXTRACT_PROMPT_TMPL,
prompt_type=PromptType.KNOWLEDGE_TRIPLET_EXTRACT,
)
MOCK_INPUT_PROMPT_TMPL = "{query_str}"
-MOCK_INPUT_PROMPT = Prompt(MOCK_INPUT_PROMPT_TMPL, prompt_type=PromptType.SIMPLE_INPUT)
+MOCK_INPUT_PROMPT = PromptTemplate(
+ MOCK_INPUT_PROMPT_TMPL, prompt_type=PromptType.SIMPLE_INPUT
+)
MOCK_PANDAS_PROMPT_TMPL = "{query_str}\n{df_str}\n{instruction_str}"
-MOCK_PANDAS_PROMPT = Prompt(MOCK_PANDAS_PROMPT_TMPL, prompt_type=PromptType.PANDAS)
+MOCK_PANDAS_PROMPT = PromptTemplate(
+ MOCK_PANDAS_PROMPT_TMPL, prompt_type=PromptType.PANDAS
+)
diff --git a/tests/prompts/test_base.py b/tests/prompts/test_base.py
index aa1507420b..8fef4f49dd 100644
--- a/tests/prompts/test_base.py
+++ b/tests/prompts/test_base.py
@@ -1,97 +1,126 @@
"""Test prompts."""
-from unittest.mock import MagicMock
-import pytest
+from llama_index.bridge.langchain import BaseLanguageModel
+from llama_index.bridge.langchain import ConditionalPromptSelector as LangchainSelector
+from llama_index.bridge.langchain import FakeListLLM
+from llama_index.bridge.langchain import PromptTemplate as LangchainTemplate
+from llama_index.llms import MockLLM
+from llama_index.llms.base import ChatMessage, MessageRole
+from llama_index.llms.langchain import LangChainLLM
+from llama_index.prompts import (
+ ChatPromptTemplate,
+ LangchainPromptTemplate,
+ PromptTemplate,
+ SelectorPromptTemplate,
+)
+from llama_index.prompts.prompt_type import PromptType
+
+
+def test_template() -> None:
+ """Test partial format."""
+ prompt_txt = "hello {text} {foo}"
+ prompt = PromptTemplate(prompt_txt)
-from llama_index.bridge.langchain import PromptTemplate
-from llama_index.llms.base import LLM
-from llama_index.llms.openai import OpenAI
-from llama_index.prompts.base import Prompt
-from llama_index.prompts.prompt_selector import PromptSelector
+ prompt_fmt = prompt.partial_format(foo="bar")
+ assert isinstance(prompt_fmt, PromptTemplate)
+ assert prompt_fmt.format(text="world") == "hello world bar"
-def is_openai(llm: LLM) -> bool:
- """Test condition."""
- return isinstance(llm, OpenAI)
+ assert prompt_fmt.format_messages(text="world") == [
+ ChatMessage(content="hello world bar", role=MessageRole.USER)
+ ]
+
+
+def test_chat_template() -> None:
+ chat_template = ChatPromptTemplate(
+ message_templates=[
+ ChatMessage(
+ content="This is a system message with a {sys_param}",
+ role=MessageRole.SYSTEM,
+ ),
+ ChatMessage(content="hello {text} {foo}", role=MessageRole.USER),
+ ],
+ prompt_type=PromptType.CONVERSATION,
+ )
+ partial_template = chat_template.partial_format(sys_param="sys_arg")
+ messages = partial_template.format_messages(text="world", foo="bar")
-def test_partial_format() -> None:
- """Test partial format."""
- prompt_txt = "hello {text} {foo}"
- prompt = Prompt(prompt_txt)
+ assert messages[0] == ChatMessage(
+ content="This is a system message with a sys_arg", role=MessageRole.SYSTEM
+ )
- prompt_fmt = prompt.partial_format(foo="bar")
+ assert partial_template.format(text="world", foo="bar") == (
+ "system: This is a system message with a sys_arg\n"
+ "user: hello world bar\n"
+ "assistant: "
+ )
- assert isinstance(prompt_fmt, Prompt)
- assert prompt_fmt.format(text="world") == "hello world bar"
+def test_selector_template() -> None:
+ default_template = PromptTemplate("hello {text} {foo}")
+ chat_template = ChatPromptTemplate(
+ message_templates=[
+ ChatMessage(
+ content="This is a system message with a {sys_param}",
+ role=MessageRole.SYSTEM,
+ ),
+ ChatMessage(content="hello {text} {foo}", role=MessageRole.USER),
+ ],
+ prompt_type=PromptType.CONVERSATION,
+ )
-def test_from_prompt() -> None:
- """Test new prompt from a partially formatted prompt."""
- prompt_txt = "hello {text} {foo}"
- prompt = Prompt(prompt_txt)
- prompt_fmt = prompt.partial_format(foo="bar")
+ selector_template = SelectorPromptTemplate(
+ default_template=default_template,
+ conditionals=[
+ (lambda llm: isinstance(llm, MockLLM), chat_template),
+ ],
+ )
- prompt_new = Prompt.from_prompt(prompt_fmt)
- assert isinstance(prompt_new, Prompt)
+ partial_template = selector_template.partial_format(text="world", foo="bar")
- assert prompt_new.format(text="world2") == "hello world2 bar"
+ prompt = partial_template.format()
+ assert prompt == "hello world bar"
+ messages = partial_template.format_messages(llm=MockLLM(), sys_param="sys_arg")
+ assert messages[0] == ChatMessage(
+ content="This is a system message with a sys_arg", role=MessageRole.SYSTEM
+ )
-def test_from_langchain_prompt() -> None:
- """Test from langchain prompt."""
- prompt_txt = "hello {text} {foo}"
- prompt = PromptTemplate(input_variables=["text", "foo"], template=prompt_txt)
- prompt_new = Prompt.from_langchain_prompt(prompt)
- assert isinstance(prompt_new, Prompt)
- assert prompt_new.prompt == prompt
- assert prompt_new.format(text="world2", foo="bar") == "hello world2 bar"
+def test_langchain_template() -> None:
+ lc_template = LangchainTemplate.from_template("hello {text} {foo}")
+ template = LangchainPromptTemplate(lc_template)
- # test errors if we specify both template and langchain prompt
- with pytest.raises(ValueError):
- prompt_txt = "hello {text} {foo}"
- prompt = PromptTemplate(input_variables=["text", "foo"], template=prompt_txt)
- Prompt(template=prompt_txt, langchain_prompt=prompt)
+ template_fmt = template.partial_format(foo="bar")
+ assert isinstance(template, LangchainPromptTemplate)
+ assert template_fmt.format(text="world") == "hello world bar"
-def test_from_langchain_prompt_selector() -> None:
- """Test from langchain prompt selector."""
- prompt_txt = "hello {text} {foo}"
- prompt_txt_2 = "world {text} {foo}"
- prompt = PromptTemplate(input_variables=["text", "foo"], template=prompt_txt)
- prompt_2 = PromptTemplate(input_variables=["text", "foo"], template=prompt_txt_2)
+ assert template_fmt.format_messages(text="world") == [
+ ChatMessage(content="hello world bar", role=MessageRole.USER)
+ ]
- test_prompt_selector = PromptSelector(
- default_prompt=prompt, conditionals=[(is_openai, prompt_2)]
- )
- test_llm = MagicMock(spec=OpenAI)
+def test_langchain_selector_template() -> None:
+ lc_llm = FakeListLLM(responses=["test"])
+ mock_llm = LangChainLLM(llm=lc_llm)
+
+ def is_mock(llm: BaseLanguageModel) -> bool:
+ return llm == lc_llm
- prompt_new = Prompt.from_langchain_prompt_selector(test_prompt_selector)
- assert isinstance(prompt_new, Prompt)
- assert prompt_new.prompt == prompt
- assert prompt_new.format(text="world2", foo="bar") == "hello world2 bar"
- assert (
- prompt_new.format(llm=test_llm, text="world2", foo="bar") == "world world2 bar"
+ default_lc_template = LangchainTemplate.from_template("hello {text} {foo}")
+ conditionals = [
+ (is_mock, LangchainTemplate.from_template("hello {text} {foo} mock")),
+ ]
+
+ lc_selector = LangchainSelector(
+ default_prompt=default_lc_template, conditionals=conditionals
)
+ template = LangchainPromptTemplate(selector=lc_selector)
+
+ template_fmt = template.partial_format(foo="bar")
+ assert isinstance(template, LangchainPromptTemplate)
- test_lc_prompt = prompt_new.get_langchain_prompt(llm=test_llm)
- assert test_lc_prompt == prompt_2
- test_lc_prompt = prompt_new.get_langchain_prompt(llm=None)
- assert test_lc_prompt == prompt
-
- # test errors if langchain prompt input var doesn't match
- with pytest.raises(ValueError):
- prompt_txt = "hello {text} {foo}"
- prompt_txt_2 = "world {text} {foo} {tmp}"
- prompt = PromptTemplate(input_variables=["text", "foo"], template=prompt_txt)
- prompt_2 = PromptTemplate(
- input_variables=["text", "foo", "tmp"], template=prompt_txt_2
- )
-
- test_prompt_selector = PromptSelector(
- prompt=prompt, conditionals=([is_openai], [prompt_2])
- )
- prompt_new = Prompt.from_langchain_prompt_selector(test_prompt_selector)
+ assert template_fmt.format(llm=mock_llm, text="world") == "hello world bar mock"
diff --git a/tests/prompts/test_utils.py b/tests/prompts/test_utils.py
new file mode 100644
index 0000000000..f09ffc2004
--- /dev/null
+++ b/tests/prompts/test_utils.py
@@ -0,0 +1,7 @@
+from llama_index.prompts.utils import get_template_vars
+
+
+def test_get_template_vars() -> None:
+ template = "hello {text} {foo}"
+ template_vars = get_template_vars(template)
+ assert template_vars == ["text", "foo"]
--
GitLab