From 606605925d78d8acc14570c581ca7dfd55b0d4e9 Mon Sep 17 00:00:00 2001 From: Jonathan Pedoeem Date: Thu, 16 Mar 2023 20:05:23 -0400 Subject: [PATCH] Adding ability to `return_pl_id` to all PromptLayer Models in LangChain (#1699) PromptLayer now has support for [several different tracking features.](https://magniv.notion.site/Track-4deee1b1f7a34c1680d085f82567dab9) In order to use any of these features you need to have a request id associated with the request. In this PR we add a boolean argument called `return_pl_id` which will add `pl_request_id` to the `generation_info` dictionary associated with a generation. We also updated the relevant documentation. --- docs/ecosystem/promptlayer.md | 18 ++++- .../examples/promptlayer_chatopenai.ipynb | 38 +++++++++- .../integrations/promptlayer_openai.ipynb | 33 ++++++++- langchain/chat_models/promptlayer_openai.py | 27 +++++-- langchain/llms/promptlayer_openai.py | 71 +++++++++++++++---- 5 files changed, 165 insertions(+), 22 deletions(-) diff --git a/docs/ecosystem/promptlayer.md b/docs/ecosystem/promptlayer.md index 06e332c6..fb2bef16 100644 --- a/docs/ecosystem/promptlayer.md +++ b/docs/ecosystem/promptlayer.md @@ -25,9 +25,25 @@ from langchain.llms import PromptLayerOpenAI llm = PromptLayerOpenAI(pl_tags=["langchain-requests", "chatbot"]) ``` +To get the PromptLayer request id, use the argument `return_pl_id` when instanializing the LLM +```python +from langchain.llms import PromptLayerOpenAI +llm = PromptLayerOpenAI(return_pl_id=True) +``` +This will add the PromptLayer request ID in the `generation_info` field of the `Generation` returned when using `.generate` or `.agenerate` + +For example: +```python +llm_results = llm.generate(["hello world"]) +for res in llm_results.generations: + print("pl request id: ", res[0].generation_info["pl_request_id"]) +``` +You can use the PromptLayer request ID to add a prompt, score, or other metadata to your request. [Read more about it here](https://magniv.notion.site/Track-4deee1b1f7a34c1680d085f82567dab9). + This LLM is identical to the [OpenAI LLM](./openai), except that - all your requests will be logged to your PromptLayer account - you can add `pl_tags` when instantializing to tag your requests on PromptLayer +- you can add `return_pl_id` when instantializing to return a PromptLayer request id to use [while tracking requests](https://magniv.notion.site/Track-4deee1b1f7a34c1680d085f82567dab9). -PromptLayer also provides native wrappers for [`PromptLayerChatOpenAI`](../modules/chat/examples/promptlayer_chat_openai.ipynb) +PromptLayer also provides native wrappers for [`PromptLayerChatOpenAI`](../modules/chat/examples/promptlayer_chat_openai.ipynb) and `PromptLayerOpenAIChat` diff --git a/docs/modules/chat/examples/promptlayer_chatopenai.ipynb b/docs/modules/chat/examples/promptlayer_chatopenai.ipynb index eb1a0675..4f7c4620 100644 --- a/docs/modules/chat/examples/promptlayer_chatopenai.ipynb +++ b/docs/modules/chat/examples/promptlayer_chatopenai.ipynb @@ -123,6 +123,40 @@ "id": "05e9e2fe", "metadata": {}, "source": [] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "c43803d1", + "metadata": {}, + "source": [ + "## Using PromptLayer Track\n", + "If you would like to use any of the [PromptLayer tracking features](https://magniv.notion.site/Track-4deee1b1f7a34c1680d085f82567dab9), you need to pass the argument `return_pl_id` when instantializing the PromptLayer LLM to get the request id. " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b7d4db01", + "metadata": {}, + "outputs": [], + "source": [ + "chat = PromptLayerChatOpenAI(return_pl_id=True)\n", + "chat_results = chat.generate([[HumanMessage(content=\"I am a cat and I want\")]])\n", + "\n", + "for res in chat_results.generations:\n", + " pl_request_id = res[0].generation_info[\"pl_request_id\"]\n", + " promptlayer.track.score(request_id=pl_request_id, score=100)" + ] + }, + { + "cell_type": "markdown", + "id": "13e56507", + "metadata": {}, + "source": [ + "Using this allows you to track the performance of your model in the PromptLayer dashboard. If you are using a prompt template, you can attach a template to a request as well.\n", + "Overall, this gives you the opportunity to track the performance of different templates and models in the PromptLayer dashboard." + ] } ], "metadata": { @@ -141,11 +175,11 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.8.8" + "version": "3.8.8 (default, Apr 13 2021, 12:59:45) \n[Clang 10.0.0 ]" }, "vscode": { "interpreter": { - "hash": "c4fe2cd85a8d9e8baaec5340ce66faff1c77581a9f43e6c45e85e09b6fced008" + "hash": "8a5edab282632443219e051e4ade2d1d5bbc671c781051bf1437897cbdfea0f1" } } }, diff --git a/docs/modules/llms/integrations/promptlayer_openai.ipynb b/docs/modules/llms/integrations/promptlayer_openai.ipynb index a50d6670..60ba4a3b 100644 --- a/docs/modules/llms/integrations/promptlayer_openai.ipynb +++ b/docs/modules/llms/integrations/promptlayer_openai.ipynb @@ -119,10 +119,39 @@ ] }, { + "attachments": {}, "cell_type": "markdown", "id": "05e9e2fe", "metadata": {}, - "source": [] + "source": [ + "## Using PromptLayer Track\n", + "If you would like to use any of the [PromptLayer tracking features](https://magniv.notion.site/Track-4deee1b1f7a34c1680d085f82567dab9), you need to pass the argument `return_pl_id` when instantializing the PromptLayer LLM to get the request id. " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1a7315b9", + "metadata": {}, + "outputs": [], + "source": [ + "llm = PromptLayerOpenAI(return_pl_id=True)\n", + "llm_results = llm.generate([\"Tell me a joke\"])\n", + "\n", + "for res in llm_results.generations:\n", + " pl_request_id = res[0].generation_info[\"pl_request_id\"]\n", + " promptlayer.track.score(request_id=pl_request_id, score=100)" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "7eb19139", + "metadata": {}, + "source": [ + "Using this allows you to track the performance of your model in the PromptLayer dashboard. If you are using a prompt template, you can attach a template to a request as well.\n", + "Overall, this gives you the opportunity to track the performance of different templates and models in the PromptLayer dashboard." + ] } ], "metadata": { @@ -145,7 +174,7 @@ }, "vscode": { "interpreter": { - "hash": "c4fe2cd85a8d9e8baaec5340ce66faff1c77581a9f43e6c45e85e09b6fced008" + "hash": "8a5edab282632443219e051e4ade2d1d5bbc671c781051bf1437897cbdfea0f1" } } }, diff --git a/langchain/chat_models/promptlayer_openai.py b/langchain/chat_models/promptlayer_openai.py index 6eb7ce56..a42acbe2 100644 --- a/langchain/chat_models/promptlayer_openai.py +++ b/langchain/chat_models/promptlayer_openai.py @@ -17,8 +17,12 @@ class PromptLayerChatOpenAI(ChatOpenAI, BaseModel): promptlayer key respectively. All parameters that can be passed to the OpenAI LLM can also - be passed here. The PromptLayerChatOpenAI LLM adds an extra - ``pl_tags`` parameter that can be used to tag the request. + be passed here. The PromptLayerChatOpenAI adds to optional + parameters: + ``pl_tags``: List of strings to tag the request with. + ``return_pl_id``: If True, the PromptLayer request ID will be + returned in the ``generation_info`` field of the + ``Generation`` object. Example: .. code-block:: python @@ -28,6 +32,7 @@ class PromptLayerChatOpenAI(ChatOpenAI, BaseModel): """ pl_tags: Optional[List[str]] + return_pl_id: Optional[bool] = False def _generate( self, messages: List[BaseMessage], stop: Optional[List[str]] = None @@ -43,7 +48,7 @@ class PromptLayerChatOpenAI(ChatOpenAI, BaseModel): response_dict, params = super()._create_message_dicts( [generation.message], stop ) - promptlayer_api_request( + pl_request_id = promptlayer_api_request( "langchain.PromptLayerChatOpenAI", "langchain", message_dicts, @@ -53,7 +58,14 @@ class PromptLayerChatOpenAI(ChatOpenAI, BaseModel): request_start_time, request_end_time, get_api_key(), + return_pl_id=self.return_pl_id, ) + if self.return_pl_id: + if generation.generation_info is None or not isinstance( + generation.generation_info, dict + ): + generation.generation_info = {} + generation.generation_info["pl_request_id"] = pl_request_id return generated_responses async def _agenerate( @@ -70,7 +82,7 @@ class PromptLayerChatOpenAI(ChatOpenAI, BaseModel): response_dict, params = super()._create_message_dicts( [generation.message], stop ) - promptlayer_api_request( + pl_request_id = promptlayer_api_request( "langchain.PromptLayerChatOpenAI.async", "langchain", message_dicts, @@ -80,5 +92,12 @@ class PromptLayerChatOpenAI(ChatOpenAI, BaseModel): request_start_time, request_end_time, get_api_key(), + return_pl_id=self.return_pl_id, ) + if self.return_pl_id: + if generation.generation_info is None or not isinstance( + generation.generation_info, dict + ): + generation.generation_info = {} + generation.generation_info["pl_request_id"] = pl_request_id return generated_responses diff --git a/langchain/llms/promptlayer_openai.py b/langchain/llms/promptlayer_openai.py index cf30270f..1cfb3177 100644 --- a/langchain/llms/promptlayer_openai.py +++ b/langchain/llms/promptlayer_openai.py @@ -17,8 +17,12 @@ class PromptLayerOpenAI(OpenAI, BaseModel): promptlayer key respectively. All parameters that can be passed to the OpenAI LLM can also - be passed here. The PromptLayerOpenAI LLM adds an extra - ``pl_tags`` parameter that can be used to tag the request. + be passed here. The PromptLayerOpenAI LLM adds two optional + parameters: + ``pl_tags``: List of strings to tag the request with. + ``return_pl_id``: If True, the PromptLayer request ID will be + returned in the ``generation_info`` field of the + ``Generation`` object. Example: .. code-block:: python @@ -28,6 +32,7 @@ class PromptLayerOpenAI(OpenAI, BaseModel): """ pl_tags: Optional[List[str]] + return_pl_id: Optional[bool] = False def _generate( self, prompts: List[str], stop: Optional[List[str]] = None @@ -40,11 +45,12 @@ class PromptLayerOpenAI(OpenAI, BaseModel): request_end_time = datetime.datetime.now().timestamp() for i in range(len(prompts)): prompt = prompts[i] + generation = generated_responses.generations[i][0] resp = { - "text": generated_responses.generations[i][0].text, + "text": generation.text, "llm_output": generated_responses.llm_output, } - promptlayer_api_request( + pl_request_id = promptlayer_api_request( "langchain.PromptLayerOpenAI", "langchain", [prompt], @@ -54,7 +60,14 @@ class PromptLayerOpenAI(OpenAI, BaseModel): request_start_time, request_end_time, get_api_key(), + return_pl_id=self.return_pl_id, ) + if self.return_pl_id: + if generation.generation_info is None or not isinstance( + generation.generation_info, dict + ): + generation.generation_info = {} + generation.generation_info["pl_request_id"] = pl_request_id return generated_responses async def _agenerate( @@ -67,11 +80,12 @@ class PromptLayerOpenAI(OpenAI, BaseModel): request_end_time = datetime.datetime.now().timestamp() for i in range(len(prompts)): prompt = prompts[i] + generation = generated_responses.generations[i][0] resp = { - "text": generated_responses.generations[i][0].text, + "text": generation.text, "llm_output": generated_responses.llm_output, } - promptlayer_api_request( + pl_request_id = promptlayer_api_request( "langchain.PromptLayerOpenAI.async", "langchain", [prompt], @@ -81,7 +95,14 @@ class PromptLayerOpenAI(OpenAI, BaseModel): request_start_time, request_end_time, get_api_key(), + return_pl_id=self.return_pl_id, ) + if self.return_pl_id: + if generation.generation_info is None or not isinstance( + generation.generation_info, dict + ): + generation.generation_info = {} + generation.generation_info["pl_request_id"] = pl_request_id return generated_responses @@ -94,8 +115,12 @@ class PromptLayerOpenAIChat(OpenAIChat, BaseModel): promptlayer key respectively. All parameters that can be passed to the OpenAIChat LLM can also - be passed here. The PromptLayerOpenAIChat LLM adds an extra - ``pl_tags`` parameter that can be used to tag the request. + be passed here. The PromptLayerOpenAIChat adds two optional + parameters: + ``pl_tags``: List of strings to tag the request with. + ``return_pl_id``: If True, the PromptLayer request ID will be + returned in the ``generation_info`` field of the + ``Generation`` object. Example: .. code-block:: python @@ -105,6 +130,7 @@ class PromptLayerOpenAIChat(OpenAIChat, BaseModel): """ pl_tags: Optional[List[str]] + return_pl_id: Optional[bool] = False def _generate( self, prompts: List[str], stop: Optional[List[str]] = None @@ -117,11 +143,12 @@ class PromptLayerOpenAIChat(OpenAIChat, BaseModel): request_end_time = datetime.datetime.now().timestamp() for i in range(len(prompts)): prompt = prompts[i] + generation = generated_responses.generations[i][0] resp = { - "text": generated_responses.generations[i][0].text, + "text": generation.text, "llm_output": generated_responses.llm_output, } - promptlayer_api_request( + pl_request_id = promptlayer_api_request( "langchain.PromptLayerOpenAIChat", "langchain", [prompt], @@ -131,7 +158,14 @@ class PromptLayerOpenAIChat(OpenAIChat, BaseModel): request_start_time, request_end_time, get_api_key(), + return_pl_id=self.return_pl_id, ) + if self.return_pl_id: + if generation.generation_info is None or not isinstance( + generation.generation_info, dict + ): + generation.generation_info = {} + generation.generation_info["pl_request_id"] = pl_request_id return generated_responses async def _agenerate( @@ -144,16 +178,27 @@ class PromptLayerOpenAIChat(OpenAIChat, BaseModel): request_end_time = datetime.datetime.now().timestamp() for i in range(len(prompts)): prompt = prompts[i] - resp = generated_responses.generations[i] - promptlayer_api_request( + generation = generated_responses.generations[i][0] + resp = { + "text": generation.text, + "llm_output": generated_responses.llm_output, + } + pl_request_id = promptlayer_api_request( "langchain.PromptLayerOpenAIChat.async", "langchain", [prompt], self._identifying_params, self.pl_tags, - resp[0].text, + resp, request_start_time, request_end_time, get_api_key(), + return_pl_id=self.return_pl_id, ) + if self.return_pl_id: + if generation.generation_info is None or not isinstance( + generation.generation_info, dict + ): + generation.generation_info = {} + generation.generation_info["pl_request_id"] = pl_request_id return generated_responses