How can I set model hyperparameters when using AzureChatOpenAI? #10954
Replies: 1 comment
-
🤖 Based on the current implementation of the Here is the class definition for reference: class AzureChatOpenAI(ChatOpenAI):
"""`Azure OpenAI` Chat Completion API.
...
"""
deployment_name: str = ""
model_version: str = ""
openai_api_type: str = ""
openai_api_base: str = ""
openai_api_version: str = ""
openai_api_key: str = ""
openai_organization: str = ""
openai_proxy: str = ""
... As you can see, the class mainly deals with setting up and interacting with the Azure OpenAI API, and does not contain any methods for setting hyperparameters. However, it's important to note that this doesn't mean it's impossible to set different hyperparameters for different functions. You could potentially extend the Please note that any modifications to the LangChain framework should be done with care, as they could potentially affect other parts of your application. If you're not comfortable making these changes yourself, I would recommend reaching out to a developer with experience in Python and the LangChain framework for assistance. Sources
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research. Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant. If you want to continue the conversation, start your reply with @dosu-bot. |
Beta Was this translation helpful? Give feedback.
-
I would like to set different hyperparameters (e.g. temperature, presence_penalty, frequency_penalty etc) for different functions that I defined for the agent. Is this supported? If so, how can I do that?
Beta Was this translation helpful? Give feedback.
All reactions