DSPy: The framework for programming—not prompting—language models
Python
docs(aws.md): added commentary on AWSModel methods
Dhar Rawalfbe22a233efa
diff --git a/docs/api/language_model_clients/aws.md b/docs/api/language_model_clients/aws.mdindex 0b9386f71..4f3c4b0b6 100644--- a/docs/api/language_model_clients/aws.md+++ b/docs/api/language_model_clients/aws.md@@ -30,7 +30,7 @@ lm = dspy.AWSMistral(sagemaker, "<YOUR_MISTRAL_ENDPOINT_NAME>", **kwargs)### Constructor-The constructor initializes the base class `LM` and the `AWSModel` class.+The `AWSMistral` constructor initializes the base class `AWSModel` which itself inherits from the `LM` class.```pythonclass AWSMistral(AWSModel):@@ -47,15 +47,40 @@ class AWSMistral(AWSModel):```**Parameters:**-- `aws_provider` (AWSProvider): The aws provider to use. One of `Bedrock` or `Sagemaker`.-- `model` (_str_): Mistral AI pretrained models. Defaults to `mistral-medium-latest`.+- `aws_provider` (AWSProvider): The aws provider to use. One of `dspy.Bedrock` or `dspy.Sagemaker`.+- `model` (_str_): Mistral AI pretrained models. For Bedrock, this is the Model ID in https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html#model-ids-arns. For Sagemaker, this is the endpoint name.- `max_context_size` (_Optional[int]_, _optional_): Max context size for this model. Defaults to 32768.- `max_new_tokens` (_Optional[int]_, _optional_): Max new tokens possible for this model. Defaults to 1500.- `**kwargs`: Additional language model arguments to pass to the API provider.### Methods-Refer to [`dspy.OpenAI`](https://dspy-docs.vercel.app/api/language_model_clients/OpenAI) documentation.+```python+def _format_prompt(self, raw_prompt: str) -> str:+```+This function formats the prompt for the model. Refer to the model card for the specific formatting required.++<br/>++```python+def _create_body(self, prompt: str, **kwargs) -> tuple[int, dict[str, str | float]]:+```+This function creates the body of the request to the model. It takes the prompt and any additional keyword arguments and returns a tuple of the number of tokens to generate and a dictionary of keys including the prompt used to create the body of the request.++<br/>++```python+def _call_model(self, body: str) -> str:+```+This function calls the model using the provider `call_model()` function and extracts the generated text (completion) from the provider-specific response.++<br/>++The above model-specific methods are called by the `AWSModel::basic_request()` method, which is the main method for querying the model. This method takes the prompt and any additional keyword arguments and calls the `AWSModel::_simple_api_call()` which then delegates to the model-specific `_create_body()` and `_call_model()` methods to create the body of the request, call the model and extract the generated text.+++Refer to [`dspy.OpenAI`](https://dspy-docs.vercel.app/api/language_model_clients/OpenAI) documentation for information on the `LM` base class functionality.+<br/>`AWSAnthropic` and `AWSMeta` work exactly the same as `AWSMistral`.\ No newline at end of filediff --git a/dsp/modules/aws_models.py b/dsp/modules/aws_models.pyindex ec5033297..ec931e156 100644--- a/dsp/modules/aws_models.py+++ b/dsp/modules/aws_models.py@@ -21,13 +21,13 @@ class AWSModel(LM):Instead, use one of the subclasses - AWSMistral, AWSAnthropic, or AWSMeta.The subclasses implement the abstract methods _create_body and _call_model and work in conjunction with the AWSProvider classes Bedrock and Sagemaker.Usage Example:- bedrock = Bedrock(region_name="us-west-2")- bedrock_mixtral = AWSMistral(bedrock, "mistral.mixtral-8x7b-instruct-v0:1", **kwargs)- bedrock_haiku = AWSAnthropic(bedrock, "anthropic.claude-3-haiku-20240307-v1:0", **kwargs)- bedrock_llama2 = AWSMeta(bedrock, "meta.llama2-13b-chat-v1", **kwargs)+ bedrock = dspy.Bedrock(region_name="us-west-2")+ bedrock_mixtral = dspy.AWSMistral(bedrock, "mistral.mixtral-8x7b-instruct-v0:1", **kwargs)+ bedrock_haiku = dspy.AWSAnthropic(bedrock, "anthropic.claude-3-haiku-20240307-v1:0", **kwargs)+ bedrock_llama2 = dspy.AWSMeta(bedrock, "meta.llama2-13b-chat-v1", **kwargs)- sagemaker = Sagemaker(region_name="us-west-2")- sagemaker_model = AWSMistral(sagemaker, "<YOUR_ENDPOINT_NAME>", **kwargs)+ sagemaker = dspy.Sagemaker(region_name="us-west-2")+ sagemaker_model = dspy.AWSMistral(sagemaker, "<YOUR_ENDPOINT_NAME>", **kwargs)"""def __init__(