Quick Start
import dspy
lm = dspy.LM('ollama_chat/llama3.2:1b', api_base='http://localhost:11434')
dspy.configure(lm=lm)
This snippet initializes a language model and configures DSPy for use.
Defining a Signature
from typing import Literal
class Categorize(dspy.Signature):
event: str = dspy.InputField()
category: Literal['Wars and Conflicts', 'Politics'] = dspy.OutputField()
confidence: float = dspy.OutputField()
Signatures define input-output structures, making your models more intuitive.
Calling the Module
classify = dspy.Predict(Categorize)
classification = classify(event="[YOUR HISTORIC EVENT]")
print(classification)
Use the Predict module to classify events with ease.
Optimizing Prompts
from dspy.teleprompt import *
tp = dspy.MIPROv2(metric=validate_category, auto="light")
optimized_classify = tp.compile(classify, trainset=trainset)
Optimize prompts with DSPy’s Teleprompt module for better performance.
Saving Optimized Systems
optimized_classify.save("optimized_event_classifier.json")
Save your optimized classification systems for later use or deployment.