Quickstart Guide

Welcome to PromptOptimizer! This guide will help you quickly get started with using PromptOptimizer in your projects. PromptOptimizer is a Python library that allows you to minimize token complexity in order to save API costs and reduce model computations.

Installation

Quick Installation

To quickly install PromptOptimizer, use the following command:

pip install prompt-optimizer

Install from Source

If you prefer to install PromptOptimizer from source, follow these steps:

1. Clone the repository:

git clone https://github.com/vaibkumr/prompt-optimizer.git

2. Navigate to the cloned repository:

cd prompt-optimizer

3. Install PromptOptimizer using pip:

pip install -e .

Prompt Optimizers

A prompt optimizer is a callable class that outputs optimized prompt data along with metrics (if requested) for given input prompt data.

Note: Optimizers output a result object with keys content to store optimized prompts and metrics to store the requested metrics computations results.

To optimize a prompt we follow three steps:

1. Import the optimizer from the range of available ../optimizers/index.html. For now, we use the EntropyOptim.

from prompt_optimizer.poptim import EntropyOptim

2. Initialize the optimizer object. Each optimizer has its own argument which can be tuned to achieve a balance between the cost and performance tradeoff.

p_optimizer = EntropyOptim(p=0.1)

3. Run the optimizer over a given prompt string and fetch the results

prompt = "In Nightmare of Mensis progress through until you reach the boss room."
result = p_optimizer(prompt)
optimized_prompt = result.content

And we’re done! We just optimized our first prompt, saved some money and if we’re smart, we had no loss in model performance.

Input Formats

Prompt optimizers support three different formats:

1. String: A basic python string. At the core, all optimizers work on python strings.

from prompt_optimizer.poptim import EntropyOptim
p_optimizer = EntropyOptim(p=0.1)
prompt = "In Nightmare of Mensis progress through until you reach the boss room."
result = p_optimizer(prompt)
optimized_prompt = result.content

2. JSON Object: APIs often accept instructions in form of sytem and human messages. JSON objects of the following format can be passed to the optimizers using the json boolean flag:

[
    {
        "role":"system",
        "content":"System instructions..."
    },
    {
        "role":"user",
        "content":"User prompt..."
    }
]

often times, it is important to skip system instructions. It can be done using the skip_system flag as follows:

from prompt_optimizer.poptim import EntropyOptim
p_optimizer = EntropyOptim(p=0.1)
prompt = [
    {
        "role":"system",
        "content":"System instructions..."
    },
    {
        "role":"user",
        "content":"User prompt..."
    }
]
optimized_prompt = p_optimizer(prompt, json=True, skip_system=True)

3. Langchain Object: Langchain agents accept prompts as a list of SystemMessage and HumanMessage. Prompt optimizers can directly be applied to these objects by using the langchain boolean flag. Again, skip_system flag can be used to skip optimizing system prompts as follows:

from prompt_optimizer.poptim import EntropyOptim
from langchain.schema import (
    HumanMessage,
    SystemMessage
)

p_optimizer = EntropyOptim(p=0.1)
prompt = [
    SystemMessage(content="You are a helpful assistant that translates English to French."),
    HumanMessage(content="I love programming.")
]
optimized_prompt = p_optimizer(prompt, langchain=True, skip_system=True)