NBC Universal Weave Onboarding Guide

os.environ["WANDB_BASE_URL"] = "https://mural.wandb.io/"
os.environ["WANDB_API_KEY"]  = "your-wandb-api-key"

pip install weave

wandb login --host <YOUR W&B HOST URL> <YOUR API TOKEN>

wandb.login(host=os.getenv("WANDB_BASE_URL"), key=os.getenv("WANDB_API_KEY"))

import weave
weave.init('<entity-name>/<project-name>')
# this ensures the project is created in the relevant team. 
# If you have WANDB_ENTITY and WANDB_PROJECT set as env vars then you won't need to specify these.
# You may need to create the team. 

import weave
from openai import OpenAI

client = OpenAI()

# Weave will track the inputs, outputs and code of this function
@weave.op()
def extract_dinos(sentence: str) -> dict:
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[
            {
                "role": "system",
                "content": """In JSON format extract a list of `dinosaurs`, with their `name`,
their `common_name`, and whether its `diet` is a herbivore or carnivore"""
            },
            {
                "role": "user",
                "content": sentence
            }
            ],
            response_format={ "type": "json_object" }
        )
    return response.choices[0].message.content


# Initialize the weave project
weave.init('<team-name>/jurassic-park')

sentence = """I watched as a Tyrannosaurus rex (T. rex) chased after a Triceratops (Trike), \
both carnivore and herbivore locked in an ancient dance. Meanwhile, a gentle giant \
Brachiosaurus (Brachi) calmly munched on treetops, blissfully unaware of the chaos below."""

result = extract_dinos(sentence)
print(result)

from openai import OpenAI

import weave

weave.init(PROJECT)

client = OpenAI()
response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[
        {
            "role": "system",
            "content": "You are a grammar checker, correct the following user input.",
        },
        {"role": "user", "content": "That was so easy, it was a piece of pie!"},
    ],
    temperature=0,
)
generation = response.choices[0].message.content
print(generation)

import weave
from weave import Evaluation

# Define any custom scoring function
@weave.op()
def exact_match(expected: str, output: dict) -> dict:
    # Here is where you'd define the logic to score the model output
    return {"match": expected == output}


# Score your examples using scoring functions
evaluation = Evaluation(
    dataset=dataset,  # can be a list of dictionaries or a weave.Dataset object
    scorers=[exact_match],  # can be a list of scoring functions
)

# Start tracking the evaluation
weave.init(PROJECT)
# Run the evaluation
print(asyncio.run(evaluation.evaluate(corrector)))

# if you're in a Jupyter Notebook, run:
# await evaluation.evaluate(corrector)