Welocalize Onboarding Guide

Table of Contents

What is Weights and Biases (W&B)? 💫

Quick Documentation Links

Access the main documentation page: https://docs.wandb.ai/
Learn how to integrate W&B with your stack: https://docs.wandb.ai/guides/integrations
Track your experiment metrics: https://docs.wandb.ai/guides/track
Track and version datasets, models, and other artifacts: https://docs.wandb.ai/guides/artifacts (Code)
Visualize your data with intearactive tables: https://docs.wandb.ai/guides/data-vis/tables-quickstart (Code)
Tune hyperparameters via sweeps: https://docs.wandb.ai/guides/sweeps (Code)
Use reports to share your findings: https://docs.wandb.ai/guides/reports

W&B Installation & Authentication

pip install wandb

wandb login <YOUR API TOKEN>

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

Track any python process or experiment with W&B's Experiment Tracking 🍽

import wandb

wandb.init(project="my-sample-project",   
           entity="<enter team name>",	       # Team
           group='my_group',	               # for organizing runs (e.g. distributed training)
           job_type='training',		       # for organizing runs (e.g. preprocessing vs. training)
           config={'hyperparam1': 24,          # Hyperparams and other config
		   'hyperparam2': 'resnet'})

What Can I log and How do I log it?

Scalar Metrics

wandb.log({"my_metric": some_scalar_value})

wandb.init()

for i in range(5):
    wandb.log({"my_metric": i})

wandb.summary["my_metric"] = 2 # 2 instead of the default 4

wandb.finish()

Distributed Training

One process: Initialize W&B (wandb.init) and log experiments (wandb.log) from a single process. This is a common solution for logging distributed training experiments with the PyTorch Distributed Data Parallel (DDP) Class. In some cases, users funnel data over from other processes using a multiprocessing queue (or another communication primitive) to the main logging process.
Many processes: Initialize W&B (wandb.init) and log experiments (wandb.log) in every process. Each process is effectively a separate experiment. Use the group parameter when you initialize W&B (wandb.init(group='group-name')) to define a shared experiment and group the logged values together in the W&B App UI.

Visualize and query dataframes via W&B Tables

# Add data as a list of lists or pandas dataframe
my_data = [
  [0, wandb.Image("img_0.jpg"), 0, 0],
  [1, wandb.Image("img_1.jpg"), 8, 0],
  [2, wandb.Image("img_2.jpg"), 7, 1],
  [3, wandb.Image("img_3.jpg"), 1, 1]
]
          
# create a wandb.Table() with corresponding columns
columns=["id", "image", "prediction", "truth"]
test_table = wandb.Table(data=my_data, columns=columns)

# Add data incrementally
for img_id, img in enumerate(mnist_test_data):
    true_label = mnist_test_data_labels[img_id]
    guess_label = my_model.predict(img)
    test_table.add_data(img_id, wandb.Image(img), \
                         guess_label, true_label)

wandb.log({"test_table": test_table})

Track and version any serialized data via Artifact Tracking and Versioning

Logging Artifacts

local files
local directories
wandb Data Types (e.g. wandb.Plotly or wandb.Tables) which will render alongside the artifact in the UI
remote files and directories (e.g. s3 buckets)

# 1. Log a dataset version as an artifact
import wandb
import os

# Initialize a new W&B run to track this job
run = wandb.init(project="artifacts-quickstart", job_type="dataset-creation")

# Create a sample dataset to log as an artifact
f = open('my-dataset.txt', 'w')
f.write('Imagine this is a big dataset.')
f.close()

# Create a new artifact, which is a sample dataset
dataset = wandb.Artifact('my-dataset', type='dataset')
# Add files to the artifact, in this case a simple text file
dataset.add_file('my-dataset.txt')
# Log the artifact to save it as an output of this run
run.log_artifact(dataset)

wandb.finish()

Consuming Artifacts

import wandb
run = wandb.init()
# Indicate we are using a dependency
artifact = run.use_artifact('dummy-team/that_was_easy/my-dataset:v3', type='dataset')
artifact_dir = artifact.download()

Tracking Artifacts By Reference

House staged/candidate models via W&B's Model Registry

Bookmark your best model versions for each machine learning task.
Automate downstream processes and model CI/CD
Move model versions through its ML lifecycle; from staging to production.
Track a model's lineage and audit the history of changes to production models.

Tune Hyperparameters via W&B Sweeps

Add wandb.init() to your training script, ensuring that all hyper-parameters are passed to your training logic via wandb.config.
Write a yaml file with your hyper-parameter search specified i.e. method of search, hyper-parameter distributions and values to search over.
Run the sweep controller, which runs in W&B through wandb.sweep or through the UI. The controller will delegate new hyperparameter values to wandb.config of the various agents running.
Run agents in however many machines you want to run the experiments with wandb.agent

Organize visualizations and share your findings with collaborators via W&B Reports

Programmatic Reports

Track and evaluate LLM applications via W&B Weave

Log and debug language model inputs, outputs, and traces
Build rigorous, apples-to-apples evaluations for language model use cases
Organize all the information generated across the LLM workflow, from experimentation to evaluations to production

Other Useful Resources

Import/Export API

Slack Alerts

FAQ

wandb.init(name="my_awesome_run")

import wandb

# Ensure you are logged in
wandb.login()

# Get the current user
user = wandb.api.viewer()

# Access the teams the user is part of
teams = user['teams']
print("Teams:", teams)

import wandb

api = wandb.Api()
projects=api.projects(entity='your-entity')
for p in projects:
    print(p.name)

os.environ["WANDB_SILENT"] = "true"

%env WANDB_SILENT=true

WANDB_SILENT=true