Welcome to Neptune!¶
Neptune is a light-weight experiment management tool that helps you keep track of your machine learning experiments.
Use Neptune to log hyperparameters and output metrics from your runs, then visualize and compare results. Automatically transform tracked data into a knowledge repository, then share and discuss your work with colleagues.
Neptune fits in any workflow, ranging from data exploration and analysis, decision science to machine learning and deep learning.
It integrates seamlessly with your machine learning infrastructure, be it AWS, GCP, Kubernetes, Azure, or on-prem machines.
Neptune client is an open source Python library that allows you to integrate your Python scripts with Neptune. Neptune client supports the following cases:
Creating and tracking experiments
Managing running experiment
Querying experiments and projects (search/download)
Track, Organize, Collaborate¶
The Neptune workflow comprises three iterative phases:
Track all objects in the data science or machine learning project. It can be model training curves, visualizations, input data, calculated features and so on. The snippet below presents an example of integration with Python code.
import neptune neptune.init('shared/onboarding', api_token='ANONYMOUS') neptune.create_experiment() neptune.append_tag('minimal-example') n = 117 for i in range(1, n): neptune.send_metric('iteration', i) neptune.send_metric('loss', 1/i**0.5) neptune.set_property('n_iterations', n)
The api_token belongs to the public user Neptuner. After running the code, your experiment will appear here.
For more information, see the Tracking How To Guide.
Organize the structure of your project:
Everything is in one place, accessible from the application or programmatically. Neptune exposes a Query API, that allows users to access their Neptune data right from the Python code.
For more information, see the Organize How To Guide.
Collaborate in the team:
Share your experiments
Comment and communicate your work
Use widgets and mentions to show your progress
Speak your language in our data-science focused interactive wiki!
For more information, see Collaborating in Neptune.
In addition to this documentation set, check out the following resources:
Hands-on tutorial: Covers installation, experiment tracking and comparison, data tracking, and Notebook use.
Sample projects like: a comparison of binary classification metrics applied to fraud detection, research on hyperparameter optimization strategies, or a step-by-step experiment tracking tutorial.
YouTube channel: Provides hands-on videos that showcase key Neptune features.
Blog: Provides in-depth articles about best practices and trends in machine learning.
Neptune user community: Meet other Neptune users and developers and start a discussion.
Presentations, talks, podcasts
Product hunt: A review helps other people find our product.
Technical support: Should you require further support, or have feature requests, reach out at firstname.lastname@example.org or click the chat icon in the bottom right corner of the Neptune UI.