Giters

KanapaZombie / examples

📝 Examples of experiment tracking, model registry, data versioning, and monitoring machine learning model training live in neptune.ai

Home Page:https://neptune.ai

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

neptune.ai examples

What is Neptune?

Neptune is a metadata store for MLOps, built for teams that run a lot of experiments.

It's used for:

  • Experiment tracking: Log, display, organize, and compare ML experiments in a single place.
  • Model registry: Version, store, manage, and query trained models, and model building metadata.
  • Monitoring ML runs live: Record and monitor model training, evaluation, or production runs live.

Examples

In this repo, you'll find examples of using Neptune to log and retrieve your ML metadata.

You can run every example with zero setup as an "ANONYMOUS" Neptune user (no registration needed).

Note: This readme is best viewed in the GitHub Light theme.

Getting started

Docs Neptune GitHub Colab
Quickstart docs neptune github colab

Use cases

Docs Neptune GitHub Colab
Time Series Forecasting neptune github

How-to guides

Experiment Tracking

Docs Neptune GitHub Colab
Track and organize model-training runs docs neptune github colab
DDP training experiments docs neptune github
Re-run failed training docs neptune github colab
Use Neptune in HPO training job docs neptune github colab
Logging from sequential ML pipelins neptune github

Model Registry

Docs Neptune GitHub Colab
Log model building metadata docs

Monitoring ML Runs Live

Docs Neptune GitHub Colab
Monitor model training runs live docs neptune github colab

Data Versioning

Docs Neptune GitHub Colab
Version datasets in model training runs docs neptune github colab
Compare datasets between runs docs neptune github colab

Neptune API

Docs Neptune GitHub Colab
Resume run or other object docs
Pass run object between files docs
Use Neptune in distributed computing docs
Use Neptune in parallel computing docs
Use Neptune in Pipelines docs
Log to multiple runs in one script docs
Create and delete projects docs github colab

Neptune app

Docs Neptune GitHub Colab
Filter and sort runs table docs
Do GroupBy on runs docs

Integrations and Supported Tools

Languages

Docs Neptune GitHub Colab
Python docs neptune github colab
R docs

Model Training

Docs Neptune GitHub Colab
Catalyst docs neptune github colab
fastai docs neptune github colab
Keras docs neptune github colab
lightGBM docs neptune github colab
Prophet docs neptune github colab
PyTorch docs neptune github colab
PyTorch Ignite docs
PyTorch Lightning docs neptune github colab
scikit-learn docs neptune github colab
skorch docs neptune github colab
🤗 Transformers docs neptune github colab
TensorFlow docs neptune github colab
XGBoost docs neptune github colab

Hyperparameter Optimization

Docs Neptune GitHub Colab
Keras Tuner docs
Optuna docs neptune github colab
Scikit-Optimize docs

Model Visualization and Debugging

Docs Neptune GitHub Colab
Altair docs neptune github colab
Bokeh docs neptune github colab
Dalex docs
HiPlot docs
HTML docs neptune github colab
Matplotlib docs neptune github colab
Pandas docs
Plotly docs neptune github colab

Automation Pipelines

Docs Neptune GitHub Colab
Kedro docs neptune github

Experiment Tracking

Docs Neptune GitHub Colab
MLflow docs
Sacred docs neptune github colab
TensorBoard docs

IDEs and Notebooks

Docs Neptune GitHub Colab
Amazon SageMaker notebooks docs
Deepnote docs
Google Colab docs neptune github colab
Jupyter Notebook and JupyterLab docs

Continuous Integration and Delivery (CI/CD)

Docs Neptune GitHub Colab
Docker docs
GitHub Actions docs

Amazon SageMaker

Docs Neptune GitHub
Setting up Neptune credentials in AWS Secrets docs
Using Neptune in training jobs with custom Docker containers neptune github
Using Neptune in training jobs with PyTorch Estimator neptune github

About

📝 Examples of experiment tracking, model registry, data versioning, and monitoring machine learning model training live in neptune.ai

https://neptune.ai

License:MIT License

Languages

Language:Jupyter Notebook 90.1%Language:Python 9.4%Language:Shell 0.4%Language:Dockerfile 0.0%Language:HTML 0.0%