neptune.ml Reviews

What our company appreciate the most about Neptune is the ability to log a rich variety of data from training runs and organize all the most essential information in a custom dashboard. Everything from use-case specific performance plots generated with Plotly to Neptune native timeseries and scalars. On top of that, Neptune requires minimal implementation overhead, has high quality documentation, great support, and it's just reliable.

Could have benefited from having some more native plotting modules. However, it is possible to upload Plotly plots or similar as a workaround.

We update our ML production model frequently. Neptune helps us keep track of how the model performance evolves over time.

The most valuable thing for us is the ability to upload a wide range of data types, such as summary plots at the end of a run or generated samples throughout model training. It allows us to use Neptune to present and share results with colleagues, all in one place, without having to manage files locally. Before Neptune, we used to log all the experimental results to a file located on our server. It was difficult to visualize, share and discuss results. We also struggled to keep track of different versions of experiments. So, we would absolutely recommend Neptune.

The integrations with other tools like PyTorch Lightning are occassionally buggy, but this is a minor issue.

Neptune allows us to quickly and easily share results from experiments across our team, without the need to constantly save, share, and track individual files.

- Every experiment on a single platform

- The best part is the integrations with other libraries

- There is no such dislike at this moment and really happy to use this tool

Tracking about the work

They respect to feedback and suggestion and update regularly the new features for a better experience.

At the moment everything is pretty useful

Everything is available to use. If you need any new features, you can simply ping them. They will consider your suggestion.

tracking my ML experiment. Hyperparameter tuning. Sharing the result with my colleagues and other tones of benefits.

one place to log all my experiments, very helpful when you have to find some results from a few months back.

It makes collaboration easier as well - just share the link to an experiment with a colleague and you can analyze the results together.

The UI for creating graphs with multiple lines could be more flexible.

* tracking data about many experiments

* easily sharing experiments within lab

* going back to old results, collecting data for report/publication

* reproducibility - code and hyperparameters are stored in one place

The real-time charts, the simple API, the responsive support

It would be great to have more advanced API functionality.

If you need to monitor and manage your machine learning or any other computational experiments, Neptune.ai is a great choice. It has many features that can make your life easier and your research more organized.

I'm mostly doing an academic research that involves the training of machine learning models, and also other long-running experiments which I need to track in real time. Without Neptune.ai, I would have waste a lot of time building a client for experiment management and monitoring. It also serves as an archive, which I also find very important for my research.

It's an easy yet powerful tool, handling notebooks is a nice feature.

Front-end details, like axis labels, could be more aesthetic.

I'm using this tool as an experiment manager, mainly as a way to choose the best model setup and store respective images.

It's an easy yet powerful tool, handling notebooks is a nice feature.

Front-end details, like axis labels, could be more aesthetic.

What problems are you solving with the product? What benefits have you realized?

We highly value Neptune's PyTorch Lightning plugin because it was the key to quick integration with our workflow and the experiment comparison view because it's critical for our ongoing hyperparameter optimization. Other than that, there are things you often take for granted in a product: reliability, flexibility, quality of support. Neptune nails those and gives us the confidence.

We miss a few minor UI features. Besides, we'd love to be notified of running out of logging minutes in advance by predicting the usage.

We monitor the training of LLMs and log our hyperparameter selection experiments.

It is an excellent tool to keep all our experiments on a single platform and easy to track the progress.

Some more features can be added like visualizations, APIs etc.

It was used to work on a Data Science project with a team of 4 people.