Compare
Neptune vs Azure ML
vs


Show differences only
Neptune

Azure ML
Commercial Requirements
Commercial requirements
Standalone component or a part of a broader ML platform?
Standalone component

Part of the Microsoft Azure ecosystem
Is the product available on-premises and / or in your private/public cloud?

Azure ML is available only as a fully managed cloud service.
Is the product delivered as commercial software, open-source software, or a managed cloud service?
Managed cloud service

Managed Cloud Service
What is the pricing model?
SLOs / SLAs: Does the vendor provide guarantees around service levels?
General Capabilities
Setup
What are the infrastructure requirements?
No special requirements other than having the neptune-client installed and access to the internet if using managed hosting. Check here for infrastructure requirements for on-prem deployment.

Only internet access needed to access the Azure ecosystem
Does it integrate with the training process via CLI/YAML/Client library?
Yes, through the neptune-client library

Yes, through the Azure ML CLI and SDKs
Serverless UI
No

No
Flexibility, speed, and accessibility
How can you access model metadata?

– Julia SDK
No

No
Supported operations

Logging modes

Mobile support
No

No
Experiment Tracking
Log and display of metadata
Dataset

Code versions

Metrics and losses

Rich format

– Neural Network Histograms
No

No
– Prediction visualization (tabular)
No

No
– Prediction visualization (image)
No

No
– Prediction visualization (image – interactive confusion matrix for image classification)
No

– Prediction visualization (image – overlayed prediction masks for image segmentation)
No

– Prediction visualization (image – overlayed prediction bounding boxes for object detection)
No

– TPU
No

No
System information

– Execution command
No

No
Environment config

Files

Comparing experiments
Parameters and metrics

– Parameter Importance plot
No

No
Rich format (side by side)

– Video
No

No
– Audio
No

No
– Plots
No

No
– Interactive visualization (HTML)
No

No
– Neural Network Histograms
No

No
– Prediction visualization (image, video, audio)
No

No
Code

– Git
No

No
– Source files
No

No
Environment

– pip requirements.txt
No

No
– conda env.yml
No

No
– Docker Dockerfile
No

No
Hardware

System information

– Execution command
No

No
Data versions

Files

– Models
No

No
– CSV
No

No
Custom compare dashboards

– Logging custom comparisons from notebooks/code
No

No
Organizing and searching experiments and metadata
Experiment table customization

Experiment filtering and searching

Custom dashboards for a single experiment

– Custom widgets and plugins
No

No
Reproducibility and traceability
One-command experiment re-run
No

No
Experiment lineage

– List of datasets used downstream
No

No
– List of other artifacts (models) used downstream
No

No
– Downstream artifact dependency graph
No

No
Model Registry
Model versioning
Model lineage and evaluation history
Access control, model review, and promoting models
Custom stage tags
No

No
Locking model version and downstream runs, experiments, and artifacts
No

No
Compatibility audit (input/output schema)
No

No
Compliance audit (datasets used, creation process approvals, results/explanations approvals)
No

No
CI/CD/CT compatibility
Model searching
Integrations and Support
Model training
Gluon
No

No
H2O
No

No
Paddle
No

No
Spacy
No

No
Spark MLlib
No

No
Statsmodel
No

No
Hyperparameter Optimization
IDEs and Notebooks
Data versioning
Model versioning
LLMs
This table refers to Azure Machine Learning SDK v1. The SDK v2 does not include any native logging functionality. This table was updated on 10 May 2023. Some information may be outdated. Report outdated information here.
What are the key advantages of Neptune then?

- Focus on the experiment tracking and model registry
- More extensive comparison functionality
- Customizable metadata structure
- Team collaboration features
See these features in action
1
Sign up to Neptune and install client library
pip install neptune
2
Track experiments
import neptune
run = neptune.init_run()
run["params"] = {
"lr": 0.1, "dropout": 0.4
}
run["test_accuracy"] = 0.84
3
Register models
import neptune
model = neptune.init_model()
model["model"] = {
"size_limit": 50.0,
"size_units": "MB",
}
model["model/signature"].upload(
"model_signature.json")
Thousands of ML people already chose their tool
It only takes 5 minutes to integrate Neptune with your code
Don’t overthink it