Overview
Sushi ML Studio - Software Sushi product illustration
Original AI-generated illustration by Software Sushi. Representative product artwork; not an application screenshot or an upstream official logo.
Sushi ML Studio - Software Sushi product illustration
Sushi ML Studio Demo
Start from a spreadsheet or an example
A plain verdict on how good the model is
Predictions for new data
Seven models compared for you
Turn a spreadsheet into predictions you can trust, without writing code, hiring a data scientist, or sending your data to anyone.
Sushi ML Studio is a web application that runs on an EC2 instance in your own AWS account. Open it in your browser, upload a file, and within minutes you have a trained model, an honest account of how good it is, and predictions for new rows.
HOW IT WORKS
- Load your data. Upload a CSV or Excel (.xlsx) file up to 50 MB, or start from one of three example datasets. The Studio reads the file, detects separators and encodings, and profiles every column: its type, blank cells, typical values, and warnings about columns that would mislead a model, such as IDs, dates, free text and duplicates of each other.
- Choose what to predict. Tick the column you want answers about. The Studio works out whether that means sorting rows into groups (classification, for example churned or stayed) or predicting a number (regression, for example a sale price), and you can override it.
- Train. Seven scikit-learn models are trained and compared for you, including gradient boosting, random forest, and linear models. An optional search tries better settings, and then tells you whether it actually helped. You can also adjust a model's settings yourself and retrain just that model. On a large file, a model that may run out of memory or take more than five minutes says so before training, and you can train it on a sample of the rows; a model that runs out of memory stops on its own, and your other models and work are kept.
- Read the results. A plain-language verdict says how good the model is, the predictions and metrics are charted, and a ranking shows which columns drive the prediction.
- Use it. Upload new rows, as CSV or Excel, and download them as a CSV file with a prediction column added. You can switch to any of the trained models right there, and the same file is scored again.
RESULTS YOU CAN TRUST
- Every score is measured on rows the model never saw during training, using cross-validation on smaller files and a held-back test set on larger ones.
- Every score sits next to what a simple guess would have achieved, so a 95% accuracy on data that is 95% one answer is called out, not celebrated.
- When the models have not learned anything useful, the Studio says so.
- When several models perform the same within noise, it says they are tied and recommends the simplest one, instead of crowning a winner by a hair.
- Column importance is measured on held-out rows, and columns that duplicate each other are named, so the ranking does not mislead.
YOU KEEP WHAT YOU MAKE
- Download the trained model as a standard scikit-learn file that loads with joblib, with no dependency on this product. Or download every model you trained in one zip, ranked, with a list of their scores.
- Download the complete Python training code, with the random seeds used, so running it reproduces the same numbers.
- The instance also includes JupyterLab and a Python environment with scikit-learn, pandas, NumPy, SciPy, matplotlib and seaborn, for when you want to continue in code. The Studio runs from its own separate environment, so changing packages cannot break it.
PRIVATE BY DESIGN
- Runs entirely on your instance, in your account and your chosen region. Uploaded data is held in memory and never written to disk or sent anywhere.
- No accounts, no telemetry, no licence server, no calls home.
- The Studio is password protected from the first boot. The initial password is your instance ID, which only people with access to your EC2 console can see, and you can change it from the Studio at any time.
- HTTPS from the first boot: the Studio is served at https:// with a trusted certificate for your instance's address, renewed automatically. No domain and nothing to set up.
- Prefer no open ports? Turn HTTPS off and reach the Studio from your own IP only, over an SSH tunnel, or through AWS Systems Manager Session Manager with no inbound ports at all.
GOOD TO KNOW
- Works with tabular data: one row per example, one column per attribute. It does not do time-series forecasting, images, or text analysis.
- Download your model before you close the browser tab. The Studio does not keep projects between sessions, and warns you before anything unsaved would be lost.
- Recommended instance type: t3.medium or larger.
Built by Software Sushi, a team that designs custom machine learning and AI systems. Support is by email on business days.
Highlights
- No-code machine learning in minutes: upload a CSV or Excel file, pick the column to predict, and seven scikit-learn models are trained and compared for you, with predictions for new rows ready to download.
- Honest results: every score is measured on rows the model never saw, compared against a simple guess, and explained in plain language, including when a model has not learned anything useful.
- Private and portable: it runs on your own EC2 instance, your data never leaves it, and you download the model file and the Python code that reproduces it. No accounts, no lock-in.
Details
Features and programs
Financing for AWS Marketplace purchases
Pricing
Free trial
Dimension | Cost/hour |
|---|---|
t3.medium Recommended | $0.05 |
c7i.2xlarge | $0.05 |
c7i.xlarge | $0.05 |
m7i.2xlarge | $0.05 |
m7i.xlarge | $0.05 |
m7i.large | $0.05 |
t3.xlarge | $0.05 |
t3.2xlarge | $0.05 |
t3.large | $0.05 |
Dimensions summary
Top-of-mind questions for buyers
Vendor refund policy
Sushi ML Studio has a 7-day free trial, so you can try it on your own data before you pay anything. During the trial there is no software charge; AWS still bills the EC2 instance as usual.
After the trial, the software is billed by AWS per instance hour, only while the instance is running. Stopping or terminating the instance stops the charge. The trial moves to paid use automatically, so stop or terminate your instance before the trial ends if you do not want to continue.
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
64-bit (x86) Amazon Machine Image (AMI)
Amazon Machine Image (AMI)
An AMI is a virtual image that provides the information required to launch an instance. Amazon EC2 (Elastic Compute Cloud) instances are virtual servers on which you can run your applications and workloads, offering varying combinations of CPU, memory, storage, and networking resources. You can launch as many instances from as many different AMIs as you need.
Version release notes
Sushi ML Studio 1.2.0.
- New: HTTPS by default. Open https:// and the Studio is served with a trusted certificate for your instance's address, renewed automatically, with nothing to set up. Ports 80 and 443 must be open in the security group. http://:8501 still works.
- New: sudo sushi-ml-studio https status, on and off, to check or change it.
- New: on large files, a model that would probably run out of memory or take more than five minutes says so before training, and can be trained on a sample of the rows you choose. A model that ran out of memory can be retrained on a sample in one click.
- New: choose which model to use right on Predict & export, without going back to Results. Switching scores your uploaded file again with that model.
- Changed: the column contribution check on the Data page explains what its numbers mean, and no longer offers to drop columns, which could cost accuracy.
- Faster: on large files, nearest neighbours finishes up to about 6 times sooner and random forest about a third sooner, with the same models and scores.
- Fixed: random forest and extra trees failed when the column to predict held numbers stored as text, as Excel often does, and some groups were much smaller than others.
- Fixed: on a large file, a model that needed more memory than the instance had could restart the Studio and lose your work. Now only that model stops, with a message saying so, and your session and other models are kept.
- Operating system and Python packages updated to the latest security patches at build time.
Additional details
Usage instructions
- Launch with the security group created from seller settings. It opens 443 and 80 for HTTPS, and 22 for SSH. Keep 80 and 443 open to every address: Let's Encrypt checks the certificate from the internet.
- Wait 1 to 5 minutes, then open https://<public-ip> in your browser.
- Sign in with your instance ID (it starts with i-, shown in the EC2 console), then choose your own password: Change password in the sidebar, or run: sudo sushi-ml-studio set-password
Optional, if you would rather not expose the Studio to the internet: run sudo sushi-ml-studio https off, remove the rules you do not need, and use one of these. Each opens the Studio at http://localhost:8501 unless noted.
- Your IP only: allow 8501 from your own IP, then open http://<public-ip>:8501 (not encrypted, password protected).
- SSH tunnel, only 22 open: ssh -i <key>.pem -L 8501:localhost:8501 ubuntu@<public-ip>
- Session Manager, no inbound ports (needs an instance role with AmazonSSMManagedInstanceCore): aws ssm start-session --target <instance-id> --document-name AWS-StartPortForwardingSession --parameters '{"portNumber":["8501"],"localPortNumber":["8501"]}'
- Inside your VPC: allow 8501 from your VPC range only.
Check HTTPS any time: sushi-ml-studio https status Setup guide: https://docs.softwaresushi.com/ml-studio/ Launch kit (Session Manager role, CloudFormation): https://docs.softwaresushi.com/launch-kit/ Help: info@softwaresushi.com
Support
Vendor support
Email support from Software Sushi, the publisher.
- Support email: info@softwaresushi.com
- Hours: weekdays. We reply within one or two business days.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.