Listing Thumbnail

    John Snow Labs - Generative AI Lab

     Info
    Generative AI Lab is the most highly used No-Code human-in-the-loop tool for AI teams. It offers End-to-End data labeling and DL model training features.
    Listing Thumbnail

    John Snow Labs - Generative AI Lab

     Info

    Overview

    Play video

    Generative AI Lab (previously known as NLP Lab and Annotation Lab) is an End-to-End No-Code platform for annotating text and training AI/ML models. It enables domain experts to extract meaningful facts from text documents, images or PDFs and train models that will automatically predict those facts on new documents. By offering out-of-the-box support for Large Language Model prompting, Zero-Shot prompting, Rules and state-of-the-art John Snow Labs pre-trained models, Generative AI Lab helps domain experts efficiently prepare training data for tuning custom AI models for specific tasks and use-cases.

    About the offer: Based on an auto-scaling architecture powered by Kubernetes, Generative AI Lab can scale to many teams and projects. Enterprise-grade security is included, with support for air-gap environments, zero data sharing, role-based access, full audit trails, MFA, and identity provider integrations. Generative AI Lab allows powerful experiments for model training and finetuning, model testing, and model deployment as API endpoints. There is no limitation on the number of users, projects, tasks, models, or trainings that can be run with this subscription. This product includes a Pay-As-You-Go license key for John Snow Labs libraries and models, that offers access to 40.000+ models and pipelines for healthcare, legal, finance, downloadable from the NLP Models Hub and with access to OCR and Visual Document understanding features. Designed to take advantage of GPU architecture, the product offers a boost in performance for model training and preannotation tasks - https://nlp.johnsnowlabs.com/docs/en/CPUvsGPUbenchmark_healthcare  You will be charged ONLY as long as you use the product. Simply stop your instance and restart it when needed it so you get charged only based on what you consume.

    Included Features:

    • Prompt engineering for Large Language Models and Zero-Shot Models - entity recognition, relation extraction, classification.
    • AI-Assisted Annotation: never start from scratch but reuse existing models to pre-annotate tasks with the latest John Snow Labs models for classification, NER, assertion status, entity resolution and relation detection;
    • High productivity annotation UI with keyboard shortcuts and pre-annotations;
    • Annotation support for Text, Image, Audio, Video and HTML;
    • Support for text annotation in 250+ languages;
    • Support for projects and teams: 30+ project templates; unlimited projects and users, project import, export and cloning, project grouping;
    • Task assignment, tagging, and comments; duplicate tasks identification; task searching and filtering;
    • Consensus analysis and Inter Annotator Agreement charts;
    • Enterprise-level security and privacy: role-based access control, role-based views, annotation versioning, full audit trail, Single Sign on;
    • Full NLP Models Hub integration: you can explore available models and embeddings, download them with the click of a button and reuse those in your project configuration.
    • Train Classification, NER, and Assertion Status models: use default parameters or easily tune them on the UI for different experiments;
    • Active Learning automatically trains new versions of your models once new annotations are available;
    • Playground - deploy, test, and update prompts, rules and models before you decide to include them in your project;
    • API access to all features for easy integration into custom data analysis pipelines;

    Who is this offer for

    • Domain experts (e.g. nurses, doctors, lawyers, accountants, investors, etc.) who want to test DL models on their data or/and tune/train new models via an easy-to-use UI, without writing a line of code;
    • Data labeling teams who want to optimize the efficiency and speed of their day-to-day work with preannotations;
    • Machine Learning engineers who need to test/train/tune NLP models;
    • Researchers who need to extract meaning from unstructured, natural language documents;
    • And anyone else interested in text and image analysis, image digitization, data extraction, document labeling and/or NLP model training.

    Target verticals The Generative AI Lab is a domain-agnostic tool that can be used to annotate documents in any vertical. Its integration with the NLP Models Hub facilitates access to over 40k pre-trained models for general-purpose text documents. The tool allows easy access and reuse of 2000+ pre-trained models covering 400+ clinical and biomedical entity types.

    Technical Specifications Operating System:Ubuntu 20.04

    3 Easy Steps to get started Subscribe to the product on the AWS Marketplace. Deploy it on a new machine. Access the login page for a guided experience on http://INSTANCE_IP. For the first login use the following credentials: Username: admin Password: INSTANCE_ID

    Highlights

    • Includes everything: - Model Hub Integration - Project Management - Role Based Access - Workflows - Analytics - Model Training and Testing - Preannotations - Security and Privacy Unlimited everything: - Users - Projects - Models - Tasks - Annotations - Pre-annotations - Training
    • Healthcare Resources - Access to 2000+ Healthcare pre-trained models covering Clinical and Biomedical NER for 400+ entity types; Assertion Status detection (positive, negative, possible, past and future facts), Clinical Relation Extraction; - De-identification NER Models - Model tuning - Build your models on existing pre-trained models - Programmatic labeling via dictionary and regex-based rules;
    • Visual Document Understanding - Pre-annotate PDF and image tasks with Visual NER models; - Tune Visual NER models for your data; - Sticky and custom annotations; - Automatic text recognition; - Support for relation annotation on top of images; - Text-based search on the image/PDF; - Zoom features;

    Details

    Delivery method

    Delivery option
    64-bit (x86) Amazon Machine Image (AMI)

    Latest version

    Operating system
    Ubuntu 22.04

    Pricing

    John Snow Labs - Generative AI Lab

     Info
    Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time.
    Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator  to estimate your infrastructure costs.

    Usage costs (3)

     Info
    Dimension
    Cost/unit
    No-Code Generative AI Lab Instance, per processor per hour
    $0.07
    Medical Model Usage for Annotation or Training, per processor per min
    $0.099
    Visual Document Import, Annotation, or Training, per processor per min
    $0.099

    Additional AWS infrastructure costs

    Type
    Cost
    EBS General Purpose SSD (gp2) volumes
    $0.10/per GB/month of provisioned storage

    Vendor refund policy

    Users need to pay price to Amazon according to the EC2 instances/servers used.

    Legal

    Vendor terms and conditions

    Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .

    Content disclaimer

    Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.

    Usage information

     Info

    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

    The release of Generative AI Lab version 6.7 introduces significant enhancements aimed at improving user control and overall project efficiency.

    A major update is the expanded de-identification feature, which now allows users to select de-identification methods on a per-entity basis. This update enables the integration of custom models, rules, and prompts, making the process of anonymizing sensitive data more flexible and tailored to specific needs.

    Additionally, admin users can now add detailed annotation instructions directly from the Customize Labels page, ensuring that annotators have clear guidelines, which enhances labeling accuracy and consistency across projects.

    Moreover, several user experience improvements have been implemented, such as automatic exit from wizard mode post-training, direct navigation from relation lines, and a new Quick Submit button for Visual NER tasks, facilitating quicker submissions.

    For additional information visit https://nlp.johnsnowlabs.com/docs/en/alab/annotation_labs_releases/release_notes_6_7_0 

    Additional details

    Usage instructions

    Ensure the IAM role attached to the AMI machine has access to both aws-marketplace:MeterUsage and ec2:DescribeInstanceTypes permission.

    Launch the AMI Generative AI Lab will then be served on http://<public ip of instance>

    To login use the following credentials

    • username: admin
    • password: <instance-id from AWS EC2>

    Support

    Vendor support

    Technical support for Generative AI Lab by Development Team support@johnsnowlabs.com 

    John Snow Labs also offers professional services to deliver custom data science work that is specific to your needs. Our team of experts is ready to assist you with various tasks, including training custom AI models, developing machine learning pipelines, annotating documents, creating Python notebooks, generating insightful reports, and much more. Our professional services are specifically designed to help you achieve remarkable results without the steep learning curve or overwhelming workload.

    In addition, when you opt for an annual NLP Libraries prepaid subscription you gain access to a host of exclusive benefits:

    A dedicated customer success manager A dedicated account manager Four hours of personalized onboarding from our data scientists Year-long customer support on a dedicated Slack channel

    Additional Resources:

    AWS Marketplace Slack Channel: https://spark-nlp.slack.com/archives/C064YR9NLBX 

    End-to-End No-Code Development of NER model for Text with Generative AI Lab: https://www.youtube.com/watch?v=jgUylZlz3uA&ab_channel=JohnSnowLabs 

    Generative AI Lab Release Notes: https://nlp.johnsnowlabs.com/docs/en/alab/release_notes  support@johnsnowlabs.com 

    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.

    Similar products

    Customer reviews

    Ratings and reviews

     Info
    0 ratings
    5 star
    4 star
    3 star
    2 star
    1 star
    0%
    0%
    0%
    0%
    0%
    0 AWS reviews
    |
    2 external reviews
    External reviews are sourced from G2  and are not included in the star rating for this product.
    Ashpreet S.

    Must try Annotating Tool

    Reviewed on Nov 25, 2023
    Review provided by G2
    What do you like best about the product?
    It is very user friendly and easy to understand in camparison to other competetive products.
    What do you dislike about the product?
    As my use case I did like all the features.
    What problems is the product solving and how is that benefiting you?
    I am using this as an Annotating Tool for image datasets
    Eric L.

    Great annotation tool

    Reviewed on Oct 06, 2022
    Review provided by G2
    What do you like best about the product?
    It makes the annotation process very simple and efficient. Easy to use. Easy to manage the work.
    What do you dislike about the product?
    The review process is a little buggy and non-intuitive. That part of the workflow should be improved.
    What problems is the product solving and how is that benefiting you?
    It solves the problem of establishing ground truth when training or testing a model.
    View all reviews