Sold by: Instituto Ingenieria Conocimiento
Deployed on AWS
RigoBERTa Clinical Classification is a state-of-the-art encoder for Spanish clinical texts, built to help you fine-tune language models on your own classification datasets.
Designed to support the development of highly accurate models, it allows you to adapt the encoder to your specific use case and labeling schema using your own clinical data.
Structured clinical text classification tasks benefit more from a specialized Natural Language Understanding (NLU) approach than from Natural Language Generation (NLG) methods.
Once fine-tuned, models can be seamlessly deployed for real-time or batch inference, enabling direct integration into clinical workflows and facilitating faster, more reliable decision-making
Overview
Easily fine-tune RigoBERTa Clinical on your own clinical data to develop highly accurate text classification models specifically tailored to your unique use case and labeling requirements.
This targeted Natural Language Understanding (NLU) approach delivers superior performance compared to traditional Natural Language Generation (NLG) methods when applied to structured classification tasks.
Once fine-tuned, your customized model is ready for seamless deployment and real-time inference, allowing for direct integration into clinical workflows and supporting faster, more reliable decision-making across healthcare applications
RigoBERTa Clinical was built by further pretraining our general-purpose RigoBERTa 2 model on a meticulously Spanish curated clinical corpus, significantly improving performance on multiple clinical NLP benchmarks while offering robust language understanding in the clinical domain.
To fine-tune RigoBERTa Clinical with your data, prepare a labeled dataset in JSON or CSV format and upload it to Amazon S3. Configure training parameters such as number of epochs, batch size, and learning rate, then launch a training job using Amazon SageMaker. After training is complete, deploy the model to a real-time endpoint to perform on-demand inference, or run batch prediction jobs for large-scale processing of clinical texts.
An open-weight version of this model, intended solely for research and non-commercial use, is available on the public Hugging Face profile of IIC
Highlights
- RigoBERTa Clinical is clinical encoder language model developed through domain-adaptive MLM pretraining on the largest publicly available Spanish clinical corpus, ClinText-SP
- We recommend using this model as a foundation for clinical NLP applications by fine-tuning it on your own data for medical text classification tasks, such as clinical note labeling
Details
Delivery method
Latest version
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
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.
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.g5.8xlarge Training Recommended | Algorithm training on the ml.g5.8xlarge instance type | $1.00 |
ml.g5.8xlarge Inference (Batch) Recommended | Model inference on the ml.g5.8xlarge instance type, batch mode | $2.70 |
ml.g5.8xlarge Inference (Real-Time) Recommended | Model inference on the ml.g5.8xlarge instance type, real-time mode | $2.70 |
ml.g5.12xlarge Training | Algorithm training on the ml.g5.12xlarge instance type | $1.00 |
ml.g5.12xlarge Inference (Batch) | Model inference on the ml.g5.12xlarge instance type, batch mode | $2.70 |
ml.g5.12xlarge Inference (Real-Time) | Model inference on the ml.g5.12xlarge instance type, real-time mode | $2.70 |
ml.g5.16xlarge Training | Algorithm training on the ml.g5.16xlarge instance type | $1.00 |
ml.g5.16xlarge Inference (Batch) | Model inference on the ml.g5.16xlarge instance type, batch mode | $2.70 |
ml.g5.16xlarge Inference (Real-Time) | Model inference on the ml.g5.16xlarge instance type, real-time mode | $2.70 |
ml.g5.4xlarge Training | Algorithm training on the ml.g5.4xlarge instance type | $1.00 |
Dimensions summary
You pay by the hour based on which AWS instance size you run and what task you perform. Four instance sizes are available: ml.g5.4xlarge, ml.g5.8xlarge, ml.g5.12xlarge, and ml.g5.16xlarge. Each size supports three activities. Training builds or adapts the model. Batch inference processes grouped text in scheduled runs. Real-time inference returns results on demand. Larger instances carry more compute, so hourly costs vary across sizes. You mix and match instance size with the activity you need. Billing follows actual host hours used.
Top-of-mind questions for buyers
One host hour is one hour that a chosen instance runs your task. Billing counts actual running time on that instance. If you run two instances at once, each accrues its own host hours. Partial hours follow AWS metering. You pay only while the instance is active for training or inference.
Batch inference processes grouped text in scheduled runs, so you pay for hours during those runs. Real-time inference stays available to return results on demand, so you pay for hours the endpoint runs. Real-time can accrue hours even between requests if the endpoint stays up.
The model handles Spanish clinical text. It supports tasks like clinical note classification and entity recognition in clinical documents. It may not generalize to other languages or non-clinical domains. Your training and inference charges apply regardless of task, since billing follows instance host hours used.
huggingface.co
Helpful?
Vendor refund policy
No refund policy
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.
Delivery details
Amazon SageMaker algorithm
An Amazon SageMaker algorithm is a machine learning model that requires your training data to make predictions. Use the included training algorithm to generate your unique model artifact. Then deploy the model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Deploy the model on Amazon SageMaker AI using the following options:
Version release notes
This is the first version of our Spanish clinical text classification model. This initial release demonstrates our commitment to making Spanish machine learning resources. Our model has been trained on a diverse Spanish dataset to ensure robust performance and accuracy in various scenarios. While this is just the beginning, we are excited about the potential applications and improvements that future iterations will bring. We look forward to refining and enhancing our model based on user feedback and continued research.
Additional details
Inputs
- Summary
The fine-tuned classification model accepts as input a JSON object containing a list of texts.
{ "inputs": [ "El paciente presenta disnea progresiva y fatiga desde hace una semana.", "Se indicó iniciar tratamiento con metformina por diagnóstico reciente de diabetes tipo 2.", "No se evidencian signos de infección en la herida quirúrgica tras 10 días de la operación." ] }
- Input MIME type
- application/json, text/csv
https://github.com/iiconocimiento/iic-aws/blob/main/notebooks/rigoberta-clinical/data/input/classification_input.json
https://github.com/iiconocimiento/iic-aws/blob/main/notebooks/rigoberta-clinical/data/input/classification_input.json
Input data descriptions
The following table describes supported input data fields for real-time inference and batch transform.
Field name | Description | Constraints | Required |
|---|---|---|---|
inputs | A list of strings to be classified by the model. | Each text must be under 512 tokens. | Yes |
Resources
Vendor resources
Support
Vendor support
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
RigoBERTa 2.0 is a state-of-the-art encoder language model for Spanish, developed through language-adaptive pretraining. This model significantly improves performance on every previous Spanish encoder model offering robust language understanding.
We offer comprehensive professional services in Generative AI on AWS, supporting projects from design to deployment. Our work includes continuous benchmarking and selection of the best AWS tools, along with the creation of specialized tests to ensure reliable performance in production. We provide expert fine-tuning of Bedrock models and advanced RAG (Retrieval-Augmented Generation) configuration deployed on AWS to boost accuracy, relevance, and context awareness. Security is a priority: we deploy with robust guardrails to prevent off-topic or unsafe outputs, and ensure full regulatory compliance (e.g., EU AI Act standards). Finally, we help clients achieve significant cost savings through targeted optimization and compression of LLMs according to their specific use cases.
Customer reviews
No customer reviews yet
Be the first to review this product . We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.