🛎️Welcome
Introduction
Knowledgator specializes in developing open-source ML models designed for information extraction (IE) tasks. Our models are trained to work for a range of IE tasks:
Text classification
Named-entity recognition
Relation extraction
Entity linking
Question-answering
Reranking of search results
Text summarization
Text cleaning
Coreference resolution
Access our models
Knowledgator currently provides two types of access to information extraction models:
Open-source access: our models are available under the Apache 2.0 License on HuggingFace. These models can be freely downloaded, modified to suit your needs and run on your custom instances.
API as a Service: for users seeking ready-to-use solutions, we provide API services with flexible pricing plans to suit various scales and requirements, available on RapidAPI.
Models
Comprehend-it
Next-generation sequence classification model designed for multi-label text classification tasks, with output scoring. Works in zero and few-shot learning settings. Could be used for:
Text classification
Reranking of search results
Named-entity recognition (NER)
Relation extraction
Entity linking
Q&A
Salamandra
Universal encoder-based token classification model capable of working with a wide range of popular IE tasks. The model is prompt-based and is seamlessly configurable through prompts. Supported IE tasks:
Named-entity recognition (NER)
Relation extraction
Summarization
Q&A
Text cleaning
Coreference resolution
📣 Custom Solutions: If you have unique requirements or need tailored solutions, our team is here to assist. Contact us to explore custom development opportunities.
Frameworks
LiqFit: Python framework used for few-shot fine-tuning of a classification model(Comprehend-it).
Support
Join our Discord community to engage in discussions about ML and information extraction, share your experiences, and connect with our engineering team. For enterprise inquiries, more information about our models, or requests for additional features, please reach out to our Business Team.
Resources
Last updated
