diff --git a/ai4-metadata.yml b/ai4-metadata.yml index 31abc4a9..4e4e7ecd 100644 --- a/ai4-metadata.yml +++ b/ai4-metadata.yml @@ -2,15 +2,15 @@ metadata_version: 2.0.0 title: AI4life model loader summary: "Support for inference of the AI4LIFE model on the marketplace." description: |- - The [BioImage Model Zoo](https://bioimage.io/#/) is a community-driven platform that + The [BioImage Model Zoo](https://bioimage.io/#/) is a community-driven platform that provides standardized deep learning models for bioimage analysis. - + This module integrates models from the **BioImage.IO** package into the **AI4EOSC** Marketplace dashboard, specifically those using PyTorch weights and following the v0.5 format. - The module allows users to seamlessly explore, deploy, and utilize these models within + The module allows users to seamlessly explore, deploy, and utilize these models within the AI4EOSC ecosystem, providing a user-friendly interface for advanced bioimage analysis. - - **Key Features** + + **Key Features** - Model Discovery: Automatically fetch and list models available in BioImage.IO that meet the criteria (PyTorch weights, v0.5 format). - Metadata Visualization: Display essential information about each model, @@ -19,13 +19,13 @@ description: |- - Seamless Deployment: Enable one-click deployment of models to AI4EOSC compute resources. - Model Preview: Provide an interactive preview to test models on sample data directly in the dashboard. - + **References**: - BioImage.Io github repository: https://github.com/bioimage-io/core-bioimage-io-python - - Documentation: https://bioimage-io.github.io/core-bioimage-io-python/bioimageio/core + - Documentation: https://bioimage-io.github.io/core-bioimage-io-python/bioimageio/core dates: - created: 2024-11-11 - updated: 2024-11-11 + created: '2024-11-11' + updated: '2024-11-11' links: ai4_template: ai4-template-adv/2.0.1 source_code: https://github.com/ai4os/ai4os-ai4life-loader @@ -37,49 +37,20 @@ links: # base_model: http://add-some-link-to-another-model.com tags: # required property, add user-defined tags that you consider relevant - deep learning -tasks: # required property, uncomment together with relevant items - - Computer Vision - # - Natural Language Processing - # - Time Series - # - Recommender Systems - # - Anomaly Detection - # - Regression - - Classification - segmentation - 2D input - 3D input - UNet - - Segment anything - - Cellpose - # - Dimensionality Reduction - # - Generative Models - # - Graph Neural Networks - # - Optimization - # - Reinforcement Learning - # - Transfer Learning - # - Uncertainty Estimation - # - Other + - Segment Anything + - cellpose +tasks: # required property, uncomment together with relevant items + - Computer Vision + - Classification categories: # required property, uncomment together with relevant items - # - AI4 trainable - # - AI4 pre trained - AI4 inference - AI4 tools libraries: # required property, uncomment together with relevant items - # - TensorFlow - PyTorch - # - Keras - Scikit-learn - - bioimageio.core - # - LightGBM - # - CatBoost - # - Other data-type: # optional, uncomment together with relevant items - Image - - npy images - # - Time Series - # - Tabular - # - Graph - # - Audio - # - Video - # - Other - \ No newline at end of file