⚠️ Note: This is an ongoing project under active development. Features, documentation and code may change.

Note: This model was used to perform analysis on the CMIP6 dataset. The code for that can be found in this github repository: mphysicus/atmospheric-rivers-cmip6

DeepAR is a deep learning model designed for Atmospheric Rivers (AR) detection and segmentation from climate data (using the Climate variable IVT, IVT_u, IVT_v). It utilizes a modified, prompt-less Segment Anything Model (SAM) to generate AR masks.

The DeepAR model processes data through a three-stage pipeline:

1. Input Generator (IVT2RGB): A CNN that converts 3 channel climate data (Integrated Vapor Transport: ivt, ivtu, ivtv) into a 3 channel RGB-like image suitable for the image encoder.
2. Segmentation (SamAR): A modified SAM model that operates without prompts. It uses a learned no_mask_embedding (This replaces the prompt encoder of the original SAM model) to generate segmentation masks from the features produced by the image encoder.

The diagram below illustrates the architecture: 

1. Clone the repository:

   git clone https://github.com/mphysicus/deep_AR.git
   cd deep_AR

2. Install the package:

   pip install -e .

The model is designed to work with NetCDF files (.nc) containing ivt, ivtu and ivtv variables. Use the dataset class ARInferenceDataset for loading and preprocessing the data during inference.

For a quick preview of how to use the model for inference, please refer to this notebook file in the repository: demo.ipynb

✨Coming Soon✨We will be uploading pre-trained model weights on Huggingface soon.

We are thankful to Segment Anything for releasing their code as open-source contributions.