jiqicn/meteovis

A jupyter-based tool for visualizing and exploring meteorological and bioecological data hosted at UvA-TCE.

MeteoVis

A jupyter-based tool for visualizing and exploring meteorological and bioecological data hosted at UvA-TCE.

Documentations of APIs can be found here.

Usage

Process files to dataset:

Visualize datasets:

Merging datasets:

Update datasets:

Installation

Using Anaconda:

conda config --add channels conda-forge
conda config --set channel_priority strict
conda create -n <name-of-your-env> python=3.8  # to install Meteovis to a new environment
conda activate <name-of-your-env>
conda install -c jiqi meteovis

When installing into JupyterLab 1 or 2, you may also need to install the labextensions if you have never done it before:

jupyter labextension install @jupyter-widgets/jupyterlab-manager jupyter-leaflet @j123npm/[email protected]

For now, MeteoVis is only available on AMD64 Linux systems, with [email protected].

meteovis's Issues

Pre-buffering of images when the views are loaded

Get the vis functions back

Realize the tag view in the dataset operation GUI, and implement the visualization tag that allows visualizing a single raster on
an interactive map.

Update the option widgets after choosing new files to process

The merging method

Merge temp files to the final dataset.

• Currently support average, maximum, and minimum.
• Delete all the temp files after finishing.
• Test performance on that, better display the merging result on maps to see if there is any offset of location.

Add more instructions within the GUIs

The following information should be shown to users

When processing data:

• Only files from the same radar can be selected when processed into a dataset

When operating dataset:

• Only one dataset can be updated at a time
• List possible operators when updating a dataset, also show examples

Dataset update by equation input

realize math and logic operations of datset

• user's input should be a string of equation, supporting normal math and logic operations
• equation string will be parsed and executed to update the dataset
• in the current stage, only allow one dataset to be updated, but could be expanded to multi-dataset computing in the future (the challenge is how to refer to multiple datasets in the equation string).
• The updated dataset will be stored as a separate dataset file

Create the pipeline of packing and uploading meteovis to Anaconda

Design of parallelism in raster merging

General idea:

1. each raster in each dataset will first be interpolated
2. the same raster of all datasets will be merged
3. merging result will be written to the output dataset file

Clean cache

In the visualization tab, add the button for cleaning the image cache.

Input all files under the selected folder

When processing data, it requires that all the files to be processed are directly under the selected folder. It's preferred that files in the recursive subfolders are also selected.

Copy of datasets

Button within the general tab to allow duplicating a dataset.
Only need to change the name and id of the dataset:

• name to copy_original_name
• id to a new uuid

Implement the interpolate function of rasters

To merge two rasters, a shared grid will be computed based on the input rasters, and both input rasters will be interpolated to this shared grid.

Take care of the nan values after interpolation.

Add bbox information in meters to the meta group

This is for raster merging, to compute on the real dimension.