Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
title: EarthView Viewer | |
emoji: 🦀 | |
colorFrom: indigo | |
colorTo: yellow | |
sdk: gradio | |
sdk_version: 4.28.3 | |
app_file: app.py | |
pinned: false | |
license: mit | |
This is a Gradio based application to browse the [satellogic/EarthView](https://huggingface.co/datasets/satellogic/EarthView) Dataset | |
![](EarthView-Satellogic-RGB.png) | |
Options in the interface | |
======================== | |
**Dataset**: satellogic/EarthView. You shouldn't need to change it, unless you want to use a compatible (forked) dataset. | |
**Subset**: Choose the subset you want to use, Satellogic, neon or Sentinel-1 | |
**Split**: This version of the dataset has only one split. | |
**Initial Shard**: Use it to jump directly to a particular shard on the subset. You can use -1 to open the whole subset, but it’s slower. | |
**Only RGB**: Subsets have more than one image type. Satellogic has RGB and NIR, Neon has RGB, Hyperspectral and CHM, etc (see the documentation for the dataset) | |
**Load**: This will load the subset and show images. Before clicking it you can choose the batch size below. | |
**Batch Size**: How many images to load and show at once. 100 is a nice number. | |
**Columns**: Controls the number of columns in the image gallery. It’s useful to align the number of columns with the images per sample. For example, Sentinel-1 has 3 images per sample, so 3, 6, 9 is a good number. When you choose to see more than just RGB images, Satellogic’s will have RGB and NIR per sample, so an even number of columns is better. | |
**Shard**: It allows you to jump to a new shard. | |
**Next Batch**: Will load and show the next batch of samples | |
The table shows the metadata for each sample. | |