Techno Blender
Digitally Yours.
Browsing Tag

Cordeiro

How to Learn Geospatial Data Science in 2023 | by Maurício Cordeiro | Feb, 2023

A step-by-step guide for those wanting to learn geospatial data analysis with PythonDigital image created with Dall-E 2. Caption: Photo of a scientist holding the Earth.Why is this topic relevant? It’s no secret that data science careers are in high demand these days. And when you add the dimension of geospatial analysis to the mix, the possibilities become even more exciting. Climate change, food production, and transitioning to a carbon-free economy are just a few of the many important issues that require a deep…

Artificial Intelligence for Geospatial Analysis with Pytorch’s TorchGeo (part 3) By Mauricio Cordeiro

An end-to-end deep learning geospatial segmentation project using Pytorch and TorchGeo packagesPhoto by NASA on UnsplashIn the previous stories (Part 1 and Part 2), we saw how to prepare a dataset of raster (multispectral) images and combined them with the corresponding labels (ground truth masks) using the IntersectionDataset provided by TorchGeo. To draw samples from it (smaller fixed-sized patches required for training), the RandomGeoSampler was used altogether with the DataLoader object (responsible for providing the…

Artificial Intelligence for Geospatial Analysis with Pytorch’s TorchGeo (part 2) | by Maurício Cordeiro | Nov, 2022

An end-to-end deep learning geospatial segmentation project using Pytorch and TorchGeo packagesPhoto by Shubham Dhage on UnsplashIn the previous story (Part 1 — here), we saw how to create a RasterDataset using TorchGeo and a RandomSampler to draw patches from it. In the current story, we are moving further and we will concatenate both the images and the masks into a IntersectionDataset, to draw tuples with training patches (images) and corresponding labels (water masks).So let’s get started.We will replicate the code…

Artificial Intelligence for Geospatial Analysis with Pytorch’s TorchGeo (Part 1) | by Maurício Cordeiro | Nov, 2022

An end-to-end deep learning geospatial segmentation project using Pytorch and TorchGeo packagesPhoto by NASA on UnsplashAccording to its documentation, TorchGeo is a “PyTorch domain library providing datasets, samplers, transforms, and pre-trained models specific to geospatial data”. The cause is noble. Make it easier for practitioners to use Deep Learning models on geospatial data. And why is that a good deal?In a last years’ presentation from Dan Morris (former principal scientist at Microsoft’s AI for Earth program) to…