Skip to content

Commit

Permalink
Update README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
Spiruel authored Mar 1, 2024
1 parent d6dd596 commit 228e48c
Showing 1 changed file with 11 additions and 16 deletions.
27 changes: 11 additions & 16 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,36 +1,31 @@
# SciML Leeds SENSE training

Welcome to the Scientific Machine Learning Training page! In this training, we will delve into the fascinating intersection of scientific research and machine learning, specifically focusing on the application of autoencoders to explore satellite imagery over the UK.

## What is Scientific Machine Learning?
Scientific Machine Learning (SciML) is an emerging field that combines the power of traditional scientific methods with modern machine learning techniques. It aims to tackle complex scientific problems by leveraging the vast amounts of data available today, enabling researchers to gain deeper insights, make predictions, and discover new patterns.

## Using Autoencoders for Satellite Imagery Exploration
In this training, we will use autoencoders, a type of artificial neural network, to analyze and extract meaningful information from satellite imagery covering the landscapes of the UK. Autoencoders are particularly useful for tasks like image denoising, feature extraction, and even generation of new images.

## What to Expect

### Introduction to Autoencoders: We will start with an overview of autoencoders, understanding how they work and their applications in image analysis.
### Introduction to Autoencoders:
We will start with an overview of autoencoders, understanding how they work and their applications in image analysis.

### Processing Satellite Imagery: Learn how to preprocess and prepare satellite imagery data for input into the autoencoder model.
### Processing Satellite Imagery:
Learn how to preprocess and prepare satellite imagery data for input into the autoencoder model.

### Training the Autoencoder: Step through the process of training an autoencoder model using TensorFlow, exploring how it learns to represent the unique features of UK landscapes.
### Training the Autoencoder:
Step through the process of training an autoencoder model using Keras, exploring how it learns to represent the unique features of UK landscapes.

### Visualizing Results: After training, we will visualize the reconstructed images, gaining insights into the model's understanding of the satellite data.
### Visualising Results:
After training, we will visualise the reconstructed images, gaining insights into the model's understanding of the satellite data.

# Getting Started
To get the most out of this training, please follow along with the provided materials:

Colab Notebook: Access our interactive Colab notebook where you can run code and experiment with the autoencoder model in real-time.

<a target="_blank" href="https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_monitoring/model_monitoring.ipynb">
<a target="_blank" href="https://colab.research.google.com/drive/1PH1J9jEvPsdK1z23YQsthtJ_KMm71PQr?usp=sharing">
<img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/>
</a>

Introductory Slides PDF: Download the introductory slides for a comprehensive overview of the topics we will cover in this training.

We are excited to embark on this journey into Scientific Machine Learning with you. Let's dive in and explore the rich world of satellite imagery over the UK through the lens of autoencoders! 🛰️🔍
[Introductory Slides](https://docs.google.com/presentation/d/1SUI94od5W9b2ikHK8bQNPYzOQ15oN8REe7p9FWBEPH8/edit?usp=sharing)

If you have any questions or need assistance, please don't hesitate to reach out to our team.
🛰️🔍

Happy learning!

0 comments on commit 228e48c

Please sign in to comment.