Skip to content
Kaggle Competition consisting in using satellite data to track the human footprint in the Amazon rainforest. (https://www.kaggle.com/c/planet-understanding-the-amazon-from-space)
Jupyter Notebook
Branch: master
Clone or download

README.md

Planet--Understanding-the-Amazon-from-Space

Kaggle Competition consisting in using satellite data to track the human footprint in the Amazon rainforest. (https://www.kaggle.com/c/planet-understanding-the-amazon-from-space)

The Powerpoint presentation "Project Presentation.pdf" outlines the objective and methodology of this project.

  1. Open Colab

Weather Classification

  1. Load the data for weather classification Run the notebook "downloading_from_kaggle.ipynb" to load the data on Colab virtual environment

Simple model from scratch

  1. Run the notebook "building_simple_model.ipynb" on Colab, without resetting the virtual environment from 2

A deeper architecture

  1. Run the notebook "model2_hidden_layers.ipynb" on Colab, without resetting the virtual environment from 2

Object Classification: Road VS Water

  1. Load the data for binary (road vs water) classification. Run the notebook "binary_downloading_from_kaggle.ipynb" to load the data on Colab virtual environment

Resnet Model and its Class Activation Maps

  1. Run the notebook "building_resnet_model_with_activation.ipynb" on Colab, without resetting the virtual environment from 5
You can’t perform that action at this time.