Skip to content

Using unsupervised machine learning, Keras, hyper parameters, dendrograms, CNNs and GANs to predict climate changes in Europe.

Tanu-Shree-git/ClimateWin-Unsupervised-Machine-Learning-

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

18 Commits
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

Hi, I am Tanu shree!

Aspiring Data Analyst focused on building a career in data analysis, with a special interest in cybersecurity trends.

πŸš€ Project Details

Using machine learning to help predict the consequences of climate change for fictional company ClimateWins based in Europe. This is a project designed to challenge the junior data analyst by introducing machine learning skills in Python. The project mimics a fictional nonprofit organization (CliamteWins) with limited funding that does not have a data scientist or data engineer team. This challenge guides the analysts as a trainee, data scientist and researcher all at once to achieve ClimateWins goals.

This project contains unsupervised learning models including:

  1. Steps to scale data to make it easier to use in machine learning models.
  2. Dendrogram and Principal Component Analysis (PCA)
  3. Deep Learning in Keras Convolution Neural Networks (CNN) and Recurrent Neural Networks (RNN)
  4. Decision Trees and Random Forest
  5. Hyperparameters and Tuning Models:
    1. Random Search
    2. Bayesian Search
  6. Handwriting Recognition with Convolution Neural Networks (CNN) & MNIST
    1. Radar Recognition with Generative Adversarial Networks (GAN)

πŸ”Ž Project Objective

  1. Identify weather patterns outside the regional norm in Europe.

  2. Determine if unusual weather patterns are increasing.

  3. Generate possibilities for future weather conditions over the next 25 to 50 years based on current trends.

  4. Determine the safest places for people to live in Europe over the next 25 to 50 years.

Data Source

This data is sourced from the European Climate Assessment & Dataset (ECA&D).

The project uses weather data from 18 weather stations across Europe, covering the period from the late 1800s to 2022. The dataset includes daily observations of variables such as: -Temperature -Wind speed -Snowfall -Global radiation -And other meteorological indicators

πŸ› οΈ Skills & Tools

Python – for data analysis, model building, and evaluation PowerPoint – for presenting findings and visualizations

πŸ“’ Notebooks

Check out my Jupyter notebooks scripts in the attached folder 'Scripts', that demonstrate my data wrangling, analysis, and visualization skills.

youtube presentation link : https://www.youtube.com/watch?v=m0KEIutMzU4&t=1s

About

Using unsupervised machine learning, Keras, hyper parameters, dendrograms, CNNs and GANs to predict climate changes in Europe.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published