kemoch Goto Github PK

course-v3

The 3rd edition of course.fast.ai

forest-type-mapping

I have used the output data of a study which classified different forest types based on their spectral characteristics at visible-to-near infrared wavelengths, using ASTER satellite imagery. The aim was to identify and/or quantify the ecosystem services (e.g. carbon storage, erosion protection) provided by the forest. I have achieved my goal leveraging on Decision Tree Classifier, Cross Validation, Random Forest & XG Boost.

ionosphere

Applying some Data Wrangling, I analysed a dataset that describes the status of free electrons passing through the ionosphere to study its Summary Statistics, Principal Component Analysis and Outliers. Finally, I created a multi-boxplot of the ionosphere measurement fields

kaggle-nyc-taxi

nytaxifare

portfolio

Portfolio Overview Cisco Webex Teams Chatbot. I have developed a virtual assistant using Dialogflow to promote automation and decrease my workload as client infrastructure management transformation lead for a world's leading financial institution. (Confidential) 'What's in the fridge?' Object Detection. Leveraging on transfer learning, I trained an image classifier which can recognize vegetables & fruits at state of the art accuracy (+90%). The algorithm is build on top of the Fastai library and run in the cloud on a Colab GPU. Titanic survival rate prediction & Data Visualization. I have worked with the Titanic dataset from Kaggle to build a model that combines supervised and unsupervised learning to predict the survival rate of an individual. I have represented my results using Matplotlib & Seaborn libraries. Forest type mapping. I have used the output data of a study which classified different forest types based on their spectral characteristics at visible-to-near infrared wavelengths, using ASTER satellite imagery. The aim was to identify and/or quantify the ecosystem services (e.g. carbon storage, erosion protection) provided by the forest. I have achieved my goal leveraging on Decision Tree Classifier, Cross Validation, Random Forest & XG Boost. Ionosphere Data Wrangling. I analysed a dataset that describes the status of free electrons passing through the ionosphere to study its Summary Statistics, Principal Component Analysis and Outliers. Finally, I created a multi-boxplot of the ionosphere measurement fields. Temperature forecast. A Simple High and Low temperature prediction using linear regression.

temperatureforecast

A Simple High and Low temperature prediction model using linear regression.

titanic

I have worked with the Titanic dataset from Kaggle to build a model that combines supervised and unsupervised learning to predict the survival rate of an individual. I represented my results using Matplotlib & Seaborn libraries.

whats-in-the-fridge

Leveraging on transfer learning, I trained an image classifier which can recognize vegetables & fruits at state of the art accuracy (+90%). The algorithm is build on top of the Fastai library and run in the cloud on a Colab GPU.