Skip to content

arialblack14/Machine-Learning

 
 

Repository files navigation

1. Linear regression

Just some basic linear regression implemented using scikit-learn as well as on our own using gradient descent

Data set

https://www.kaggle.com/gregorut/videogamesales

2. Emotion classification

Emotion recognition using CNNs. Attempt 1 scored a 83.5%, Attempt 2 scored a 84.0% in kaggle competition.

Data set

https://www.kaggle.com/c/cs4990-assignment-2-option-3-facial-expression/data

Model

VGG19 inspired model

  • Layers: 64 - 128 - 256 - 512 - 1024, 4096 full connected

3. Hair segmentation

Hair segmentation using Unets. Best result was 77.9%.

Data set

https://www.kaggle.com/c/cs4990-assignment-3-option-1-hair-segmentation/data

Model

Medical processing inspired Unet

  • Layers: 64 - 128 - 256 - 512 - 1024 - 512 - 256 - 128 - 64

4. Spam detection

A practice project. Recognizing spam text messages using a recurrent neural net.

Data set

https://www.kaggle.com/uciml/sms-spam-collection-dataset

Tutorial

https://www.kaggle.com/kentata/rnn-for-spam-detection/notebook

About

CS4990 - Special Topics Course

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 94.7%
  • Python 5.3%