Skip to content

Latest commit

 

History

History
70 lines (34 loc) · 2.36 KB

README.md

File metadata and controls

70 lines (34 loc) · 2.36 KB

Machine Learning & Deep Learning

Portofolio about Machine Learning and Deep Learning Project

1. Churn Prediction

rf

  • first excercise machine learning
  • using dataset about telco churn dataset
  • training modelling using Logistic Regression, Random Forest, and Gradient Boosting
  • Random Forest make better result for training model

2. Customer Segmentation

kpro

  • modelling using K-prototypes
  • using 3 categories features and 2 numerical features

3. Twitter's Sentiment Analysis

twitter

  • sentiment analysis using tweepy and python readlines
  • with search topic about "uu ite" and "kritik"
  • the result about topic is dominate by negative sentiment

4. Face recognition using OpenCV

opencv

  • Face recognition using Haarcascade
  • face recognition training using 4 difference face

5. AUS's Weather prediction

opencv

  • Weather prediction using ANN modelling using pytorch
  • had 2 hidden layer each layer with 5 neuron and 3 neuron
  • after 1000 epoch of training model the result get :
  • accuracy : 83%, loss : 0.404
  • had a good recall for no raining prediction (96%) but worst for raining prediction (40%)

6. MNIST: LRvsANNvsCNN

mnist

  • MNIST prediction using Logistic Regression,ANN and CNN modelling using pytorch
  • after 15 epoch of training model the result get :
  • CNN had better result between each other (acc:88%,loss:0.40)
  • ANN get lower result from logistic regression, maybe due to had unpotimized ANN model (1 hidden layer with 50 neuron)