javedali99/Machine-Learning-Notes

The repository contains notes related to different methods of Machine Learning

Machine Learning Notes

The repository contains the work related to Machine Learning and some notes from what i have learned in machine learning.

Custom Function

1. Description Function
2. Visual Function
3. Advance Visual Function
4. Feature Engineering Function
5. Feature Importance Function
6. Scoring Function
7. Feat_imp Function
8. Image Processing Function
9. Statistic Function
10. Model Function
11. Map Function
12. Time Series Function
13. Tunning Function
14. Interpretable Function
15. Cluster Function
16. Generate Data Function

Supervised Learning

1. Decision Tree
2. Random Forest
3. SVM
4. Gradient Boosting
   1. Extreme Gradient Boosting Learning Curve
   2. GPBoost_Tree-Boosted Mixed Effects Models
   3. Gradient Boosting Classifiers
5. Multi-Class Classification
   1. One-vs-Rest and One-vs-One for Multi-Class Classification
6. Ensemble Methods
   1. Ensemble Method (Hard and Soft Voting)
   2. Weighted Average Ensembles
   3. Dynamic Ensemble Selection (DES)
   4. Super Learner Ensembles
7. Thesis
   1. Decision Tree
   2. Bagging for DT
   3. Random Forest
   4. Gradient Boosting
   5. Adaboost
   6. Discriminant Analysis
   7. K-Nearest Neighbors
   8. Logistic Regression
   9. Naive Bayes
   10. Neural Networks
   11. SVM Linear
   12. SVM RBF
   13. SVM Sigmoid
   14. Stochastic Gradient Descent (SGD)
   15. ROC
8. Use cases
   1. Ruang_Guru - Predict Probability of Correct Answer
      1. Dataset
      2. Result
      3. Test Context
      4. Result in ipynb
      5. Result in PDF

Preprocessing

1. Weight of Evidence (WOE) and Information Value (IV)

NLP

1. Text Classification
   • AT_PA text classification
   • Text Classification - Simple NLP with multilabel
2. Text Generation
3. Text Summarization
4. Text Manipulation
5. Facebook FastText Library
   • Semantic Similarity
   • Text Classification
6. Bag of word
7. TF-IDF
8. Word2Vec
   • Word2Vec with Gensim
   • Word2Vec with skip-gram

Tunning Model

1. Grid Search
2. GridSearchCV
3. Tunning with Nature Inspired Algorithms
4. Hyperopt
5. Keras tuner
6. Optuna

Neural Network

1. Basic Neural Network
2. Deep Neural Networks
3. Generative Adversarial Networks (GANs)
4. Improved GANs

Autoencoders

1. AutoEncoders vanilla
2. AutoEncoders denoiser
3. Autoencoders are related to PCA [ongoing]
4. Denoising - Autoencoder

Generative Adversarial Networks (GANs)

1. Deep Convolutional GAN - (DCGAN)
2. Conditional GAN - (CGAN)
3. Wasserstein GAN - (WGAN)
4. Least-squares GAN - (LSGAN)

Novelty, Anomaly and Outlier Detection

1. Anomaly detection vanilla
2. Anomaly detection using PCA [ongoing]
3. Anomaly detection using Scikit-learn
4. Anomaly detection using AutoEncoders (PYOD)
5. Python Outlier Detection (PYOD)
6. Anomaly detection using Isolation Forest
7. Anomaly Detection using Benford’s Law
8. Anomaly Detection using LSTM
9. Anomaly Detection using OCSVM
10. Fraud detection using DBSCAN
11. Anomaly detection using AutoEncoders
12. Anomaly detection using Variational AutoEncoders
13. Denoising - Autoencoder
14. Anomaly Detection using the Restricted Boltzmann machines (RBM)
15. Anomaly detection with LSTM (time series data)
16. Anomaly Detection with Temporal convolutional networks (TCN)
17. Anomaly Detection with Encoder Decoder - Temporal convolutional networks (ED - TCN)

Association Rule

1. Apriori
2. Eclat Algorithm
3. Frequent Itemsets
4. Frequent Itemsets via Apriori Algorithm
5. Frequent Itemsets via FP-Growth Algorithm
6. Maximal Itemsets via FP-Max Algorithm

Interpretable Model

1. Feature Interaction
2. Individual Conditional Expectation (ICE)
3. Local Interpretable Model-agnostic Explanations (LIME)
4. Partial Dependence Plot (PDP)
5. Random Forest Interpretation [ongoing]
6. SHapley Additive exPlanations (SHAP)
7. Interpretable Function

Dimensional Reduction

1. PCA
2. RBF Kernel Principal Component Analysis (RBF PCA)

Computer Vision

1. Simple CNN
   1. Rock_paper_scissors (simple CNN with ImageDataGenerator)
   2. CIFAR 10 (simple CNN)
   3. CIFAR 10 v2 (simple CNN)
   4. MNIST (simple CNN)
2. Simple CNN with Image Data Generator
   1. Malaria (simple CNN with ImageDataGenerator)
   2. Fashion MNIST (simple CNN)
3. LeNet
   • LeNet for classification
4. EAST - Text Detector
5. Y-Network CNN
6. [ResNet]
   • Resnet on the CIFAR10 dataset
7. [DenseNet]
   • DenseNet on the CIFAR10 dataset
8. Automatic Colorization Autoencoder
9. [Mobile Net V2]
   • Image Classification - MobileNetV2 - imagenette2
10. Image Processing Function

Time Series

1. Seasonal and Decomposition
2. Decomposition and Exponential Smoothing
3. AR, MA and ARIMA
4. Time Series Forecast with Machine Learning
5. Time series with ARIMA
6. Change and convert time and date data
7. Sequence Problems with LSTM
8. Time Series Prediction using LSTM and PyTorch
9. Predict Visitor Forecasting

Clustering

1. Simple Customer Segmentation Using RFM Analysis
2. Select Optimal Number of Clusters
3. Customer Segmentation Using Cohort and RFM Analysis
4. Cluster Function

Quality Control Case

1. Predictive Maintenance detect Faults from Sensors with CNN
2. Predictive Maintenance detect Faults from Sensors with CRNN and Spectrograms
3. Quality Control with Machine Learning

Rare Event

1. Extreme Event Forecasting with LSTM Autoencoders
2. Remaining Life Estimation
3. Survival Analysis with LightGBM plus Poisson Regression
4. LSTM Autoencoder for Extreme Rare Event Classification

Recommendation Engine

1. Movie Lens Collaborative Filtering
2. Cosine similarity

Sound

1. Speech Recognition

Geospacial

1. Change Detection in Multitemporal Remote Sensing Images
   1. Unsupervised PCA & K-Means
   2. Unsupervised Deep Slow Feature Analysis
   3. Supervised DSMS-FCN
2. Image-to-Image Translation
   1. Pix2Pix GAN
3. Map Data Visualization
   1. Map Function