sub-rat / 3-MONTHS-ON-MachineLearning

Learning Machine learning each 2 hours a day for a 3 months

3-MONTHS-ON-MachineLearning

Learning Machine learning each 2 hours a day for a 3 months

Month1

Week 1:

Watched the linear algebra session of 3Blue1Brown. link of video to learn linear regression
Here explained is done in graphical way which makes good understanding of Linear Algebra, vectors , basis vectors, Composition Matrix for transformation , rotation ,shearing etc , Matrix multiplication, dot product, cross product, Eigen value and Eigen vector.

Implementation of data preprocessing using python library and some dataset in colab
link to the data preprocessing code

Learning and Implementation of Linear Regression using LinearRegression() model
link to the linear regression

Learning and Implementation of Multiple Linear Regression model
link to the multiple linear regression

Started learning basis of calculus. Essence of calculus

• Watched the video of basis of calculus ie. derivative and integration .

• How the circle, square or cube area is derived using the derivative.

• learned about the derivation of how velocity actually works with derivative rule.

• learned about role of the tangent in any kind of curve of graph.

• watched session Essence of calculus

• learned about Eulers number.

• learned about Implicit differentation

• http://www.math.ucsd.edu/~ebender/proofs.html

source: 3blueOneBrown

continue on Essence of calculus

• Limits
• L'Hopital's rule
• epsilon delta definitions
• Integration
• fundamental theorem of calculus

Week2:

• logistic Regression

• sigmoid function

• implementation of logistic Regression

• Logistic Regression on Titanic Data

• Logistic Regression Social Ads

• probabily

• data structure and algorithm

• KNN algorim and implementation

• knn with irisdataset

• Matrix Rank,multiplicaiton and operation

• Matrix Transformation

• Naive Bayes Classification

• SVM Classification on cancer data

Week3

• Random Forest
• OpenCV basics
  • Image
  • Video loading
  • Video Capturing
  • Drawing Function
  • link of above
• Pixel operation in Image
• XGBoost study
• Studying on NLP
  • Component of NLP
    • Morphological and Lexical Analysis
    • Syntactic Analysis
    • Semantic Analysis
    • Discourse Integration
    • Pragmatic Analysis

week4

• Topic Frequency-Inverse Document Frequency (TF-IDF)
• Basic and Simple workout on Non-negative Matrix Factorization (NMF)
• hands on knowledge of Singular Value Decomposition (SVD)
• knowledge on Stochastic Gradient Descent (SGD)

Month2

Week 1:

• Machine Learning Crash course started

  • key ML Terminology

  • Linear Regession

  • Training and loss

  • Gradient Descedent , learning rate, epoch, batch

  • SGD

  
  

  • Introduciton with tensor flow

  • Generalization

  • Training, Test and validation

  • Representation of data

  • Cleaning data

  
  

  • Feature Crosses( Encoding Nonlinearity and Crossing One Hot Vector)

  • Regularization ( L2 Regularization and lambda)

  • Logistic Regression ( Loss and Regulariztion) (log loss)

  
  

  • TP,TN, FP,FN

  • Classification

  
  

  • ROC Curve and AOC

  • L1 Regularization

  • Neural Network

  • Neural Netork pracice

  • Activation Function(Relu, Linear, Sigmoid, tanh)

• PCA watched youtube videos and done some Python implementation

• PCA on Breast Cancer Dataset

• PCA on CIFAR-10 Dataset

Week 2

• Convolutional Neural Networks in Python with Keras
• Speedup deep learning with pca on CIFAR-10 dataset
• MultiClass Neural nets
• softmax
• Collaborative filtering
• Diving deep into Convolutional Neural Network link to reference
• using pre trained model pytorch

week 3

• using pre trained model
• pre trained model tensorflow
• hugging face
• transfer learning

week 4

• TFlite android samples exploring
• Object detection in android
• exploring on Ml with Mobile
• Try other differnet TFLite sample and with programming implementation

Month3

Week 1:

• Detectorn2
• Human Hair Segmentaion research paper reading
• Exploring Mlkit and vision api overview
• mediapipe exploring https://github.com/google/mediapipe

week2:

• knowledge on RNN
• LSTM network
• Understanding Encoders and Decoders Sequence to Sequence model
• Understanding GRU Networks