Skip to content

Latest commit

 

History

History
112 lines (110 loc) · 2.85 KB

README.md

File metadata and controls

112 lines (110 loc) · 2.85 KB

21-DAYS-PROGRAMMING-CHALLENGE-ACES

@@ Exploring sklearn! @@

Bit intro About library

A python library built upon NumPy ,SciPy and Matplotlib orignal name scikit-learn.

Installation

pip install scikit-learn

Features of Sklearn
  1. Supervised Learning Model
  2. Unsupervised learning Model
  3. Clustering
  4. Dimenstionality Reduction
  5. Ensemble Methods
  6. Feature Extraction
  7. Feature Selection
  8. Open Source
💠 Day 1 :Sklearn Modelling Process:
  1. Loading ,splitting data
  2. Training Model
  3. Model Persistence
  4. Preprocessing the Dataset(Binarisation,Mean Removal ,Scaling,Noemalisation(L1,L2 normalisation))
💠 Day 2:Linear Modelling :
  1. Linear Regression (SL)(Regression) ( logit or MaxEnt Classifier)
💠 Day 3:Linear Modelling :
  1. Logistic Regression (SL)(Classification)
  2. Lasso
  3. Ridge
  4. ElasticNet
💠 Day 4:Gradient Descent Algorithm
  1. Batch Gradient Descent
  2. Stochastic Gradient Descent
  3. Mini Batch Gradient Descent
💠 Day 5:Suppot Vector Machine
  1. SVM (SL,Classification+Regression)
💠 Day 6:KNN Algorithm
  1. KNN as Classifier (SL,Classification+Regression)
  2. KNN as Regressor
💠 Day 7:Metrics and scoring

(Not did much read a bit theory)

  1. Confusion_matrix
  2. Accuracy
  3. Precision
  4. Recall or Sensitivity
  5. Specificity
💠 Day 8:PCA
  1. Incremental PCA (UL + dimensionality Reduction)
  2. Kernel PCA
💠 Day 9:Tree
  1. Decision Tree (ID3 iterative dichotomiser 3)(SL,CART)
  2. Random Forest
💠 Day 10:Naive Bais
  1. Gaussian Naive Bayes (Classification)
💠 Day 11:Dimension Reduction
  1. Principal Component Analysis(PCA)
💠 Day 12:Dimension Reduction
  1. Singular Vector Decomposition(SVD) [not did much today kam hai kafi!]
💠 Day 13:Ensemble methods
  1. Voting Classifier
    Soft Voting + with GridSearchCV
💠 Day 14:Gradient Boosting

Read theory about all

  1. GBA
💠 Day 15:DATA PROCESSING

Steps involved in data processing

  1. Treating up missing values
  2. Treating outliners
  3. Dimentionality Reduction
  4. Variable Transformation and Feature Engineering
💠 Day 16:Recommender System
  1. Simple REcommende using IBM formula
💠 Day 17:Recommender System
  1. Content based Recommendation(tfid)
💠 Day 18:Mean shift Clustering Algorithm
💠 Day 19: Not a good day
  1. Not having laptop with me😥 signed in through phone will read about different types of regression. no code today 😔.
💠 Day 20: Pipeline
  1. How to create one and use.
    Laptop didn't come today.
    Now I am pro at using GitHub on phone.💁
💠 Day 21 Anomaly detection

RESOUCES

Tutorial

Happy to complete this Chanllenge and for sure will continue Learning! 😊