Skip to content

abhishekkgp/Machine-learning

Repository files navigation

Machine-learning

Code from scratch and also using Scikit-learn

Linear Regression

Understanding the data and simple curve fitting, Visualization of the fitted curves, Regularization all this part as given in Question(you can see question in Linear regression directory) has been included in "17CE10003_ML_A1.ipynb".

Logistic Regression and Decision Trees

(As per Problem Given)

1.All the 3 questions are solved in different part Nameing:

"17CE10003_A2_a,b,c" for question number 1,2,3 respectively.

2.To excute all this only chnage the path of 'winequality-red' in the initial of all three part and only once in each part.

3.All the part or each question are cleary solved sepertely in the same .py document with proper headlines and comment.

4.Also I have attached notebook code(naming ''ML_A2_Q1,2,3'') too so which is already excuted.

Clustering and Dimension Reduction

(As per Problem Given)

1.All the 5 questions are solved in different part Nameing: A.py, B.py, C.py, D.py and E.py

2.Name of data set is 'AllBooks_baseline_DTM_Labelled.csv' ,to excute all this only change the path of 'AllBooks_baseline_DTM_Labelled.csv' in the initial of all five part.

3.All the part or each question are cleary solved sepertely in the same .py document with proper headlines and comment.

  1. Clustres of different part is included in "clusters" folder with propely naming for each part in ".txt" file.

Name of clusters are: agglomerative.txt ---> from Agglomerative Clustering

              kmeans.txt 		     ---> from kmeans clustring
	      
	      kmeans_reduced.txt	     ---> from Kmeans Clustring on reduced data
	      
	      agglomerative_reduced.txt	     ---> from Agglomerative Clustering on reduced data

NMI score of agglomerative.txt : 1.8461794521809038

NMI score of kmeans.txt : 0.9187470250416958

NMI score of kmeans_reduced.txt : 1.0244221726618001

NMI score of agglomerative_reduced.txt : 1.8461794521809038

6.Also I have attached fully excuted notebook code for each part naming:

-->17CE10003_A.ipynb

-->17CE10003_B.ipynb

-->17CE10003_C.ipynb

-->17CE10003_D.ipynb

-->17CE10003_E.ipynb

Neural Network

(As per Problem Given)

1.All the parts are solved in different folder Nameing:

    dataset.py--> In Dataset folder

(part1_a.py, part1_b.py)---> In Part 1 folder

    part2.py-->In part 2 folder

2.Change the path of taking input and the place where you want to save training and testing data

3.All the part for each question are cleary solved sepertely in the same .py document with proper headlines and comment.

4.Also I have attached fully excuted notebook code for each part naming with the same name and same folder with ".ipynb" extension.