SWAGATIKA GIRI's repositories
DECISION-TREE
The aim of this project is to print steps for every split in the decision tree from scratch and implementing the actual tree using sklearn. Iris dataset has been used, the continuous data is changed to labelled data. In this code gain ratio is used as the deciding feature to split upon.
TEXT-CLASSIFICATION
The aim of this project is: 1.Perform Text Classification using Multinomial Naive Bayes 2. Implement Naive Bayes from scratch for Text Classification. 3. Compare Results of self implemented code of Naive Bayes with one in Sklearn. dataset used is 20_newsgroups
CAB-CANCELLATION
The goal of the project is to create a predictive model for classifying new bookings as to whether they will eventually get cancelled due to car unavailability. This is a classification task that includes misclassification costs. Dataset used: https://www.kaggle.com/c/predicting-cab-booking-cancellations/overview
COLOR-EXTRACTION
This projects aims to extract color from an image.
DICTIONARY
A dictionary is made using selenium in python.
FACE-CLASSIFICATION
The aim of this project is to classify the faces. Olivetti Faces dataset has been used. In this dataset there are ten different images of each of 40 distinct subjects. For some subjects, the images were taken at different times, varying the lighting, facial expressions (open / closed eyes, smiling / not smiling) and facial details (glasses / no glasses). All the images were taken against a dark homogeneous background with the subjects in an upright, frontal position (with tolerance for some side movement). The “target” for this database is an integer from 0 to 39 indicating the identity of the person pictured. Each of the sample images needs to be classified in the classes ranging from 0 to 39. PCA has been applied to reduce the dimensionality. Then various classification and regression techniques are used with and without using PCA and the accuracy and time taken by the algorithms are recorded. Algorithms used: SVM, KNN, logistic regression, neural networks, linear regression and random forests.
K-MEDIODS
The aim of this project is to implement k-mediods algorithm of unsupervised learning from scratch. 3 random numpy arrays(2-D) have been taken into consideration for this project. This code can be used to partition any given dataset into 'n' clusters where n can be any real number of user's choice.
Lungs-X-ray-Classification
The aim of this project is to import image files from Google Drive, the images are chest X-rays. Each X-ray in the set is that of a patient with either pneumonia or a normal lung field. There are two types of infection: bacterial and viral. The aim is to predict if a previously unseen chest X-ray is normal or if it depicts either bacterial or viral pneumonia.
TRAFFIC-COUNTER
This project aims to count the number of vehicles passed from an area using OpenCV library.