sunandabiswas Goto Github PK

100daysofmlcode

This repo is for the Siraj's 100 Days Of ML Code Challenge

abigsurvey

A collection of 700+ survey papers on Natural Language Processing (NLP) and Machine Learning (ML)

advanced-sql_

alzheimer-s-disease-prediction---multiclass-image-classification

This is a multiclass image classififcation problem. Here, to classify four types of disease different image classification models has been implemented.

bangladeshi-restaurants-food-reviews-analysis

The vectorization architecture, Bag-of-words(BOW) has been introduced for text classification problem. Count vectorizer, the simplest way of text vectorization method has been used in this project.

banglalm-dataset

basic-calculator-using-python

basics-of-neural-networks-using-tensorflow

The project is all about the basics of Neural Networks. Here I have implemented the model MLP for Binary Classification for creating a deep learning model. The project has two portions. At the first place I have focused on implementing a basic neural network model rather than concentrating on improving accuracy results. Lastly, I have worked on improving the accuracy result tuning the hyperparameters. Moreover, the model prediction results are described through model evaluation matrix, the confusion matrix.

certificates

cheatsheets

classification-of-fruits-using-knn-algorithm

K-NN Classification algorithm has been used here to predict fruits . Dataset has been collected

computer-vision-basics

cv-image-nanda

data-and-business-analytics

Excel practice project for business analyst

data-scraping-and-automation-

Documentations for basic understanding on Data scraping / Web Scraping from different sources.

data-visualization-

Visualization reports, dashboards created by Google Sheets, Google Studio, Power BI, Tableau.

diabetes-prediction-with-machine-learning-models

The dataset is a real dataset on Pima Indian diabetes data that consist of several medical predictor (independent) variables and one target (dependent) variable, Outcome. Independent variables include the number of pregnancies the patient has had, their BMI, insulin level,glucose level,level of Bloodpressure, skin thickness age, and so on. To predict Diabetes on diagonstic, I have used several models such as Logistic Regression, KNN Classifier, Decision Tree Classifier, Random Forest Classifier, Support Vector Classifier and Gradient Boosting Classifier. Moreover, the important part of building any machine learning model is Exploratory Data analysis(EDA) that has been described in this project shortly.

fake-news-classification-system

This project is about Fake News Classification problem. Several methods such as text vectorization methods, Bag-of-words, TFIDF has been used. Furthermore, a deep learning approach has been used with different RNN architectures (LSTM, Bidirectional LSTM) for classification. Dataset : https://www.kaggle.com/c/fake-news/data

fasttext-library-

This repository is all about the basic understanding on the applications of fastText library for Natural Language Processing problems. I have used several datasets to solve different problems such as Finding Semantic Similarities, Spelling Corrector/ word suggestion, understanding document distances, Classification for sentiment analysis and so on.

firstfrontendwithangular

This is a first project with angular for frontend web development.

gaussian-naive-bayes-algorithm-on-iris-dataset

I have used Gaussian Naive Bayes Model on Iris dataset that can predict the flower species based on the selected feature values. Moreover, Scikit-learn library has been used here to train the model as well as Seaborn has been used for visualization through pair plot and heatmap.

geospatial-data-analysis-using-zomato-restaurants-data

The project is about EDA with Geographic analysis and Interactive plots. Here I have analyzed data to find out area based restaurants, if they are top rated or if online delivery is available or not. In addition analyzed the restaurant ratings, most served food items, food items in a specific budget and so on. Moreover, learnt the uses of the libraries, seaborn, plotly, folium, iplot, geopy etc.

grouping-data-using-pandas-library

handwritten-digit-classification-using-cnn-architecture

For Image classification problem, the most common dataset MNIST dataset has been used here, which is a set of 70000 small images of hand written digits. The model has been trained with CNN Architecture and the project signifies the basic understanding of CNN Architecture.

java-project

Tution management system with java swing and MySQL.

json-parsing

leetcode-problems-

machine-learning-basics

Fundamentals of different ML Algorithms and their implementations.

ml-dl-model-templates-

FOR OWN WORK PURPOSE

nlp-basics

Basic understanding on Textual data preprocessing methods(Tokenization, Removing Stop words, Stemming & Lemmatization ) and feature engineering methods (Bag of Words, TF-IDF, Word Embedding).