lohith0501 Goto Github PK

adaboost-and-hyperparameter-tuning

This effort was to understand how boosting technique performs with Decision tree of depth 2 as a weak learner.

analytics-and-automation-app-using-streamlit-for-avihs-technologies

This is a web app created using streamlit library in python. This app helps to fast track analytics of day to day tasks. Even automation engines such as auto email and auto messaging are created to ease the daily process.

ann-with-hyperparameter-tuning

Churn analysis with applying droput regularization and tuning hyper parameters for better results.

apache-kafka-course

This repository is for the Apache Kafka Course

banking-data-analytics-using-tableau

This effort was to perform data analytics by creating a beautiful business dashboard in Tableau

bike-_sharing_count_prediction_deep-learning

This project is to find prediction of bike sharing count using neural network. Here neural network is built from scratch.There are easier ways through Pytorch,tensorflow or Keras but this attempt was to have a better understanding on building neural nets from scratch.

bootstrap

The most popular HTML, CSS, and JavaScript framework for developing responsive, mobile first projects on the web.

bootstrap-sampling

This effort was to check how bootstrap sampling can work on smaller data set for Regression model

bouldin-index-formula-coding-using-python

The Davies–Bouldin index (DBI) is a metric for evaluating clustering algorithms.[1] This is an internal evaluation scheme, where the validation of how well the clustering has been done is made using quantities and features inherent to the dataset.

cat-dog-identification-using-cnn-computer-vision

This is very simple cat/dog picture identification using CNN coded with Keras API. This is a simple CNN architecture

comparison-of-randomforest-gausian-nb-multinomialnb-

This was an effort to compare the decision boundaries formed by generative classifier models during BITS Pilani Mtech

comparison-of-smoothing-models-in-time-series

This effort was made to better understand how Simple exponential, double exponential & triple exponential smoothing along with additive and multiplicative models vary over a training data set. This helps us to finalize the model for further forecasting.

computer-vision-capabilities

Contains all my basic learning codes that i have practiced to understand technicalities of cv.

data-app

This is an app with simple work flow in data analytics. This will be deployed in Heroku and further similar ideology can be used to create an end2end product with ML or DL algorithms.

data-collector-using-flask-and-python-anywhere-cloud

This effort was made to collect data from market for AVIHS technologies. These surveys will be automatically stored in csv file in python anywhere cloud as and when person submits the survey. Further this data set was used for machine leaning models.

data-transformation

Data transformation is the process of converting data from one format or structure into another format or structure. It is a fundamental aspect of most data integration and data management tasks such as data wrangling, data warehousing, data integration and application integration.

data-visualization-using-matplotlib-seaborn

A casual guide to use matplotlib and seaborn library for data analysis

data_visualization

Python_Tutorials_and_Courses

dendrogram-analysis-in-decision-trees

This effort was made to plot the dendrogram and check how decision trees will look after applying decision tree algorithm

detection-of-spam-ham-using-linear-svm-classification

Classification of Spam or ham using linear SVM algorithm. Slack variable is tuned using Gridsearchcv technique.

discretisation-or-binning

Discretisation is the process of transforming continuous variables into discrete variables by creating a set of contiguous intervals that span the range of the variable's values. Discretisation is also called binning, where bin is an alternative name for interval.

dlab-hackathon

Dlab Hackathon on zomato rating prediction which we won 5th place

document_similarity_using_lsi_approach_nlp-

Similarity of documents using Latent semantic indexing is used to get similarities between the documents. This similarity is obtained by comparing keywords with that of BOW dictionary.

face-keypoint-detection-using-cnn-pytorch-computer-vision

This is the project which i completed in Udacity nano degree program. In this project i have created a CNN with 4 layers, maxpooling, dropouts as regularization. This core model helps to detect face and key points/ key areas on the face like eyes,mouth,ears,nose etc. This project helped to recognize any humans in between other objects. I have used Pytorch with Adam optimizer MSE technique to complete the model.

farolcovid

🚦🏥. Ferramenta de monitoramento do risco de colapso no sistema de saúde em municípios brasileiros com a Covid-19 • Monitoring tool & simulation of the risk of collapse in Brazilian municipalities' health system due to Covid-19

feature-encoding

Convert categorical variables to multiple variables based on label encoding techniques

feature-scaling

Feature Scaling is a technique to standardize the independent features present in the data in a fixed range. It is performed during the data pre-processing.

fine-tuning-bert

gausian-normality-test

Gausian/Normal form is very important for all models. This is how we can check for time series data set with hypothesis testing

getting-started-with-streamlit-for-data-science