Madhur Devkota's Projects
Adopting Recurrent Neural Network, Long-Short Term Model to build Air pollution forecasting Deep Neural Network
Clustering for mixed-type data
Codes related to activities on AV including articles, hackathons and discussions.
Mastering Image Classification with Advanced CNN Techniques in Keras: A Deep Dive into Network Architectures and Performance Enhancement Strategies
Catboost model implementation
A small tutorial to demonstrate the power of CatBoost Algorithm
Building CNN
Credit Card Fraud Detection models using Machine Learning and Deep Learning on a PCA-transformed dataset. Addressing class imbalance, the models include ensemble methods and a Deep Neural Network (DNN), evaluated with standard metrics. Ensuring robust fraud detection through comprehensive preprocessing and balanced sampling.
basic to advanced datascience manipulation functions and techniques in pandas
Datascience notebook usage personal notes
DataViz-Python-Notebook
Python implementation of Density-Based Clustering Validation
neural network models
Andrew Ng深度学习Jupyter Notebook (Exercise notebooks for Deep Learning)
Multivariate time series forecasting that aims to predict the next hour's weather forecast based on the previous 24 hour window. Final deep learning project for CSC 578 (Neural Networks and Deep Learning). Base model consisted of a recurrent neural network (RNN) with a single LSTM layer, followed by hyperparameter tuning and variations in depth intended to achieve a closer fit to the data.
Deep Neural Network modeling for CivilEngineering problem
DeepLearning-Python-Notebook
GitHub Pages for Automating GIS-processes
Source files and notebooks for a paper on accelerating HDBSCAN*
🤖 Python examples of popular machine learning algorithms with interactive Jupyter demos and math being explained
experimenting with several imputation techniques
Create, Send, Transform JSON Data
The Language Interpretability Tool: Interactively analyze NLP models for model understanding in an extensible and framework agnostic interface.
various models developed using different machine learning algorithms
adopted multiple ML models using python libraries