Amit Satpathy's repositories
Topic-Modeling-with-Automated-Determination-of-the-Number-of-Topics
My version of topic modelling using Latent Dirichlet Allocation (LDA) which finds the best number of topics for a set of documents using ldatuning package which comes with different metrics
Discriminative-Robust-Local-Binary-and-Ternary-Patterns
Contains the codes for Discriminative and Robust Local Binary Pattern and Discriminative and Robust Local Ternary Pattern for object recognition developed by me during my PhD studies.
Extended-Histogram-of-Gradients
Contains the codes for Extended Histogram of Gradients for object recognition developed by me during my PhD studies.
Knapsack-solution-for-Resource-Allocation
A solution for resource allocation based on limited constraints
Identifying-Commuter-Travel-Patterns-In-Bus-Services
A project I did with Land Transport Authority, a statutory board, whose main role is to manage the transportation infra of Singapore which includes public transport like bus and trains. The agency was interested to understand how the bus services were being utilized by commuters during peak hours and if interventions could be introduced to further enhance commuter experience on bus services e.g. shorter waiting time, faster trips with skipping of bus stops etc. This required understanding archetypes of travel patterns by commuters in bus services. This project is an extension of what was previously done here: https://blog.data.gov.sg/fingerprint-of-a-bus-route-73e5be53dcf0
Supervised-Classification-of-Text-Categories
This repo describes a supervised approach to text classification using different features and classifiers. This, obviously, is good to use if there is labelled data available.
AI-102-AIEngineer
Lab files for AI-102 - AI Engineer
annotator-app
This flask app is a simple personal project (done by Andrew Tan[@a-tanman] and myself) that is meant to make it easier to label/annotate text data for creating datasets for supervised machine learning.
Difference-of-Gaussian-Edge-Texture-Based-Background-Modeling-for-Dynamic-Traffic-Conditions
Note: This is a very old Visual 2005 project which has not been tested to work with later versions of Visual Studio. Use at your own risk.
gds-webinar-scale-demo
Demo for GDS 2022 February Webinar
gpt-2
Code for the paper "Language Models are Unsupervised Multitask Learners"
gpt-3
GPT-3: Language Models are Few-Shot Learners
graph-book
The Code Examples and Notebooks for The Practitioners Guide to Graph Data
interpret
Fit interpretable models. Explain blackbox machine learning.
LSTM-Sentiment-Analysis
Sentiment Analysis with LSTMs in Tensorflow
Machine-Learning-with-Python
Python codes for common Machine Learning Algorithms
ml-pipeline-generator-python
ML Pipeline Generator is a tool for generating end-to-end pipelines composed of GCP components so that any customer can easily migrate their local ML models onto GCP and start realizing the benefits of the cloud quickly.
mlflow-example
An example MLflow project
mlopsworld2020-kubeflow
workshop materials for the mlopsworld 2020
mlopsworld2020-mlflow
mlflow workshop material for mlopsworld 2020
mslearn-ai900
Lab files for AI-900: Azure AI Fundamentals
mslearn-dp100
Lab files for Azure Machine Learning exercises
notebooks-contrib
RAPIDS Community Notebooks
Optical-Flow-Histogram-Features-for-Anomaly-Movements-in-Video
This repository contains codes for a project I did where we detect hypermotor seizures in an unsupervised environment.
recommenders
Best Practices on Recommendation Systems
reinforcement-learning-an-introduction
Python Implementation of Reinforcement Learning: An Introduction
spark-rapids
Spark RAPIDS plugin - accelerate Apache Spark with GPUs