experiments
Experiments done for hands-on learning with dummy data
Science Score: 67.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
✓DOI references
Found 2 DOI reference(s) in README -
✓Academic publication links
Links to: arxiv.org -
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (10.3%) to scientific vocabulary
Last synced: 10 months ago
·
JSON representation
·
Repository
Experiments done for hands-on learning with dummy data
Basic Info
- Host: GitHub
- Owner: atalv
- License: gpl-3.0
- Language: Jupyter Notebook
- Default Branch: main
- Size: 5.64 MB
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Created about 3 years ago
· Last pushed over 2 years ago
Metadata Files
Readme
License
Citation
README.md
Overview
This repo is to store the experiments done for hands-on learning with dummy data. It is never too late to learn and showcase!
- Each sub-directory in the root is named as the main topic of the experiment.
- All the contents are created solely by me with guidance from official resources and academic experts.
Citation
If you use any of this work then please add a referrence to this repository 'Experiments by Vivek Atal' as a fair usage policy.
Some highlighs
GraphNetwork:
- Predicted whether a user of LastFM would follow another user and serve as a recommendation.
- Implemented multiple node embedding approaches for link prediction - Graph Factorization, DeepWalk, Node2Vec, Adamic-Adar index - and compared their performance for link prediction task.
- Referred to excellent materials by Stanford CS224W course on Machine Learning with Graphs.
MachineLearning:
- Predicted NYC taxi trip duration.
- Implemented typical machine learning models from scikit-learn (GammaRegressor, RandomForestRegressor, HistGradientBoostingRegressor) with intelligently derived features, viz., traffic information in an area at a given time window based on average active number of trips originating or ending.
ReinforcementLearning:
- Simulated multiple UCB (Upper Confidence Bound) policies for MAB (Multi Armed Bandit) problems and MDP (Markov Decision Process) and compared their performance.
- Learned to do simulation of multiple states Markov Chain and calculate average reward, expected present value, estimate steady state probabilities, etc.
- Most of the research papers referred for simulation exercises are authored by Dr. Michael Katehakis.
TimeSeries:
- Forecasted 2 weeks ahead grocery store sales of 33 product groups across 54 stores, approx. 1.8K time series.
- Engineered multiple sensible features, viz., cross-store, cross-product elements, algorithmically short-listed important events for a given store-product, etc.
- Some Seasonal ARIMA models were built manually, and then scaled it using ARIMA where seasonal components were extracted beforehand for faster execution.
- Experimented with DeepAR on AWS Sagemaker to build a single global model instead of 1.8K ARIMA models.
Owner
- Name: Vivek Atal
- Login: atalv
- Kind: user
- Repositories: 3
- Profile: https://github.com/atalv
Citation (CITATION.cff)
# This CITATION.cff file was generated with cffinit.
# Visit https://bit.ly/cffinit to generate yours today!
cff-version: 1.2.0
title: Experiments by Vivek
message: >-
If you utilize this work, please cite it using the
metadata from this file
type: software
authors:
- given-names: Vivek
family-names: Atal
email: atalvivek@yahoo.co.in
orcid: 'https://orcid.org/0000-0002-9948-7458'
license: GPL-3.0