114 lines
6.8 KiB
HTML
114 lines
6.8 KiB
HTML
{% extends "base.html" %}
|
||
|
||
{% block title %}
|
||
{{ title }}
|
||
{% endblock %}
|
||
|
||
{% block content %}
|
||
|
||
<!-- Primary Page Layout
|
||
–––––––––––––––––––––––––––––––––––––––––––––––––– -->
|
||
<div class="container">
|
||
<!-- <section class="header">
|
||
|
||
</section> -->
|
||
|
||
<section>
|
||
|
||
<h2>Projects</h2>
|
||
|
||
|
||
<div class="row item">
|
||
<a class="item-screenshot-wrapper" target="_blank" href="https://activerecaller.com">
|
||
<img class="item-screenshot" src="static/public/images/active_recaller.png" style="object-fit:fill;">
|
||
</a>
|
||
<div class="one-half offset-by-one-half column">
|
||
<h6 class="item-header">A web app for studying better powered by AI and MERN stack</h6>
|
||
<p class="item-project-summary">
|
||
Recently, I have been developing a web app for active recall learning, You can Upload a file, or CSV, or manually generate you own decks, to help you study better. I'm going to make it more powerful by adding more features and integrations with other tools. You can upload your book,
|
||
it will automatically make it into chunks, you can use an Agent from OpenAI, Gemini, or Claude to generate the decks. cards have 3 difficulty levels, and you can review them in a spaced repetition manner based on that difficulty.
|
||
</p>
|
||
<p class="docs-subheader">Productivity Project <span class="date">Oct 2025</span></p>
|
||
|
||
<a class="button" href="https://activerecaller.com">activerecaller.com</a>
|
||
</div>
|
||
</div>
|
||
|
||
<div class="row item">
|
||
<a class="item-screenshot-wrapper" target="_blank" href="/projects/run-yolo">
|
||
<img class="item-screenshot" src="static/public/images/repair.png" style="object-fit:fill;">
|
||
</a>
|
||
<div class="one-half offset-by-one-half column">
|
||
<h6 class="item-header">Yolov8 on Repair Dataset</h6>
|
||
<p class="item-project-summary"> The main goal of <a href="https://www.repairproject.eu/">RePAIR project</a> is to develop a ground-breaking technology to virtually eliminate one of the most labour intensive and frustrating steps in archaeological research, namely the physical reconstruction of shattered artworks. Indeed, countless vases, amphoras, frescos and other ancient artefacts, all over the world, have not survived intact and were dug out from excavation sites as large collections of fragments, many of which are damaged, worn out or missing altogether.</p>
|
||
<p class="docs-subheader">Introduction to Machine Learning <span class="date">Sep 2024</span></p>
|
||
|
||
<a class="button" href="/projects/run-yolo">Run</a>
|
||
<a class="button" href="https://github.com/saeedkhosravi94/resume" target="_blank">Source</a>
|
||
</div>
|
||
</div>
|
||
|
||
|
||
<div class="row item">
|
||
<a class="item-screenshot-wrapper" target="_blank" href="https://n8n.saeedkhosravi.it">
|
||
<img class="item-screenshot" src="static/public/images/article_automation.png" style="object-fit:fill;">
|
||
</a>
|
||
<div class="one-half offset-by-one-half column">
|
||
<h6 class="item-header">Article Reading Automation</h6>
|
||
<p class="item-project-summary"> I started to read new articles on a daily basis. I made an automation using n8n, Google Gemini, Postgres, and a website using Fast-API that made this more productive, and time efficient.
|
||
I would appreciate if you check it out and give me feedback, to make it better.
|
||
|
||
If you are interested in using n8n automation, contact me to launch one for you. ;)
|
||
</p>
|
||
<p class="docs-subheader">Productivity Project <span class="date">Oct 2025</span></p>
|
||
|
||
<a class="button" href="https://art.saeedkhosravi.it">Visit</a>
|
||
<a class="button" href="static/public/files/Computer Vision Articles Summary.json" download>Download n8n file</a>
|
||
</div>
|
||
</div>
|
||
|
||
|
||
<div class="row item">
|
||
<a class="item-screenshot-wrapper" target="_blank" href="/projects/run-models">
|
||
<img class="item-screenshot" src="static/public/images/S3VM_vs_NUTSVM.png">
|
||
</a>
|
||
<div class="one-half offset-by-one-half column">
|
||
<h6 class="item-header">Semi-Supervised SVM vs Newton Universum Twin SVM</h6>
|
||
<p class="item-project-summary"> In this project, we implemented two models: the Semi-Supervised SVM (S3VM) and the Newton-based Universum Twin SVM (Newton-UTSVM). S3VM strengthens learning with unlabeled data, while Newton-UTSVM improves generalization using Universum data. After comparing their performance, we propose a new method—the Unconstrained S3VM—that combines the advantages of both approaches for a more flexible solution. </p>
|
||
<p class="docs-subheader">Introduction to Artificial Intelligence <span class="date">Apr 2023</span></p>
|
||
|
||
<a class="button" href="static/public/files/S3VM_vs_NUTSVM.pdf" target="_blank">Report</a>
|
||
<a class="button" href="/projects/run-models">Run</a>
|
||
<a class="button" href="https://github.com/saeedkhosravi94/resume" target="_blank">Source</a>
|
||
</div>
|
||
</div>
|
||
<div class="row item">
|
||
<a class="item-screenshot-wrapper" target="_blank" href="/projects/run-models">
|
||
<img class="item-screenshot" src="static/public/images/ndcc.jpg">
|
||
</a>
|
||
<div class="one-half offset-by-one-half column">
|
||
<h6 class="item-header">Multi-Class Normally Distributed Cluster Centers Data Generator</h6>
|
||
<p class="item-project-summary"> NORMALLY DISTRIBUTED CUBIC CLUSTERS is a data generator. It generates a series of random centers for multivariate normal distributions. NDC randomly generates a fraction of data for each center, i.e. what fraction of data points will come from this center. NDC randomly generates a separating plane. Based on this plane, classes for are chosen for each center. NDC then randomly generates the points from the distributions. NDC can increase inseparability by increasing variances of distributions. A measure of "true" separability is obtained by looking at how many points end up on the wrong side of the separating plane. All values are taken as integers for simplicity.</p>
|
||
<p class="docs-subheader"><a href="https://scholar.google.com/citations?user=4aZwjNUAAAAJ&hl=en">Hossein Moosaei</a>, Saeed Khosravi, <a href="https://www.cs.carleton.edu/faculty/dmusicant/">Dave Musicant</a> <span class="date">Sep 2021</span></p>
|
||
|
||
<a class="button" href="/projects/generate-ndcc">Run</a>
|
||
<a class="button" href="https://github.com/saeedkhosravi94/resume" target="_blank">Source</a>
|
||
</div>
|
||
|
||
</div>
|
||
|
||
|
||
</section>
|
||
|
||
|
||
|
||
|
||
</div>
|
||
|
||
|
||
|
||
|
||
<!-- End Document
|
||
–––––––––––––––––––––––––––––––––––––––––––––––––– -->
|
||
|
||
{% endblock %} |