GDG Devfest'24 Notes

I’ve joined GDG devfest’24 event yesterday. There were remarkable speakers. I would like to add my notes. You may find the link of the conference here

ECDP '24 Poster

I’m excited to share that our poster is in 20th European Congress on Digital Pathology 2024 | ECDP2024! ✨

quantization-2

TL;DR? Read the full version on Medium Understanding quantization can feel like learning two different languages when comparing TensorFlow and PyTorch. In this post, I’ll guide you through the differences and similarities, providing clear explanations along the way. You can find all the code references and notes in the provided notebook

quantization-1

TL;DR? Read the full version on Medium Why do we need quantization?

• Shrink models to a small size.
• DL architectures are bigger and bigger.
• A model can have 70 billion parameters.
• NVIDIA T4 GPUs have 16 GB RAM.
• Running models are still a challenge.
• The aim is to get a smaller model.

whats-new-in-openai

TL;DR? Read the full version on Medium In this post, I will highlight some of the recent updates of OpenAI SDK.

Ogrenme_Teknikleri

TL;DR? Read the full version on Medium Uzun yıllar önce belirlediğim hayat felsefem hayat boyu öğrenci olmaktı. Şu an yapay zeka ile ilgileniyorum ve makinelerin nasıl öğrendiği üzerine oldukça kafa yoruyorum. Aynı zamanda, ben nasıl öğreniyorum diye uzun uzun düşünme fırsatım oldu. Kendi tecrübelerimden bazı anekdotlar aktarmak istedim. Umarım, yazının tamamında olmasa da bazı kısımlarında kendinizden parçalar bulursunuz :)

Temperature_parameter

Temperature is a parameter which is injected into the softmax function, enabling the users to manipulate the output probabilities. It helps us to control the creativeness of a Large Language Model.

LLM — Q&A

TL;DR? Read the full version on Medium If you’ve been curious about LLMs and still have questions, this article is just for you. Get ready to dive deep into the world of language models!

LLM — Finetuning

TL;DR? Read the full version on Medium Finetuning doesn’t have to be a mystery anymore! In this article, I’ve created a simple notebook that breaks down the process in an easy-to-understand way.

NER

TL;DR? Read the full version on Medium Build your NER data from scratch and learn the details of the NER model.

LLM — Evaluation

TL;DR? Read the full version on Medium This article is a practical guide to measure the model performance and to make sure that the model gives good results after fine-tuning.

LLM — Few-Shot Learning

TL;DR? Read the full version on Medium Explanation of Zero-Shot One-Shot and Few Shot on code. Click here to watch the video on YouTube

ChatGPT - Code Interpreter

TL;DR? Read the full version on Medium Code Interpreter of GPT-4 is available to all ChatGPT Plus users. You can make a quick analysis in just 5 minutes, this is a breakthrough in data science and data analysis!:)

LLM - Configurations

TL;DR? Read the full version on Medium What are the configuration parameters that can influence the model’s output during inference? Click here to watch the video on YouTube

LLM - Introduction

TL;DR? Read the full version on MediumIf you think that, now that ChatGPT handles everything and you are late to learn it, this article is for you:)

Cross-Entropy Loss

TL;DR? Read the full version on Medium I will explain the cross-entropy loss with a very simple example.

Easy OCR with Hugging Face & Streamlit

TL;DR? Read the full version on Medium You can start creating your apps on Hugging Face platform today!

K-means with Numpy

In this post, I will explain the K-means algorithm via Numpy. Continue reading by clicking here

Line_chart_with_plotly

In this file I will explain how to use plotly with markers and annotation. It is very easy to plot chart in plotly express!:)

Word_vectors

Anomaly_detection_with_numpy_v2

Anderson_test

Anomaly_detection_with_numpy

K_means_with_numpy

Audio Classifier with Edge Impulse

In this post you can find my second project on Edge Impulse with Arduino Tiny ML kit. I used audio dataset this time.

Motion Classifier with Edge Impulse

Hello! Happy new year everyone:) On the very first day of the year, I completed a motion classifier project with Edge Impulse & Arduino. I would like to tell you the steps of the project, some problems I’ve faced with and how to solve them.

Fast fourier tranform

How to Deploy Model to Arduino?

In this post, I will explain the deployment of edge impulse to Arduino.

Set Arduino Tiny ML Kit up in Edge Impulse

I’ve bought Arduino Tiny Machine Learning Kit for my tiny machine learning projects. As a start, I’m taking the lecture: Introduction to Embedded Machine Learning from Coursera. To set the device up to Edge impulse you need to install Edge Impulse CLI and Arduino CLI. Here you can find some tips so that you don’t deal with errors during the installation: Remember that those tips are valid for Windows OS.

Style Transfer with PyTorch

In this post, I would like to share my notes from the ‘Style Transfer’ lecture given by Udacity, Intro to Deep Learning with PyTorch. You may find the references below.

Image Loading and Transformation with Numpy & PyTorch & PIL

In this notebook, I will explain how to load an image with PIL and explain some operations with PyTorch and Numpy. These processes are part of Style Transfer in Convolutional Neural Networks, there are two images: content and style. The next post defines how to transfer the style of one image to the content of the other image.

Stanford cs231n: Optimization Algorithms in Neural Networks

In this post, Optimization Algorithms in Neural Network will be explained. Thanks to Stanford University, the lecture videos of cs231n can be found in YouTube and all materials are freely available (see the references).