DEEP LEARNING TUTORIALS BY MURAT KARAKAYA AKADEMI

• Murat Karakaya Akademi Youtube
• Murat Karakaya Akademi Medium
• Murat Karakaya Akademi Kaggle
• Murat Karakaya Akademi GitHub Pages

TUTORIALS

I prepare the below tutorials to help you at solving Deep Learning problems with TensorFlow and Keras. You can also watch all the tutorials in English or Turkish at the Murat Karakaya Akademi Youtube channel.

I hope you find these tutorials helpful and useful.

Enjoy!

Murat Karakaya

GOOGLE GEMINI

• A COMPREHENSIVE INTRODUCTION TO GEMINI API FOR TEXT GENERATION & CHAT
• GOOGLE GEMINI: Upload Your Document Ask Questions Create Report For Free
• A SIMPLE RAG PIPELINE BASED ON GEMINI & CHROMADB

Retrieval Augmented Generation RAG

• PART A: A COMPREHENSIVE INTRODUCTION TO GEMINI API FOR TEXT GENERATION & CHAT
• PART B: CODE CHROMADB FOR VECTOR STORAGE & SIMILARITY SEARCH
• PART C: CODE WITH CHROMADB FOR PERSISTENT VECTOR DB
• PART D: A SIMPLE RAG PIPELINE BASED ON GEMINI & CHROMADB

HUGGING FACE TRANSFORMERS

• How to Train a Hugging Face Causal Language Model from Scratch with a Custom Dataset and a Custom Tokenizer?

SEQ2SEQ LEARNING

If you are interested in Seq2Seq Learning, I have a good news for you. Recently, I have been working on Seq2Seq Learning and I decided to prepare a series of tutorials about Seq2Seq Learning from a simple Multi-Layer Perceptron Neural Network model to Encoder Decoder Model with Attention.

• PART A: Introduction & A Simple NN Solution
• Part B: Using LSTM layer in a Recurrent Neural Network
• Part C: Basic Encoder Decoder Model
• Part D: Encoder Decoder with Teacher Forcing
• Part E: Encoder Decoder for Variable Input Output Size with Teacher Forcing
• Part F: Encoder Decoder with Bahdanau & Luong Attention Mechanism

TENSORFLOW DATA PIPELINE

In this series, we will focus on how to Build Efficient TensorFlow Input Pipelines in Deep Learning with Tensorflow & Keras. We will review the tf.data module. Using tf.data.Dataset methods, we will learn how to map, prefetch, cache, and batch the datasets correctly so that the data input pipeline will be efficient in terms of time and performance. We will discuss how map, prefetch, cache, and batch functions affect the performance of the tf.data.Dataset input pipeline performance.

Moreover, we will see how to use TensorBoard add-on “TF Profiler” for monitoring the performance and bottlenecks of the tf.data input pipeline.

• Build an Efficient TensorFlow Input Pipeline for Char-Level Text Generation
• Build an Efficient TensorFlow Input Pipeline for Word-Level Text Generation
• tf data: Build Efficient TensorFlow Input Pipelines for Image Datasets

TIME SERIES ANALYSIS

• A Time Series Analysis: Covid-19 Death Toll Prediction
• Anomaly Detection in Time Series

MODEL EVALUATION & METRICS

• How to Evaluate Models in Multi-Label Classification
• How to Evaluate a Classifier Trained with an Imbalanced Dataset? Why Accuracy is not Enough?

CONVOLUTIONAL NEURAL NETWORKS (CNN) & KERAS CONVOLUTIONAL LAYERS

• What are Convolution, Filters, and Feature Map? How Keras Conv2d Layer Works?
• Understand and Use Keras Conv1d Layer: Predict House Prices

CLASSIFICATION IN DEEP LEARNING WITH TENSORFLOW & KERAS

• How to solve Classification Problems
• How to solve Multi-Class Image Classification Problems
• How to solve Multi-Label Image Classification Problems
• How to solve Binary Image Classification Problems
• How to solve Binary Text Classification Problems
• Multi-Class Text Classification with a GPT3 Transformer block: An End-to-End Example
• Using Keras Text Vectorization Layer For a Multi-Class Text Classification
• How to Evaluate a Classifier Trained with an Imbalanced Dataset? Why Accuracy is not Enough?
• Multi Topic (Multi-Class) Text Classification With various Deep Learning Models

LSTM IN KERAS

• LSTM: Introduction
• LSTM: Understanding Output Types
• LSTM: Understanding the Number of Parameters
• Using LSTM layer in a Recurrent Neural Network Model
• Basic Encoder-Decoder Model with LSTM
• Character Level Text Generation with an LSTM Model

TEXT GENERATION WITH KERAS

In this series, we have been covering all the topics related to Text Generation with sample implementations in Python, Tensorflow & Keras.

• Text Generation: Fundementals
• Build an Efficient TensorFlow Input Pipeline for Char-Level Text Generation
• Build an Efficient TensorFlow Input Pipeline for Word-Level Text Generation
• Text Generation: Sampling
• Character Level Text Generation with an LSTM Model
• Character Level Text Generation with an Encoder-Decoder Model

CONTROLLABLE TEXT GENERATION WITH TRANSFORMERS

In this series, we have been covering all the topics related to Controllable (Conditioned) Text Generation with sample implementations in Python, Tensorflow & Keras.

• Fundamentals of Text Generation

WEB SCRAPING

• Web Scrapping: Car Price Prediction
• Web Scrapping: Find Your Dream Car

TensorFlow & Keras Layers

• Save and Load Keras Models with Custom Layers: TensorFlow SavedModel & HDF5 formats
• Keras Text Vectorization Layer: Configure, Adapt, Use, Save, Load, and Deploy to Hugging Face Spaces using Gradio

Large Language Models (LLMs)

• ZIPPING LLMs: Quantization & Off-loading
• GOOGLE GEMINI: Upload Your Document Ask Questions Create Report For Free
• A COMPREHENSIVE INTRODUCTION TO GEMINI API FOR TEXT GENERATION & CHAT