Welcome to the Resources repository! This repository contains tutorial notebooks on various Deep Learning (DL) and Natural Language Processing (NLP) topics.
- Intro to Deep Learning: Comparison between Simple Model and CNN Model Architecture
- Notebook: Basics of Deep Learning Classification
- Description: This notebook provides an introduction to deep learning and compares the architecture of a simple model with a convolutional neural network (CNN) model.
- CNN: Multi-class Classification using CNNs and Explained Architecture
- Notebook: CNN MultiClass Classification
- Description: In this notebook, you will learn how to perform multi-class classification using convolutional neural networks (CNNs). The notebook also includes a detailed explanation of the CNN architecture used.
- ResNet: Multi-class Classification using ResNet (Pre-trained) and Explained Architecture
- Notebook: ResNet MultiClass Classification
- Description: This notebook demonstrates multi-class classification using a pre-trained ResNet model. It also provides an explanation of the ResNet architecture used.
- Natural Language Processing Basics
- Notebook: Hospital Comments EDA
- Description: This notebook focuses on exploratory data analysis (EDA) of a private dataset containing hospital comments. It includes preprocessing techniques such as data cleaning and text normalization. Additionally, it explores topics within the comments using Latent Dirichlet Allocation (LDA) and identifies named entities using Named Entity Recognition (NER) techniques.
- Generative Adversarial Networks
- Notebook: Introduction to GAN (Basic GAN and DCGAN Tutorial)
- Description: This notebook serves as an introduction to Generative Adversarial Networks (GANs). It provides a basic GAN tutorial, explaining the fundamental concepts and architecture. It also covers Deep Convolutional GANs (DCGANs), which are a variation of GANs designed for image generation tasks.
- Unsupervised Sentiment Analysis using Word2Vec and BERT
- Notebook: Sentiment Analysis (Clustering + Word2Vec Approach + BERT Approach)
- Description: This notebook focuses on sentiment analysis using different approaches. It covers clustering-based sentiment analysis techniques, leveraging unsupervised learning algorithms. It also explores the Word2Vec approach, which captures word embeddings to analyze sentiment. Lastly, it incorporates the BERT (Bidirectional Encoder Representations from Transformers) approach, a state-of-the-art model for sentiment analysis.
- Sentiment Analysis Using LSTM , RNN and Bidirectional LSTM
- Notebook: Movie Reviews Sentiment Analysis (Pre-processing + LSTM + RNN + BiLSTM Comparison)
- Description: This notebook specifically addresses sentiment analysis of movie reviews. It covers various steps of the analysis pipeline, including data preprocessing techniques. It then compares the performance of different recurrent neural network (RNN) architectures, such as Long Short-Term Memory (LSTM) and Bidirectional LSTM (BiLSTM), to determine their effectiveness in sentiment analysis tasks.