This repository contains implementation code for important research papers, starter guides for common deep learning tools, and explorations of various algorithms and techniques in machine learning, deep learning, and related fields. All relevant notebooks can be found in the nbs folder.

1. Fundamentals
2. Computer Vision
3. Natural Language Processing
4. Audio Processing
5. Machine Learning Algorithms
6. Deep Learning Architectures
7. Embeddings and Similarity Search
8. Model Fine-tuning
9. Multimodal Learning
10. Data Preprocessing and Augmentation
11. Mathematics for Machine Learning
12. Algorithms
13. Tool-specific Guides
14. Miscellaneous Topics

• Matrix Multiplication: Implementation and optimization of matrix multiplication operations.
• Backpropagation: Step-by-step implementation of the backpropagation algorithm.
• Mini-batch Training: Techniques for efficient mini-batch training in deep learning.
• Datasets: Handling and preprocessing various types of datasets.
• Foundations: Core concepts and building blocks of deep learning.
• Tensor Puzzlers: Exploration of tensor operations and manipulations.

• YOLO Object Detection and Tracking: Implementation of YOLO for real-time object detection and tracking.
• Vision Transformer (ViT): Exploration of the Vision Transformer architecture for image classification.
• CLIP Model: Implementation and usage of OpenAI's CLIP (Contrastive Language-Image Pre-training) model.
• Zero-shot Image Classification and Object Detection: Techniques for classifying images and detecting objects without prior training.
• Image Search using Color Histograms: Basic image retrieval system using color histogram features.
• Bag of Visual Words: Implementation of the Bag of Visual Words technique for image classification.
• ImageNet: Working with the ImageNet dataset.

• Transformers: Implementation and analysis of the Transformer architecture.
• GPT Architecture and Pretraining: Exploration of GPT (Generative Pre-trained Transformer) models and pretraining techniques.
• Sentence Embeddings: Various techniques for generating sentence embeddings.
• Cross-encoders: Implementation and usage of cross-encoders for text pair classification and ranking.
• Question Answering Systems: Building end-to-end question answering systems.
• NLP Data Augmentation: Techniques for augmenting text data to improve model performance.
• Attention Mechanism: Detailed exploration of attention mechanisms in neural networks.
• Mamba Implementation: Implementation and analysis of the Mamba architecture.

• Speech Recognition: Implementing and fine-tuning speech recognition models.
• Text-to-Speech: Building text-to-speech systems using deep learning.
• Speech-to-Speech Translation: End-to-end speech translation systems.
• Voice Assistant: Creating a basic voice assistant using deep learning techniques.
• Audio Classification: Classifying audio samples into predefined categories.
• Audio Generation: Generating audio using deep learning models.
• Audio Preprocessing: Techniques for preparing audio data for machine learning tasks.
• Audio Exploration: Analyzing and visualizing audio data.
• Spectrograms: Conversion between audio and spectrogram representations.

• Mean Shift Clustering: Implementation of the Mean Shift clustering algorithm.
• Principal Component Analysis (PCA): Dimensionality reduction using PCA.
• Linear Discriminant Analysis (LDA): Implementation of LDA for classification and dimensionality reduction.
• Kernel PCA: Non-linear dimensionality reduction using Kernel PCA.
• Support Vector Machines: Implementation and analysis of SVMs for classification.
• Logistic Regression: Building and training logistic regression models.
• Collaborative Filtering: Implementing collaborative filtering for recommendation systems.
• Perceptron and Adaline: Implementation of basic neural network models.

• Convolutional Neural Networks: Implementation and analysis of CNNs for image-related tasks.
• Recurrent Neural Networks: Exploring RNNs for sequence modeling tasks.
• Autoencoders: Implementation of various autoencoder architectures, including Variational Autoencoders (VAEs).
• Vector Quantized Variational Autoencoders (VQ-VAE): Exploring VQ-VAE for discrete representation learning.

• FAISS: Using Facebook AI Similarity Search (FAISS) for efficient similarity search and clustering of dense vectors.
• Locality Sensitive Hashing: Implementing LSH for approximate nearest neighbor search.
• Product Quantization: Exploring product quantization for compact encoding of high-dimensional vectors.
• Hierarchical Navigable Small World (HNSW): Implementation of HNSW for efficient approximate nearest neighbor search.

• QLoRA: Implementing Quantized Low-Rank Adaptation for efficiently fine-tuning large language models.
• LoRA: Exploring Low-Rank Adaptation for fine-tuning large models.
• Fine-tuning Sentence Transformers: Techniques for fine-tuning sentence transformer models on specific tasks.

• CLIP Model: Exploring OpenAI's CLIP model for joint vision and language understanding.
• Auto Video Captioning System: Building a system that automatically generates captions for video content.
• Video Data Handling: Techniques for processing and analyzing video data.

• Data Cleaning: Techniques for cleaning and preprocessing raw data.
• Feature Scaling: Implementing various feature scaling methods.
• NLP Data Augmentation: Exploring techniques for augmenting text data to improve model performance.

• Matrix Decomposition: Implementing and analyzing various matrix decomposition techniques.
• Fourier Transform: Exploring the Fourier transform and its applications in signal processing and machine learning.

• Multiprocessing: Techniques for parallel processing in Python.

• RAG System: Implementation of a Retrieval-Augmented Generation system.
• Tabular Data Processing: Techniques for handling and analyzing tabular data.

This project is licensed under the Apache-2.0 License.