Deep Learning
Neural networks.
Contents
Courses
A free five-weekend plan to self-learners to learn the basics of deep-learning architectures like CNNs, LSTMs, RNNs, VAEs, GANs, DQN, A3C and more (2019)
A free deep reinforcement learning course by OpenAI (2019)
Google AI
Jeremy Howard - Fast.ai
a 3-6 month Udacity nanodegree, spanning multiple courses (2018)
Google AI
by Hakan Cebeci
A great introductory course on Deep Learning by Udacity and Facebook AI
Kaggle's free course on Deep Learning
Papers
Tutorials
Tutorials
Researchers
Websites
Textual QA corpus from CNN and DailyMail. More than 300K documents in total. Paper for reference.
Open Images is a dataset of ~9 million URLs to images that have been annotated with labels spanning over 6000 categories.
MNIST like fashion product dataset consisting of a training set of 60,000 examples and a test set of 10,000 examples. Each example is a 28x28 grayscale image, associated with a label from 10 classes.
Contains about 10 million news articles classified using opensources.co types
15488 visible-infrared paired images (30976 images) for low-light vision research, Project_Page
Over over 5 million images from 5 different domains for multi-source ocr/text recognition DA research, Project_Page
Sberbank released ~90,000 Russian QA pairs
https://research.facebook.com/tomas-mikolov
Frameworks
Application-oriented deep reinforcement learning framework addressing real-world decision problems.
Tools
Easy-to-use library to boost deep learning inference leveraging multiple deep learning compilers.
Visualizer for deep learning and machine learning models
TensorFlow's Visualization Toolkit
Debugging and visualization for deep learning
All-in-one web-based IDE for machine learning and data science.
A little logger for machine learning research. Log any object to the console, CSVs, TensorBoard, text log files, and more with just one call to logger.log()
Deep learning training platform with integrated support for distributed training, hyperparameter tuning, smart GPU scheduling, experiment tracking, and a model registry.
Fastest unstructured dataset management for TensorFlow/PyTorch by activeloop.ai. Stream & version-control data. Converts large data into single numpy-like array on the cloud, accessible on any machine.
Miscellaneous
Easy to install and use deep learning Faster R-CNN face detection for images and video in a docker container.
Curated list of articles related to deep learning scientific research applied to music
Curated list of articles related to deep learning scientific research on graph structured data at the graph level.
Curated list of articles related to deep learning scientific research on graph structured data at the node level.
contains examples, utilities and best practices for building recommendation systems. Implementations of several state-of-the-art algorithms are provided for self-study and customization in your own applications.
Keras Implementation of Ladder Network for Semi-Supervised Learning
Roadmap to becoming an Artificial Intelligence Expert
Table of Contents
Magenta is a project by Google exploring the role of machine learning in the process of creating art and music.
The CaffeNet model, a popular deep learning model for image classification. This is a PyTorch implementation.
Keras is a simple, modular, and extensible high-level Python API for deep learning.
fastai is a library for deep learning research and production. It's built on top of PyTorch.
Microsoft Cognitive Toolkit (CNTK) is an open-source deep-learning toolkit.
Open Neural Network Exchange (ONNX) is an open format built to represent machine learning models.
State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch.
A toolkit for developing and comparing reinforcement learning agents.
Models and examples built with TensorFlow.
JAX is an alpha library for high performance machine learning research.
Detectron2 is FAIR's next-generation library for object detection and segmentation.
YOLOv5 is a state-of-the-art, real-time object detection system.
PyTorch Image Models (timm) is a collection of pre-trained image classification models.
DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.
NeMo is a conversational AI toolkit built from the ground up for researchers and developers to create and experiment with the world's leading pre-trained conversational AI models.
An open source machine learning framework for everyone
OpenMMLab detection toolbox and benchmark
OpenMMLab semantic segmentation toolbox and benchmark
High-resolution image synthesis with latent diffusion models
Port of Facebook's LLaMA model in C/C++
Tensor library for machine learning
An open implementation of the OpenAI CLIP model
A Gradio web UI for Stable Diffusion.
High-Resolution Image Synthesis with Latent Diffusion Models
InvokeAI is a leading creative engine for Python, providing tools for AI-powered creativity.
CLIP (Contrastive Language–Image Pre-training)
A small, high-quality language model for reasoning and language understanding.
A sequence modeling toolkit.
Detectron is a tool for object detection and segmentation.
Nvdiffrast is a library for differentiable rasterization.
Efficient Geometry-aware 3D Generative Adversarial Networks
StyleGAN3 is an efficient and high-fidelity generative adversarial network for image synthesis.
StyleGAN2 is an efficient and high-fidelity generative adversarial network for image synthesis.
StyleGAN is a generative adversarial network for image synthesis.
Google's Generative Adversarial Networks implementation.
A PyTorch Video library for video understanding research.
FastText is a library for efficient learning of word representations and sentence classification.
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
OpenAI's GPT-2 model implementation.
OpenAI's GPT-3 model implementation.
Raw audio sequence to sequence models.
The official PyTorch implementation of the wav2vec 2.0 paper.
A PyTorch implementation of BERT for pre-trained models.
Text-to-Text Transfer Transformer
OpenAI's summarization models.
A large-coverage multilingual neural machine translation model.
Google's Seq2Seq models.
A fast and easy-to-use neural style transfer implementation.
Neural style transfer implementation.
A deepfake implementation.