Deep-Learning | Deep-Reinforcement-Learning | NLP-IE-Pipeline | MLOPS-Real-World-ML-System | Knowledge-Graph | AI-Security-&-Privacy | Serverless-AI/ML | AI-Datasets | AutoML-&-HPO |
This repository is my humble endeavour to take a you through The Best of Enterprise AI by introducing you to very many constructs, design elements, real life AI applications, codesets, FAQs from the captivatingly beautiful landscape of AI. Gear up for the Universe called "AI". Bon Vaoyage!!
A Solemn Commitment:
I, The Captain of the Space Cruiser- DeepHiveMind, would continue to enrich this repository in my relentless persuit of democratizing the field of AI.
Welcome Onboard!
Your journey begins with **The wonderland of AI** wherein you get to relish the gigantic-yet-so-beautiful landscape of AI. Herein, you will be pampered with Infographic panoramic view of the design, the sweeping constructs of AI, and many more such delightful introductory elements. I'm sure, You'll relish the gateway to AI wonderland!
Post that, In this voyage, you will be treated with ultra-luxurious warmth of The REAL world of AI applications, wherein pleathora of actual codesets with varied flavours will welcome you in detail. Please cherish the tangy-yet-so-sweet flavour of AI application codesets! Then comes the next pitstop. Dress up for to be lavishly treated with amazing-awesome-powerful-&-yet-so-beautiful world of Deep Reinforcement Learming (DeepRL), Automated Machine Learning (AutoML), Productionizing of ML Model (MLOps. Yes, this is global AI shift! Let us together unleash the power of DeepRL, AutoML, MLOps, and transcend to the new era of AI!
Every destination of this tour is delicately curated to make your AI experience wonderful. This carefully-curated AI itinerary is planned by me based on my 10+ dedicated years of rich, deep hands-on engagement with various streams of "Enterprise AI, ML, NLP-Generation, Knowlede Graph, Converstaional AI, DL(Deep Learning) & DRL(Deep Reinforcement Learning)" - be it
- **Cognitive AI** , powered by Deep Neural networks, such as
- Computer vision,
- Natural Language Proessing, Understanding, Generataion (NLP-NLU-NLG),
- Audio & Speech analytics,
- Conversational AI,
- Multimodal analytics
- **classical Predictive machine learning**,
- **Strcutured Data AI/ML - Predictive ML** , powered by Deep Neural networks, such as
-Structured data classification from scratch
-Collaborative Filtering for Recommendations
-credit card fraud detection: Heavily Imbalanced classification
-Timeseries anomaly detection
- **AutoML**,
- AutoKeras (NAS with Bayesian Optimization)
- Google AutoML (NAS with Reinforcement Leaning)
- Hyper Parameter Optimization
- Randmoized/Grid Serach optimization
- **Knowledge Graph**,
- Ontology based on RDF/OWL subject-object-predicate semantic triples
- Apache JENA/ AWS NEPTUNE with SPARQL
- GRAKN with GRAQL
- Semantic lexicon for
- BioMedical ontology (used for Precision Medicine & Disease Network - BioGrakn)
- English language (used extensively by computational linguists and cognitive scientists - WordNet/ SynSet
- Law (legal ontologies for Law based knowledge formalization - JurWordNet/ DOLCE+)
- **Deep Reinforcement-learning** (self learning autonomous systems),
- **Distributed & Parallel ML**,
- **AI-as-a-Service**,
- RESTful API
- gRPC API
- **Custom MLOps** (the Next-Gen MLOPS),
- **Comprehensive AI GOVERNANCE**,
- 'AI TRUST'
- 'AI Collaboration, Sharing & Exchange'
- 'AI Security & Privacy'
- 'AI Scalability'
- 'AI Inference Model Update/ Roll out' Mechanism
- **SERVERLESS AI/ML**
and so on so forth for variuos industries such as ***Banking & FS, Insurance, Halthcare, Retail, Energy and utility.***
Please note:
- This page serve as the Index page.
- Please click on hyperlinks of the respective items to delve deep into it, i.e, Click to land at my other delicately handcrafted Repos/info-files for deep dive into respective items..
- Please keep checking your compass (this index page) for to seamlessly steer your way to the next milestone of this wonderful journey.
In this repository, sharing with you delicately hand crafted exquisite bouquet of offerings: Below is the list of hyperlinks to my other Repo/Readme recepies/notebook/codesets. Just Click on hyperlinks of the respective items to delve deep into it.
-
World of Popular Datasets for AI SOTA Models
- Audio SOTA Model Datasets
- Image SOTA Model Datasets
- Image Classification SOTA Model Datasets
- Image Object detection SOTA Model Datasets
- Visual Question answering SOTA Model Datasets
- NLP Text SOTA Model Datasets
- NLP Text Summarization SOTA Model Datasets
- NLP Text Translate NMT SOTA Model Datasets
- Video SOTA Model Datasets
-
World of Deep_NeuralNet AI_Applications Keras_TF2.0
-
-Image classification from scratch -Simple MNIST convnet -Image segmentation with a U-Net-like architecture -Next-frame prediction with Conv-LSTM -Grad-CAM class activation visualization -Model interpretability with Integrated Gradients -Metric learning for image similarity search -Point cloud classification with PointNet -Few-Shot learning with Reptile -Visualizing what convnets learn -
-Text classification from scratch -Sequence to sequence learning for performing number addition -Bidirectional LSTM on IMDB -Character-level recurrent sequence-to-sequence model -Using pre-trained word embeddings -Text classification with Transformer -BERT (from HuggingFace Transformers) for Text Extraction -
-Structured data classification from scratch -Collaborative Filtering for Movie Recommendations -Imbalanced classification: credit card fraud detection -
-Timeseries anomaly detection using an Autoencoder -
-Variational AutoEncoder -GAN overriding Model.train_step -WGAN-GP overriding Model.train_step -Neural style transfer -Deep Dream -Character-level text generation with LSTM -PixelCNN -Text Generation with miniature GPT
-
-
Deep NeuralNet AI Applications Additional Codesets
-Image Object Detection SSD -Image Segmentation UNET -Object Detection and Segmentation RCNN MaskRCNN -Skin Cancer Lesion detection from dermoscopic images -Automated Brain Tumour Segmentation -Scene Text detection from wild Images using CRNN CTPN DenseNet-OCR -Custom OCR Engine -Text detection from IDCard -Automated Number Plate Detection -Microservice Microweb framework for AIaaS -
World of Deep Reinforcement Learning Self evolving systems
-
Introduction to High level constructs of DeepRL
-Agent -Environment -Action -State -Reward -Policy -Value function -Function approximator -Markov decision process (MDP) -Dynamic programming (DP) -Monte Carlo methods -Temporal Difference (TD) algorithms -Model -
Introduction to Significant DeepRL algorithms
-Value Optimization Agents Algorithms -Deep Q Network (DQN) -Double Deep Q Network (DDQN) -Mixed Monte Carlo (MMC) -Policy Optimization Agents Algorithms -Policy Gradients (PG) -Asynchronous Advantage Actor-Critic (A3C) -Deep Deterministic Policy Gradients (DDPG) -Proximal Policy Optimization (PPO) -General Agents Algorithms -Direct Future Prediction (DFP) -
Introduction to Advanced DeepRL algorithms
- Imitation Learning Agents Algorithms -Behavioral Cloning (BC) -Conditional Imitation Learning - Hierarchical Reinforcement Learning Agents Algorithms -Hierarchical Actor Critic (HAC) - Memory Types Algorithms -Hindsight Experience Replay (HER) -Prioritized Experience Replay (PER) -
Comparison of Main DeepRL frameworks
Keras-RL (Developed by Matthias Plappert- Employed with OpenAI) OpenAI Gym Facebook Horizon Google Dopamine Google DeepMind TensorFlow Reinforcement Learning (TRFL) Coach (Developed by Intel Nervana Systems) RLLib (Highly customizable open source DeepRL framework with support for TF2.0/PyTorch 1.4, customization for Environments/Policy/Action) Tensorforce (Tensorforce is built on top of Google's TensorFlow framework version 2.0 by Alexander Kuhnle - currently with BluePrism) -
DeepRL HANDS-ON with Keras-RL & TF & OpenAI-Gym
Deep Q Learning (DQN) Double DQN Deep Deterministic Policy Gradient (DDPG) Continuous DQN (CDQN or NAF) Cross-Entropy Method (CEM) Dueling network DQN (Dueling DQN) Deep SARSA Asynchronous Advantage Actor-Critic (A3C) Proximal Policy Optimization Algorithms (PPO)
-
-
Detailed Insight into the constructs of AutoML ecosystems
- Automated Feature Engineering - [Expand Reduce] - [Hierarchical Organization of Transformations] - [Meta Learning] - [Reinforcement Learning] - Architecture Search - [Evolutionary Algorithms] - [Local Search] - [Meta Learning] - [Reinforcement Learning] - [Transfer Learning] - Hyperparameter Optimization - [Bayesian Optimization] - [Evolutionary Algorithms] - [Lipschitz Functions] - [Local Search] - [Meta Learning] - [Particle Swarm Optimization] - [Random Search] - [Transfer Learning]
-
Hyper Parameter Optimization (HPO) in Classical ML - Grid Serach/Random Search
-
Real World ML System Powered by MLOPS
- INSPIRATION for MLOPS
- MLOPS VS DATAOPS VS AIOPS VS PLATFORMOPS
- STATE OF MACHINE LEARNING OPERATIONS IN Y2019
- MLOPS Reference Architecture
- High level constructs of MLOPS
- Detail References to the Constructs and Tools of MLOPS
- MLOPS Architecture based on KUBEFLOW
- Train and Serve TensorFlow Models at Scale with KUBERNETES and KUBEFLOW
- Docker - Containerizatioin of AI Modules
- Kubernetes - K8S for AI
- Helm Charts on K8S for AI complex deployment
- Kubeflow for AI MLOPS
- JupyterHub on K8S- AI Build Notebook
- TFJob- K8S custom kind for AI Training on GPU/CPU/TPU
- Distributed Tensorflow
- Hyperparameters Sweep with Helm
- AI Model Serving for prediction/scoring
- Train and Serve TensorFlow Models at Scale with KUBERNETES and KUBEFLOW
Click below items for safe ladning on the planets of the respective items. These planets are delicately curated handcrafted my preferred list of tools. Takes to my other repos/info-files.
Hola! You have arrived at your destination. Hope you chrished the journey.
Hurray! You are now Black belt Data Scientist.
Sayonara!! Before, you take a leave, here is a quick recap of interesting journey through Infographics. No Words! Just the Visuals!
A quick recap of the some of the Cognitive models shared in this Repo
Please pay attention to the announcement:
- Some of these Models repos shared hereiwth me (DeepHiveMind) are delicately curated and handcrafted.
- There are many other Repos which I have shared in this project beyond what is listed out below. Takes to my other repos/info-files.
Click below items for happy ladning on & surfing some of the very many Cognitive models shared by in the DeepHiveMind repo.
During KubeCon 2017, Vicki Cheung and Jonas Schneider delivered a keynote explaining how OpenAI manage to handle training at very large scale with Kubernetes, it is worth listening to:
Please refer to Additional references for learning and career transitioning to the world of 'DL,RL,DRL,NLP,ML,OCR'
Feel free to contact me to discuss any issues, questions, or comments.
My contact info can be found on my GitHub page.
I, The DeepHiveMind, am providing code and resources in this repository to you under custom Copyright & license (Copyright 2019 DeepHiveMind & Creative Commons Legal Code CC0 1.0 Universal). Please Refer to the Copyright 2019 DeepHiveMind License for further details as to this. Thanks!