collection of resources to study and learn ML & AI
fast.ai
- Deep learning 1: practical
- Deep learning 2: cutting edge
- Computational linear algebra
- Teaching philosophy and approach
Self-paced, good level of depth.
Based on jupyter books and AWS high-spec dedicated instances
Learn Machine Learning in 3 months
https://github.com/llSourcell/Learn_Machine_Learning_in_3_Months
- Fundamentals: calculus, algebra, probability, algorithms
- Python and machine learning tools
- Deep learning
Informal, compendium of resources and playlists
Mostly self-learning, free style
AI in a week
https://www.oxfordinsights.com/aiinaweek
7 days of insights, ideas, concepts, predictions for AI from academia
NVIDIA Deep Learning Institute
https://www.nvidia.com/en-us/deep-learning-ai/education/
Online, affordable (some free), hands-on training resources
General
- Fundamentals of Deep Learning for Computer Vision
- Deep Learning Workflows with TensorFlow, MXNet, and NVIDIA Docker
- Image Segmentation with TensorFlow
- Linear Classification with TensorFlow
Industry specific
- Image Creation Using GANs with TensorFlow and DIGITS
- Deployment for Intelligent Video Analytics using TensorRT
- Medical Image Analysis with R and MXNet
Additional resources: https://www.nvidia.com/en-us/deep-learning-ai/developer/
☑️ Udacity's Intro to ML
Free, Self-contained and hands-on with scikit-learn
- Naive Bayes, SVM, Decision Trees
- Regression, Outliers, Clustering, Feature scaling
- Text Learning, Feature Selection, PCA
- Validation, Evaluation Metrics
ML from scratch
Extensive library full of ML algo implementations with comprehension in mind, not efficiency
some interesting examples:
- Classification With CNN
- Density-Based Clustering
- Generating Handwritten Digits
- Deep Reinforcement Learning
- Image Reconstruction With RBM
great simple implementations Supervised, unsupervised, reinforcement & Deep Learning algos
https://github.com/eriklindernoren/ML-From-Scratch
Explained.ai
- The Matrix Calculus You Need For Deep Learning: Just the right amount of math you need for DL
Tools
Scikit-learn Machine Learning
^
|
+
+--> Keras <--+ Deep Learning
| |
+ +
Theano Tensorflow GPU
^ ^
| |
+--+ cuDNN +---+ CUDA