heatdrop.ai

A grab-bag of classic AI coursework—neural nets, genetic algorithms, adversarial attacks—sitting in Jupyter notebooks waiting for the curious.

Six standalone deep learning implementations for developers who'd rather read code than prose.

A from-scratch walkthrough that treats backprop through time as a computation graph problem, not a math ritual.

A 2016 tutorial repo that still runs on TensorFlow 2.x, comparing how feedforward, convolutional, and recurrent networks handle city noise.

A curated collection of paper summaries that saves you from reading the original ResNet paper at 2 AM.

Side-by-side exercises for developers who think in ndarrays and need to learn TensorFlow's dialect.

A reading list that tracks how to compress models and cram them onto FPGAs/ASICs without the marketing fluff.

Someone finally organized the firehose of NLP research into a single, obsessively maintained list.

A readable, extensible re-implementation of the Differentiable Neural Computer for researchers who want to poke at the memory gates, not just cite the paper.

A dead-simple TensorFlow implementation that trades bleeding-edge complexity for actual comprehension.

A grab-bag of reference implementations for classic deep-learning papers, aimed at learners who already finished the official tutorial.

A Korean-language talk and its code examples that predate TensorFlow's official debugger, yet remain weirdly relevant for anyone still untangling graph execution mysteries.

Complete lecture notes and working assignment solutions for CS 231n, the course that launched a thousand computer vision careers.

Someone is manually tracking the entire AI tooling landscape so you don't have to.

An old-school awesome-list that catalogs books, papers, courses, and tools for text mining and NLP.

A monthly ranked list of the top 10 machine learning articles, because nobody has time to skim 1,500 papers.

A collection of Jupyter notebooks teaching deep learning with TensorFlow's imperative mode, frozen in time at version 1.7.

A TensorFlow implementation of IndRNN, where each recurrent neuron carries its own weight and its own baggage.

A Microsoft Researcher's decade of ML lecture slides, paper summaries, and cheat sheets, living in a Jekyll repo.

Someone scraped the free HTML of the canonical deep learning textbook and glued it into a single PDF, then asked you to buy the paper copy anyway.