What is this?
A curated set of short, standalone PyTorch example projects covering computer vision, text, and reinforcement learning, each meant to be emulated in your own work.
A curated set of short, standalone PyTorch example projects covering computer vision, text, and reinforcement learning, each meant to be emulated in your own work.Is it a library I install, or something else?Neither — it is a repository of independent example scripts, not an installable package. Each one is short, needs few or no extra dependencies, and serves as a template to copy into your own code.
Neither — it is a repository of independent example scripts, not an installable package. Each one is short, needs few or no extra dependencies, and serves as a template to copy into your own code.How do I get one of the examples running?Clone the repo, make sure PyTorch is installed, and run the script inside the example folder you want with Python. Every example is self-contained in its own directory.
Clone the repo, make sure PyTorch is installed, and run the script inside the example folder you want with Python. Every example is self-contained in its own directory.Do any examples need more than PyTorch itself?A few: the SNLI example uses torchtext with GloVe vectors, and the reinforcement learning examples run on OpenAI Gym. Otherwise the collection is deliberately kept dependency-light.
A few: the SNLI example uses torchtext with GloVe vectors, and the reinforcement learning examples run on OpenAI Gym. Otherwise the collection is deliberately kept dependency-light.If I'm new to PyTorch, is this the right place to learn it?It is a complement, not a course: the README points to separate tutorials, production recipes, a model hub, and a Q&A forum, leaving this repo purely to example code.
It is a complement, not a course: the README points to separate tutorials, production recipes, a model hub, and a Q&A forum, leaving this repo purely to example code.Does it cover distributed training?Yes — there are examples using Distributed Data Parallel and RPC, plus Hogwild-style training of shared ConvNets across multiple processes on MNIST.
Yes — there are examples using Distributed Data Parallel and RPC, plus Hogwild-style training of shared ConvNets across multiple processes on MNIST.What is the Forward-Forward example about?It is an image classification example trained with the Forward-Forward algorithm, sitting alongside the repo's conventional convnet and transformer examples.
It is an image classification example trained with the Forward-Forward algorithm, sitting alongside the repo's conventional convnet and transformer examples.Is there anything for time-series data?Yes — a time sequence prediction example where an LSTM learns to reproduce sine waves.
Yes — a time sequence prediction example where an LSTM learns to reproduce sine waves.What are similar projects?DeepLearning, leedl-tutorial, train-llm-from-scratch, AI-For-Beginners, TensorFlow-Examples, SolarWM, DeepLearning-500-questions, awesome-deep-learning
DeepLearning, leedl-tutorial, train-llm-from-scratch, AI-For-Beginners, TensorFlow-Examples, SolarWM, DeepLearning-500-questions, awesome-deep-learningWhich topics is it in?gan, cnn, model-training, tutorials, distributed-training, examples, reinforcement-learning, neural-network, transformer, nlp, bsd, generative-ai
gan, cnn, model-training, tutorials, distributed-training, examples, reinforcement-learning, neural-network, transformer, nlp, bsd, generative-ai