Catalyst is a PyTorch framework for Deep Learning Research and Development.
It focuses on reproducibility, rapid experimentation, and codebase reuse so you can create something new rather than write yet another train loop.
Break the cycle - use the Catalyst!
Deep learning – introduction, backpropagation algorithm. Optimization methods. Neural Network in numpy.
Regularization methods and deep learning frameworks. Pytorch basics & extras.
CNN. Model Zoo. Convolutional kernels. ResNet. Simple Noise Attack.
Object Detection. (One, Two)-Stage methods. Anchors. Image Segmentation. Up-scaling. FCN, U-net, FPN. DeepMask.
Metric Learning. Contrastive and Triplet Loss. Samplers. Cross Entropy Loss modifications. SphereFace, CosFace, ArcFace.
AutoEncoders. Denoise, Sparse, Variational. Generative Models. Autoregressive models.
Generative Adversarial Networks. VAE-GAN. AAE. Energy based model.
Embeddings. RNN. LSTM, GRU.
Attention Mechanism. Transformer Model.
Pretrained Transformers. BERT. GPT. Data Augmentation in Texts. Domain Adaptation.
Collaborative Filtering. FunkSVD. Neural Collaborative Filtering.
Reinforcement Learning. DQN Algorithm. DDPG Algorithm. Wolpertinger.
Research & Deploy. Config API. Reaction.
Pipeine for Image Super-Resolution task that based on a frequently cited paper, ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks (Wang Xintao et al.), published in 2018.
Brain image analysis with Catalyst
Fast R&D prototyping and Kaggle farming.
Bert Distillation with Catalyst.
Generative Adversarial Networks with Catalyst
Accelerated Python code formatter
Experiments logging & visualization
Convenient deep learning models serving
Image segmentation with Catalyst
Object detection with Catalyst
Image classification with Catalyst
Catalyst.RL: A Distributed Framework for Reproducible RL Research
Accelerated deep learning research and development