Tensorpack

Tensorpack - train neural networks at scale | Awesome Repos

Features

Neural Network Training Frameworks - Provides a high-performance framework for building and training flexible neural network architectures.
Deep Learning Training Toolsets - Offers a specialized toolset for training deep neural networks with a focus on execution speed and flexibility.
Distributed Deep Learning Frameworks - Serves as a unified platform for distributed model training and deployment across GPU clusters.
Distributed Training - Scales training workloads across multiple GPUs and compute clusters using data-parallel strategies.
Data-Parallel Training - Distributes training workloads across multiple GPUs by replicating models and splitting input data.
Large Scale Dataset Processing - Processes massive datasets using parallel strategies to maximize throughput and minimize research bottlenecks.
Multi-Process Data Loading - Utilizes multi-process data loading and parallelization strategies to maximize throughput for large datasets.
Neural Network Training Toolkits - Provides a suite of high-efficiency execution patterns and tools for optimizing deep learning model development.
Execution Pattern Optimizations - Increases throughput and reduces overhead by implementing high-efficiency execution patterns instead of standard interfaces.
Training Data Prefetchers - Implements background threads to load and buffer training data batches, minimizing GPU stalling due to CPU bottlenecks.
Parallel Batch Loading Pipelines - Implements a concurrent reader-writer pipeline for loading and preprocessing large datasets during training.
Training Speed Optimizations - Reduces training overhead and increases throughput using high-efficiency execution patterns.
Parameter Synchronization - Coordinates model weights across distributed workers using synchronization primitives to ensure consistent updates.
Framework-Agnostic Pipelines - Provides a decoupled loading mechanism that handles massive datasets independently of the underlying deep learning framework.
Low-Overhead Training Interfaces - Provides a low-overhead interface that interacts directly with the execution engine to increase training speed and throughput.
Dataset Streaming Buffers - Provides buffered streaming of large-scale datasets from disk to memory to maintain high throughput while managing memory consumption.
Large Dataset Optimizations - Implements efficient data loading and preprocessing pipelines optimized for handling massive amounts of information.
Deep Learning - Interface for efficient neural network training.
Frameworks and Libraries - Toolbox focused on training speed and large datasets.

Open-source alternatives to Tensorpack

Similar open-source projects, ranked by how many features they share with Tensorpack.

liguodongiot/llm-action
liguodongiot/llm-action
23,169View on GitHub
This project is a comprehensive framework for the training, fine-tuning, and deployment of large language models. It functions as a distributed deep learning platform that enables users to scale model workflows across multiple hardware nodes while providing tools for model evaluation and performance benchmarking. The platform distinguishes itself by offering specialized utilities for model compression and weight transformation, allowing users to reduce memory footprints and latency through quantization and pruning. It supports the adaptation of large models for consumer-grade hardware, facili
HTMLllmllm-inferencellm-serving
View on GitHub23,169
microsoft/swin-transformer
microsoft/Swin-Transformer
15,715View on GitHub
Swin-Transformer is a deep learning framework designed for training and deploying hierarchical vision transformer models. It serves as a research library and toolkit for computer vision tasks, providing the infrastructure to build models that replace standard convolution operations with sliding window self-attention mechanisms. By utilizing a multi-scale feature hierarchy, the framework enables the processing of visual data at varying resolutions and spatial scales. The project distinguishes itself through its implementation of shifted window partitioning, which facilitates global information
Pythonade20kimage-classificationimagenet
View on GitHub15,715
pytorch/tutorials
pytorch/tutorials
9,202View on GitHub
The PyTorch Tutorials repository is a collection of educational resources that provides step-by-step guidance on building, training, and deploying neural networks using the PyTorch framework. It covers the complete machine learning workflow, from data loading and model definition through optimization loops and model persistence, with dedicated guides for distributed training, model fine-tuning, and deployment. The tutorials offer practical demonstrations of adapting pre-trained models to new tasks through transfer learning, scaling training across multiple GPUs or machines using PyTorch's dis
Python
View on GitHub9,202
gabrieleangeletti/deep-learning-tensorflow
gabrieleangeletti/Deep-Learning-TensorFlow
958View on GitHub
This project is a deep learning library and neural network training framework built for the TensorFlow ecosystem. It functions as a structured repository of algorithms and tools designed to execute iterative learning routines, fit complex datasets to predictive models, and manage the deployment of trained neural networks. The library provides a standardized interface for machine learning research prototyping, allowing users to experiment with various architectures and validate data models. It supports the full lifecycle of model development, from the initial training of neural networks on cus
Pythondeep-learningtensorflow
View on GitHub958

See all 30 alternatives to Tensorpack

ppwwyyxxtensorpack

Features

Open-source alternatives to Tensorpack

liguodongiot/llm-action

microsoft/Swin-Transformer

pytorch/tutorials

gabrieleangeletti/Deep-Learning-TensorFlow

Star history

Open-source alternatives to Tensorpack

liguodongiot/llm-action

microsoft/Swin-Transformer

pytorch/tutorials

gabrieleangeletti/Deep-Learning-TensorFlow