Search Results for "throughput" - Page 3
Sort By:
Showing 80 open source projects for "throughput"
View related business solutions-
Stop Storing Third-Party Tokens in Your DatabaseRolling your own OAuth token storage can be a security liability. Token Vault securely stores access and refresh tokens from federated providers and handles exchange and renewal automatically. Connected accounts, refresh exchange, and privileged worker flows included.
-
Train ML Models With SQL You Already KnowBuild and deploy ML models using familiar SQL. Automate data prep with built-in Gemini. Query 1 TB and store 10 GB free monthly.
-
1
Hiera
A fast, powerful, and simple hierarchical vision transformer
Hiera is a hierarchical vision transformer designed to be fast, simple, and strong across image and video recognition tasks. The core idea is to use straightforward hierarchical attention with a minimal set of architectural “bells and whistles,” achieving competitive or superior accuracy while being markedly faster at inference and often faster to train. The repository provides installation options (from source or Torch Hub), a model zoo with pre-trained checkpoints, and code for evaluation... -
2
TensorHouse
A collection of reference Jupyter notebooks and demo AI/ML application
TensorHouse is a scalable reinforcement learning (RL) platform that focuses on high-throughput experience generation and distributed training. It is designed to efficiently train agents across multiple environments and compute resources. TensorHouse enables flexible experiment management, making it suitable for large-scale RL experiments in both research and applied settings. -
3
Punica
Serving multiple LoRA finetuned LLM as one
...The system includes specialized CUDA kernels that enable batched GPU operations across different LoRA models simultaneously. This design allows a single GPU cluster to host many task-specific models while maintaining high throughput and minimal latency. The architecture also includes scheduling mechanisms that coordinate requests from multiple tenants and distribute workloads efficiently across available resources. -
4
LLaMA-MoE
Building Mixture-of-Experts from LLaMA with Continual Pre-training
LLaMA-MoE is an open-source project that builds mixture-of-experts language models from LLaMA through expert partitioning and continual pre-training. The repository is centered on making MoE research more accessible by offering smaller and more affordable models with only about 3.0 to 3.5 billion activated parameters, which helps reduce deployment and experimentation costs. Its architecture works by splitting LLaMA feed-forward networks into sparse experts and adding gating mechanisms so... -
Forever Free Full-Stack Observability | Grafana CloudBuilt on open standards like Prometheus and OpenTelemetry, Grafana Cloud includes Kubernetes Monitoring, Application Observability, Incident Response, plus the AI-powered Grafana Assistant. Get started with our generous free tier today.
-
5
Medusa
Framework for Accelerating LLM Generation with Multiple Decoding Heads
Medusa is a framework aimed at accelerating the generation capabilities of Large Language Models (LLMs) by employing multiple decoding heads. This approach allows for parallel processing during text generation, significantly enhancing throughput and reducing response times. Medusa is designed to be simple to implement and integrates with existing LLM infrastructures, making it a practical solution for scaling LLM applications. -
6
FastViT
This repository contains the official implementation of research
FastViT is an efficient vision backbone family that blends convolutional inductive biases with transformer capacity to deliver strong accuracy at mobile and real-time inference budgets. Its design pursues a favorable latency-accuracy Pareto curve, targeting edge devices and server scenarios where throughput and tail latency matter. The models use lightweight attention and carefully engineered blocks to minimize token mixing costs while preserving representation power. Training and inference recipes highlight straightforward integration into common vision tasks such as classification, detection, and segmentation. The codebase provides reference implementations and checkpoints that make it easy to evaluate or fine-tune on downstream datasets. ... -
7
FFCV
Fast Forward Computer Vision (and other ML workloads!)
ffcv is a drop-in data loading system that dramatically increases data throughput in model training. From gridding to benchmarking to fast research iteration, there are many reasons to want faster model training. Below we present premade codebases for training on ImageNet and CIFAR, including both (a) extensible codebases and (b) numerous premade training configurations. -
8
OmicSelector
Feature selection and deep learning modeling for omic biomarker study
OmicSelector is an environment, Docker-based web application, and R package for biomarker signature selection (feature selection) from high-throughput experiments and others. It was initially developed for miRNA-seq (small RNA, smRNA-seq; hence the name was miRNAselector), RNA-seq and qPCR, but can be applied for every problem where numeric features should be selected to counteract overfitting of the models. Using our tool, you can choose features, like miRNAs, with the most significant diagnostic potential (based on the results of miRNA-seq, for validation in qPCR experiments). -
9
Apple Neural Engine (ANE) Transformers
Reference implementation of the Transformer architecture optimized
...The repository targets practitioners who want to keep familiar PyTorch modeling while preparing models for Core ML/ANE execution paths. Documentation highlights reported improvements in throughput and memory residency, while releases track incremental fixes and packaging updates. The project sits alongside related Apple ML repos that focus on deploying attention-based models efficiently to ANE-equipped hardware. In short, it’s a practical blueprint for adapting Transformers to Apple’s dedicated ML accelerator without rewriting entire model stacks. -
Gemini 3 and 200+ AI Models on One PlatformBuild, govern, and optimize agents and models with Gemini Enterprise Agent Platform.
-
10
DockStream
A Docking Wrapper to Enhance De Novo Molecular Design
...The flexilibity to specifiy a large variety of docking configurations allows tailored protocols for diverse end applications. DockStream can also parallelize docking across CPU cores, increasing throughput. DockStream is integrated with the de novo design platform, REINVENT, allowing one to incorporate docking into the generative process, thus providing the agent with 3D structural information. -
11
MToolBox
A bioinformatics pipeline to analyze mtDNA from NGS data
MToolBox is a highly automated bioinformatics pipeline to reconstruct and analyze human mitochondrial DNA from high throughput sequencing data. MToolBox includes an updated computational strategy to assemble mitochondrial genomes from Whole Exome and/or Genome Sequencing (PMID: 22669646) and an improved fragment-classify tool (PMID:22139932) for haplogroup assignment, functional and prioritization analysis of mitochondrial variants. MToolBox provides pathogenicity scores, profiles of genome variability and disease-associations for mitochondrial variants. ... -
12
PerfKit Benchmarker
PerfKit Benchmarker (PKB) contains a set of benchmarks
PerfKitBenchmarker is an open-source benchmarking framework designed to measure and compare the performance of cloud infrastructure across multiple providers in a consistent and reproducible way. It allows users to evaluate metrics such as latency, throughput, provisioning time, and system performance using a standardized set of benchmarks. The tool supports a wide range of environments, including major cloud platforms, Kubernetes clusters, and even local hardware, making it highly versatile for performance analysis. It simplifies the process of running complex benchmarks by providing unified command-line workflows that handle resource provisioning, execution, and result collection. ... -
13
HyLiTE
Hybrid Lineage Transcriptome Explorer
HyLiTE (Hybrid Lineage Transcriptome Explorer) analyzes high-throughput transcriptome data from allopolyploid species. Allopolyploidy describes the formation of a new hybrid organism from the union of two or more different parents. Allopolyploid species carry multiple copies of each gene (homeologs), which often exhibit unusual expression patterns. Homeolog expression levels can technically be determined from next generation sequencing data (RNA-seq), but in practice, assigning reads to one homeolog over another is extremely challenging, particularly on a whole-genome scale. ... -
14
PyTorch-BigGraph
Generate embeddings from large-scale graph-structured data
...PBG supports multi-relation graphs (knowledge graphs) with relation-specific scoring functions, negative sampling strategies, and typed entities, making it suitable for link prediction and retrieval. Its training loop is built for throughput: asynchronous I/O, memory-mapped tensors, and lock-free updates keep GPUs and CPUs fed even at extreme scale. The toolkit includes evaluation metrics and export tools so learned embeddings can be used in downstream nearest-neighbor search, recommendation, or analytics. In practice, PBG’s design lets practitioners train high-quality graph embeddings. -
15
Scalable Distributed Deep-RL
A TensorFlow implementation of Scalable Distributed Deep-RL
Scalable Agent is the open implementation of IMPALA (Importance Weighted Actor-Learner Architectures), a highly scalable distributed reinforcement learning framework developed by Google DeepMind. IMPALA introduced a new paradigm for efficiently training agents across large-scale environments by decoupling acting and learning processes. In this architecture, multiple actor processes interact with their environments in parallel to collect trajectories, which are then asynchronously sent to a... -
16
D-Tailor
D-Tailor: automated analysis and design of DNA sequences
Recent advances in DNA cloning and synthesis technologies afford high throughput implementation of designed sequences into living cells. However, our ability to design sequences to interrogate multifactorial biological processes and further engineer biological functions is lagging behind. DNA-Tailor (D-Tailor) is a fully extendable software framework for biological sequence analysis and multi-objective sequence design. -
17
gain
Asyncio-based Python framework for building fast web crawling spiders
Gain is a Python web crawling framework designed to simplify the process of building efficient and scalable web scrapers. It is built on top of asynchronous technologies such as asyncio, aiohttp, and uvloop to support high-performance crawling with concurrent network requests. It provides a structured framework for creating spiders that can navigate websites, extract structured data, and process the collected results. Developers define crawlers using components such as spiders, parsers, and... -
18
SnowyOwl
RNA-Seq based gene prediction pipeline for fungal genomes
...Sensitivity is gained by repeatedly running the HMM gene predictor Augustus with varied input parameters, and selectivity by choosing the models with best homology to known proteins and best agreement to the RNA-Seq data. SnowyOwl has successfully predicted genes in 26 novel fungal genomes. The pipeline can be installed locally for high throughput and control over configuration. It can also be run on a remote server through a convenient web interface for occasional use. -
19
HTSeq
Analysing high-throughput sequencing data with Python
This SourceForge page is outdated! HTSeq has been moved to github: https://github.com/simon-anders/htseq General information and documentation on HTSeq; http://www-huber.embl.de/users/anders/HTSeq -
20
vipR is a program to screen for sequence variants (SNPs, deletions) in sequence data generated by high-throughput-sequencing platforms. Information on this and other projects can be found on: http://www.altmann.eu
-
21
Mi-perf
Miperf is an automated enhancement of iperf tool
iperf is a tool that can create TCP&UDP data streams and measure the throughput of a network that is carrying them. basically to do this iperf testing, we need 2 machines to act as server and client. we run a command -s in server side and then launch a client connection to a server by typing a command -c in client side. before that we need to enter the webgui of the wifi AP to do some configurations such as, change the frequency mode and channel. so we need to repeat the same process/steps for next test cases.imagine if we have hundred test cases, how much time we have to spend for this testing. ... -
22
...However I will be happy to answer any queries or provide support and comments if you are interested in extending this algorithm. The method is fully described in Deshpande, V., Fung, E. D., Pham, S., & Bafna, V. (2013). Cerulean: A hybrid assembly using high throughput short and long reads. arXiv preprint arXiv:1307.7933. http://arxiv.org/abs/1307.7933
-
23
ParaSim
Parallelized calculation of molecular similarities
...ParaSim addresses this challenge by parallelizing the calculations according to the number of computing cores available on a single machine. It is optimized for the throughput of very large numbers of query structures against very large numbers of reference structures. As as special feature, ParaSim allows to store and and to access frequently queried datasets as persistant objects in memory for short response times. ParaSim calculates chemical similarities based on binary structural fingerprints. ... -
24
AMBIENT
Find active modules in metabolic networks using high-throughput data
IMPORTANT: Since publication of the AMBIENT method in BMC Sys Bio, several updates have been made. If you wish to use the version used in the paper it is v0.6.3, however I recommend using the latest version which works in the same way but with additional options and has stability and performance improvements. Thanks for your interest! AMBIENT (Active Modules for Bipartite Networks) is a Python module that uses simulated annealing to find areas of a metabolic network (modules) that have... -
25
colonyzer image analysis software
Image analysis estimating cell density in arrayed microbial cultures
...Specialises in detection of extremely low cell densities. Forms part of the Quantitative Fitness Analysis (QFA) workflow: http://research.ncl.ac.uk/qfa/ Suitable for high-throughput, genome-wide analysis of culture libraries when combined with the following qfa R package: http://qfa.r-forge.r-project.org/ This is the version presented, used and demonstrated in the following manuscripts: Lawless et al. 2010 http://dx.doi.org/10.1186/1471-2105-11-287 Addinall et al. 2011 http://dx.doi.org/10.1371/journal.pgen.1001362 Chang et al. 2011 http://dx.doi.org/10.1534/g3.111.000216 Banks et al. 2012 http://dx.doi.org/10.3791/4018 Development has been moved to github, where Colonyzer has undergone several recent improvements, particuarly making installation easier and analysis faster: https://github.com/CnrLwlss/Colonyzer
