eth-easl / modyn
Modyn is a research-platform for training ML models on growing datasets.
☆46Updated last week
Alternatives and similar repositories for modyn:
Users that are interested in modyn are comparing it to the libraries listed below
- ML Input Data Processing as a Service. This repository contains the source code for Cachew (built on top of TensorFlow).☆37Updated 6 months ago
- Multi-Instance-GPU profiling tool☆57Updated last year
- A resilient distributed training framework☆89Updated 11 months ago
- ☆52Updated 9 months ago
- ☆35Updated 4 years ago
- ☆43Updated 3 years ago
- ☆44Updated 8 months ago
- Model-less Inference Serving☆85Updated last year
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆82Updated last year
- Stateful LLM Serving☆46Updated this week
- LLM Serving Performance Evaluation Harness☆70Updated 2 weeks ago
- Microsoft Collective Communication Library☆61Updated 3 months ago
- A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems☆148Updated 4 months ago
- ☆23Updated last year
- An experimental parallel training platform☆54Updated 11 months ago
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆112Updated last year
- ☆53Updated 4 years ago
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆148Updated 5 months ago
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆49Updated 2 years ago
- SOTA Learning-augmented Systems☆35Updated 2 years ago
- Surrogate-based Hyperparameter Tuning System☆28Updated last year
- ☆31Updated 2 years ago
- Dynamic resources changes for multi-dimensional parallelism training☆24Updated 4 months ago
- Efficient Interactive LLM Serving with Proxy Model-based Sequence Length Prediction | A tiny BERT model can tell you the verbosity of an …☆27Updated 9 months ago
- ☆24Updated last year
- Artifacts for our ASPLOS'23 paper ElasticFlow☆52Updated 10 months ago
- (NeurIPS 2022) Automatically finding good model-parallel strategies, especially for complex models and clusters.☆38Updated 2 years ago
- Ultra | Ultimate | Unified CCL☆47Updated 3 weeks ago
- FastFlow is a system that automatically detects CPU bottlenecks in deep learning training pipelines and resolves the bottlenecks with dat…☆26Updated last year
- A Cluster-Wide Model Manager to Accelerate DNN Training via Automated Training Warmup☆34Updated 2 years ago