adaptdl
determined
Our great sponsors
adaptdl | determined | |
---|---|---|
4 | 10 | |
395 | 2,861 | |
0.0% | 3.8% | |
0.0 | 9.9 | |
about 1 year ago | 3 days ago | |
Python | Go | |
Apache License 2.0 | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
adaptdl
-
Introduction to PyTorch
AdaptDL
-
[Discussion] Open source scheduler and queuing system for model training/inferencing tasks?
check out https://github.com/petuum/adaptdl. It natively supports AWS/EKS (for the autoscaling feature), otherwise it can run anywhere on Kubernetes.
-
Reduce cost by 3x in the cloud and improve GPU usage in shared clusters with AdaptDL for PyTorch
AdaptDL monitors training job performance in real-time, and elastically re-scales resources (GPUs, compute instances) while jobs are running. For each training job, AdaptDL automatically tunes the batch size, learning rate, and gradient accumulation. In the cloud (e.g. AWS), AdaptDL can auto-scale the number of provisioned Spot Instances. We’ve seen shared-cluster training jobs at Petuum and our partners complete 2–3x faster on average, with 3x cheaper cost in AWS using Spot Instances!
-
How we were able to achieve hyper-parameter tuning (HPT) for deep learning workflows at 1.5x faster in our clusters and 3x cheaper on AWS
To tackle the problem of long and expensive HPT workflows, our team at Petuum collaborated with Microsoft to integrate AdaptDL with Neural Network Intelligence (NNI). AdaptDL is an open-source tool in the CASL (Composable, Automatic, and Scalable Learning) ecosystem. AdaptDL offers adaptive resource management for distributed clusters, and reduces the cost of deep learning workloads ranging from a few training/tuning trials to thousands. NNI from the Microsoft open-source community, is a toolkit for automatic machine learning (AutoML) and hyper-parameter tuning.
determined
-
Open Source Advent Fun Wraps Up!
17. Determined AI | Github | tutorial
-
ML Experiments Management with Git
Use Determined if you want a nice UI https://github.com/determined-ai/determined#readme
- Determined: Deep Learning Training Platform
-
Queueing/Resource Management Solutions for Self Hosted Workstation?
I looked up and found [Determined Platform](determined.ai), tho it looks a very young project that I don't know if it's reliable enough.
-
Ask HN: Who is hiring? (June 2022)
- Developer Support Engineer (~1/3 client facing, triaging feature requests and bug reports, etc; 2/3 debugging/troubleshooting)
We are developing enterprise grade artificial intelligence products/services for AI engineering teams and fortune 500 companies and need more software devs to fill the increasing demand.
Find out more at https://determined.ai/. If AI piques your curiosity or you want to interface with highly skilled engineers in the community, apply within (search "determined ai" at careers.hpe.com and drop me a message at asnell AT hpe PERIOD com).
-
How to train large deep learning models as a startup
Check out Determined https://github.com/determined-ai/determined to help manage this kind of work at scale: Determined leverages Horovod under the hood, automatically manages cloud resources and can get you up on spot instances, T4's, etc. and will work on your local cluster as well. Gives you additional features like experiment management, scheduling, profiling, model registry, advanced hyperparameter tuning, etc.
Full disclosure: I'm a founder of the project.
-
[D] managing compute for long running ML training jobs
These are some of the problems we are trying to solve with the Determined training platform. Determined can be run with or without k8s - the k8s version inherits some of the scheduling problems of k8s, but the non-k8s version uses a custom gang scheduler designed for large scale ML training. Determined offers a priority scheduler that allows smaller jobs to run while being able to schedule a large distributed job whenever you need, by setting a higher priority.
-
Cerebras’ New Monster AI Chip Adds 1.4T Transistors
Ah I see - I think we're pretty much on the same page in terms of timetables. Although if you include TPU, I think it's fair to say that custom accelerators are already a moderate success.
Updated my profile. I've been working on DL training platforms and distributed training benchmarking for a bit so I've gotten a nice view into the GPU/TPU battle.
Shameless plug: you should check out the open-source training platform we are building, Determined[1]. One of the goals is to take our hard-earned expertise on training infrastructure and build a tool where people don't need to have that infrastructure expertise. We don't support TPUs, partially because a lack of demand/TPU availability, and partially because our PyTorch TPU experiments were so unimpressive.
[1] GH: https://github.com/determined-ai/determined, Slack: https://join.slack.com/t/determined-community/shared_invite/...
-
[D] Software stack to replicate Azure ML / Google Auto ML on premise
Take a look at Determined https://github.com/determined-ai/determined
-
AWS open source news and updates No.41
determined is an open-source deep learning training platform that makes building models fast and easy. This project provides a CloudFormation template to bootstrap you into AWS and then has a number of tutorials covering how to manage your data, train and then deploy inference endpoints. If you are looking to explore more open source machine learning projects, then check this one out.
What are some alternatives?
HandyRL - HandyRL is a handy and simple framework based on Python and PyTorch for distributed reinforcement learning that is applicable to your own environments.
ColossalAI - Making large AI models cheaper, faster and more accessible
alpa - Training and serving large-scale neural networks with auto parallelization.
Dagger.jl - A framework for out-of-core and parallel execution
FedML - FEDML - The unified and scalable ML library for large-scale distributed training, model serving, and federated learning. FEDML Launch, a cross-cloud scheduler, further enables running any AI jobs on any GPU cloud or on-premise cluster. Built on this library, FEDML Nexus AI (https://fedml.ai) is your generative AI platform at scale.
aws-virtual-gpu-device-plugin - AWS virtual gpu device plugin provides capability to use smaller virtual gpus for your machine learning inference workloads
tutorials - PyTorch tutorials.
cfn-diagram - CLI tool to visualise CloudFormation/SAM/CDK stacks as visjs networks, draw.io or ascii-art diagrams.
PyTorch-NLP - Basic Utilities for PyTorch Natural Language Processing (NLP)
goofys - a high-performance, POSIX-ish Amazon S3 file system written in Go
torchlambda - Lightweight tool to deploy PyTorch models to AWS Lambda