dsrhaslab/prisma
A data prefetching storage data plane for accelerating DL training performance.
This tool helps machine learning engineers accelerate the training of deep learning models. It takes your existing training datasets and processes them to ensure data is available instantly, drastically reducing the time models spend waiting for data to load. The primary users are researchers and practitioners who build and train large-scale deep learning models and are looking to optimize their training infrastructure.
No commits in the last 6 months.
Use this if your deep learning model training is bottlenecked by the speed at which data is read from storage.
Not ideal if your deep learning training is slow due to computational complexity or model architecture rather than data loading.
Stars
8
Forks
—
Language
C++
License
—
Category
Last pushed
Jan 16, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/dsrhaslab/prisma"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
deepspeedai/DeepSpeed
DeepSpeed is a deep learning optimization library that makes distributed training and inference...
helmholtz-analytics/heat
Distributed tensors and Machine Learning framework with GPU and MPI acceleration in Python
hpcaitech/ColossalAI
Making large AI models cheaper, faster and more accessible
horovod/horovod
Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.
bsc-wdc/dislib
The Distributed Computing library for python implemented using PyCOMPSs programming model for HPC.