Deep Learning Frameworks: Difference between revisions
Jump to navigation
Jump to search
| Line 6: | Line 6: | ||
=== Popular Deep Learning Framework === | === Popular Deep Learning Framework === | ||
{| class="wikitable" | {| class="wikitable sortable mw-collapsible" | ||
!Framework | !Framework | ||
! | !HPCMATE custom build<ref>HPCMATE supports custom build service to optimize binary performance on given system based on customer requirements </ref> | ||
!Developer | !Developer | ||
!Language | !Language | ||
| Line 19: | Line 19: | ||
|- | |- | ||
|TensorFlow | |TensorFlow | ||
| | |Yes | ||
|Google | |Google | ||
|Python | |Python | ||
| Line 30: | Line 30: | ||
|- | |- | ||
|PyTorch | |PyTorch | ||
|Pytorch version 2.0<ref>https://github.com/pytorch/pytorch/releases</ref> released as of 12/2/22 | |Yes | ||
Pytorch version 2.0<ref>https://github.com/pytorch/pytorch/releases</ref> released as of 12/2/22 | |||
|Facebook | |Facebook | ||
|Python | |Python | ||
Revision as of 12:37, 26 March 2023
There have been many Deep Learning (DL) frameworks, like Theano, CNTK, Caffe2, and MXNet. Nowadays, they appear to be dead or dying, as just two frameworks heavily dominate the DL scene: Google TensorFlow (TF), which includes Keras and PyTorch from Meta aka FaceBook.
However, there is no reason to believe such a duopoly will persist forever. All the time, new DL frameworks are proposed. We have no idea which DL framework will be popular in, say, ten years.
Google has at least two (perhaps more) competing AI teams: Google Brain and DeepMind. Even in the TensorFlow era, DeepMind used their own layer API called Sonnet (instead of the usual Keras)
Popular Deep Learning Framework
| Framework | HPCMATE custom build[1] | Developer | Language | GPU Support | Distributed Computing | Auto-differentiation | Pre-trained models | Visualization | Deployment |
|---|---|---|---|---|---|---|---|---|---|
| TensorFlow | Yes | Python | Yes | Yes | Yes | Yes | Yes | Yes | |
| PyTorch | Yes
Pytorch version 2.0[2] released as of 12/2/22 |
Python | Yes | Yes | Yes | Yes | Yes | Yes | |
| Keras | Python | Yes | Yes | Yes | Yes | Yes | Yes | ||
| MXNet | Apache | Multiple | Yes | Yes | Yes | Yes | Yes | Yes | |
| Caffe | Berkeley AI Research | C++ | Yes | No | Yes | Yes | No | Yes | |
| Theano | Université de Montréal | Python | Yes | No | Yes | Yes | No | No | |
| Torch | Facebook AI Research | Lua | Yes | No | Yes | Yes | No | No |
Reference
<references/>
- ↑ HPCMATE supports custom build service to optimize binary performance on given system based on customer requirements
- ↑ https://github.com/pytorch/pytorch/releases