I think if you’re interested in doing it for more than a few months it would make sense to build your own server.
Any feedback on https://www.leadergpu.com/gpu_instances ? Any other serious options besides this and Google Cloud for GPU instances (for TF&Torch)?
-
-
-
Right now I only need a beefy instance for a couple hours to build a model
- 3 more replies
New conversation -
-
-
Awesome! Let’s see how these perform with Tensorflow!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Include the cuda toolkit, cuDNN and pycuda in the default image for these instances? These take forever to install.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
And that’s why I would just give up and use Google ML cloud at this point :/
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@awscloud also provides GPU based instances. -
Still a technology preview, and insanely expensive.
- 1 more reply
New conversation -
-
-
aws. mostly because I bid on short lived instances
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Azure N preview http://gpu.azure.com , Baidu (preview) via google translate https://cloud.baidu.com/product/gpu.html … and Softlayerhttp://www.softlayer.com/gpu
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.