You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,
I was looking at your snippet and I was wondering if you have experimented with other kind of data.
As far as I can see this DALI library (I've been trying to find examples) is intented to be used for specific kind of data, right?
Pytorch dataset is flexible as you can code almost everything. Is it possible such flexibility here? is it restricted to premade image/audio/video loader?
As far as I can see it's only possible to use the system as dataloader. Is it possible to make use only of GPU accelerated video/image decoding tools in a normal pytorch dataset class?
The text was updated successfully, but these errors were encountered:
This example is purely for image loading. DALI itself has added support for many other datatypes, such as audio. I suggest you check out the official docs, there's some examples in there now for other datatypes
Hi,
I was looking at your snippet and I was wondering if you have experimented with other kind of data.
As far as I can see this DALI library (I've been trying to find examples) is intented to be used for specific kind of data, right?
Pytorch dataset is flexible as you can code almost everything. Is it possible such flexibility here? is it restricted to premade image/audio/video loader?
As far as I can see it's only possible to use the system as dataloader. Is it possible to make use only of GPU accelerated video/image decoding tools in a normal pytorch dataset class?
The text was updated successfully, but these errors were encountered: