Jeremy covered that. There is nothing that is going to make your activations go high for the absence of classes, so you are making it a harder task for your model.
3 Likes
with binomial function whatās the limit to say there is / there isnāt a āthingā (say fish) in the image. is that 0.5?
1 Like
You will have to decide on this 
For instance, on the planet dataset, 0.2 worked rather well.
1 Like
thanks ā¦for the multiclass classification also we can have this approach ?assumption every image has to be one class
Did we already create nn.Conv2d last week?
1 Like
No, we didnāt. And I donāt think we will, Jeremy is cheating
8 Likes
Is it better to build a layer dictionary and pass that to nn.Sequential or list the layers directly?
1 Like
As you prefer, this is really a styling issue.
1 Like
Thanks
Is there ever any reason not to send things to CUDA?
2 Likes
GPU Memory Restrictions
1 Like
If you donāt have a GPU? More seriously, things like metrics or stored losses need to be put back on the CPU to avoid OOM errors.
2 Likes
If you want to test your model in inference mode to see how long it takes to run inference on a CPUā¦for implementation purposes.
2 Likes
Jeremy also mentioned that inferrence is often better done on CPU.
Why would that be the case?
Isnāt there acceleration that can be had if you have a small sliver of GPU? i.e. for the forward pass
How did he decide on the order=2 for that batch transform cb?
1 Like
I sometimes use CPU for inference. I donāt need the speed of the GPU.
havenāt checked fastai code recently but image augmentation is or can also be implemented as callback?
1 Like