Is this “keep the last 10 or 20 approximate hessians”?
1 Like
Is there any work on pre-training networks for generic style learning?
3 Likes
Just thinking of how some of this semi-relates to Feature Visualization from distill.pub
2 Likes
Is @ matrix multiplication or dot product?
did we skip the part where the vgg model is actually trained on the painting? I feel like we skipped straight to extracting the layers of interest.
1 Like
This uses the pretrained VGG model. Probably no fine-tuning needed, since we’re using the features extracted by some layer in the middle of the network.
2 Likes
@ is matrix multiplication (new in python 3)
5 Likes
I am enjoying today’s lesson a lot 
1 Like
I’m really glad about this too…For anyone that might have missed it I actually started a thread a couple days back specifically for an open discussion on this subject in the forums so welcome to anyone that wants to join in or share insights there 
11 Likes
first came to know about it on your computational algebra course
2 Likes
Please ask him to talk about Multi-GPU
7 Likes
what was the name of the paper with style-transferring captain america’s shield?
It’s matrix multiplication. It’s a pure python operator. But dot product and matrix multiply are the same, no?
2 Likes
Deep painterly harmonization https://arxiv.org/abs/1804.03189
6 Likes
I CANNOT get enough of Jeremy’s frustration with Math notation. I would watch a friggin’ netflix special about that.
Can we make this happen?
13 Likes
Is there a name for the type of operation where you just multiply each pair without adding, like this
[[a,b],[c,d]] * [[e,f],[g,h]] = [ [a * e, b * f], [ c * g,d * h ] ]
There is a convention (especially loved by physicists) called Einstein summation. Numpy can do it: https://docs.scipy.org/doc/numpy/reference/generated/numpy.einsum.html
2 Likes
I guess it is Hadamard Product
3 Likes
element-wise multiplication. And yeah, that’s just * in pytorch/numpy/tensorflow
1 Like