Skip to main content

Posts

Showing posts from June, 2020

Testing Ubuntu 20 on old laptop

So I wanted to test Ubuntu 20 on old laptop I have and see if things will work. To my surprise, it does. It's not as swift as on new laptop, but considering the age (9 years?), it still works. There are no issues with drivers and everything works smoothly.
Ubuntu 20.04 running on Lenovo X220
The only recommendation I would have is to allow different scale for different displays, but this can be achieved with choosing different resolutions. Little confusing, but at the end it works.
One more thing - does it really need to use 1.3 GB of memory without running anything? For old systems this might be a little bit of issue.

Good phones for good price (399$)

It's nice that Apple & Google decided to make some lower end (price), but good phones:
Apple iPhone SE

Google Pixel 3a
Or if you want really cheap (99$), you can go for Nokia:
Nokia 3.1

Who would have thought after three months things will be still getting worse

Exact Match Convolution

So, I tried to test out exact match convolution, which can be defined as:
It consists from three parts: pixel offsetstandard convolutionconvolution biasIf you test it out with DeeperThought, with config:
convolutionsq,1,28,28,500,8,8,0.0,1.0,-0.001 max,500,21,21,21,21 matrix,500,200,0.1,-0.001 sigmoid,200 dropout,200,0.5 matrix,200,10,0.1,-0.001 softmax,10
where convolutionsq is exact match convolution (emc), you get pretty good results:


Top accuracy on test data: 98.68 % (Dataset - MNIST digits).
Not as good as with convolution (up to 99.41 % accuracy), but pretty close. Are we really just looking for exact matches and we do so by convolution? Hint: Linear Regression on a Set of Selected Templates from a Pool of Randomly Generated Templates.

When you find etalon of perfect singing

When you meet a curious squirrel (not a monkey)

In Toronto, quite normal In USA, they usually keep distance But not this one

DeeperThought now running on Windows

Deeper Thought can now run on windows. You only need NVidia CUDA & Visual Studio Community to compile it.

Open solution:
Build:
Run from command line:

Same for Wide Open Thoughts - now can run on Windows as well.