Post has attachment
12 ways to deeply learn
Photo

Post has attachment

I recently read on an answer on StackOverflow that HSV color space is better suited for image classification tasks than RGB since it has a separate light channel which makes it less sensitive to light in the image.

Has anyone tried using HSV color scheme for image classification/regression
problems instead of RGB?

If I use HSV, for it to be lighting insensitive, should I just use the other two "HS" channels and skip the light channel "V"?

Thanks in advance :)

Post has attachment

Post has attachment

Post has attachment

Post has attachment
The first article in a series exploring Deep Learning in Computer Vision using PyTorch. In this, we set the goal of getting into the top 1% in a Kaggle Competition (you need to be in top 10% to be an expert):

http://theaijournal.com/2018/04/20/pytorch-for-computer-vision-1-introduction-to-computer-vision-and-deep-learning/

Post has attachment
Using already existing models in #ML/#DL libraries might be helpful in some cases. But to have better control and understanding, you should try to implement them yourself. This article shows how a #CNN is implemented just using #NumPy.
Code is available at #github: https://github.com/ahmedfgad/NumPyCNN #MachineLearning #DeepLearning #Python #NeuralNetwork #ANN #NN #ConvNet

Post has attachment
ver 0.16 astyleすぐにパラメータを忘れるのでシェルスクリプトa.shに。20180920

Post has attachment
An interesting TWIML&AI podcast interview of Jeff Dean, head of Google Brain, the Google’s deep learning research team. By the way, Jeff gave a keynote at TechAide AI conference, yesterday in Montréal. The pretty high scientific standing conference was organized by Hugo Larochelle, researcher at Google Brain Montréal, for the benefit of Centraide which fights against poverty. In addition to his time, Dr. Larochelle generously offered to double donations received for the event up to $ 25,000
Wait while more posts are being loaded