https://towardsdatascience.com/advances-in-few-shot-learning-a-guided-tour-36bc10a68b77 great brief summary over matching networks, prototypical networks, model-agnostic meta-learning(MAML). well, the first two topics are well explained but MAML section needs a lot more thinking to understand. Also, I think MAML is closer to the topic of meta-learning Read more…
key point compared to resnet, the residual blocks are upgraded to have multiple “paths” or as the paper puts it “cardinality” which can be treated as another model architecture design hyper parameter. resnext architectures that have sufficient cardinality shows improved Read more…
you can apply proxy settings to python requests like this:
instlal PyYAML in python file, load file like the following the data in yaml file will be saved in data variable as a python dict.
download latest poppler package from here. unzip the downloaded file, and obtain the path of ‘poppler-xx.xx.x/Library/bin’ directory. Add this path to windows environment variable PATH.
reference: https://www.urlencoder.io/python/
good simple explanation and figure: https://www.kaggle.com/residentmario/cosine-annealed-warm-restart-learning-schedulers
arxiv: https://arxiv.org/pdf/1905.11946.pdf key point propose ‘compound scaling method’ which scales all width/depth/resolution together which is an efficient scaling method that can be applied to any existing structure introduce a new family of baseline structure called ‘EfficientNets’. The very smallest baseline Read more…
https://arxiv.org/pdf/1611.08408.pdf key point use adversarial net to distinguish if input masks are from segmentation net or ground truth. this allows to add auxiliary loss, like below: where mce: multi class cross-entropy bce: binary cross entropy s(): segmentation network a(): adversarial Read more…