Daesoo Lee's Blog

Let's say we have the following three files: config.py, main.py and experiments/test.sh. config.py is in charge of (hyper-)parameters in main.py and we'd like to run main.py using test.sh. # config.py import argparse def load_args(): parser = argparse.ArgumentParser() parser.add_argument('--lr', default=1e-3, type=float) parser.add_argument('--h_layers', default=[2, 2, 2], type=lambda s: [int(it..

The overall information about AWS and how to create an instance can be found in my Naver blog. How to connect to an instance using Putty and WinSCP can be found here.

Binary Cross Entropy is used for multi-label classification, and it's involved with a sigmoid function. (Note that cross-entropy is used for multi-class classification that's involved with a softmax function) (reference) Example of a simple neural network model for multi-label classification in Keras (link) - Size of an output layer must be a number of labels. [PyTorch web] BCEWithLogitLoss (link)

AUC (Area Under the Curve) [1] developers.google.com/machine-learning/crash-course/classification/roc-and-auc?hl=ko The area under the ROC (receiver operating characteristic) curve. The ROC curve consists of TPR (True Positive Rate) and FPR (False Positive Rate) : $$ TPR = \frac{TP}{TP + FN} $$ $$ FPR = \frac{FP}{FP + TN} $$ The ROC curve can be drawn by adjusting a classification threshold from..