PaDiM Several methods have been proposed to combine anomaly detection(Give anomaly score to images) and localization(Assign each pixel an anomaly score to output anomaly map) in a one-class learning setting(Whether an image is normal or not). However, either they require…
Anomaly Detection Nearest neighbor (kNN) methods utilizing deep pre-trained features exhibit very strong anomaly detection performance when applied to entire images. A limitation of kNN methods is the lack of segmentation map describing where the anomaly lies inside the image.…
Simple SGD vs Cyclic Learning Rate I compared the training speed between two optimizers by training a UNet Model. Simple SGD optimizer = torch.optim.SGD(model.parameters(), lr=0.01) Cyclic Learning Rate optimizer = torch.optim.SGD(model.parameters(), lr=0.01) scheduler = torch.optim.lr_scheduler.CyclicLR(optimizer, base_lr=0.1, max_lr=1e-4) Cyclic Learning Rate…
Learning Rate Range Test Learning rate may be the most important hyper-parameter in deep learning and you can use this test to find the right learning rate; Run your model and record accuracy/loss for several epochs while letting the learning…
Cyclical Learning Rate It is known that the learning rate is one of the most important hyper-parameters for training deep neural networks. Unlike previous methods such as monotonically decreasing the learning rate, the Cyclical learning rate practically eliminates the need…
Feature Pyramid Network Feature pyramids are a basic component for detecting objects on different scales. Before this paper, a lot of research has been avoiding these pyramid structures due to their high computational and memory costs. Feature Pyramid Network tackles…
Creating Your Original Dataset Here is 1 way you can create your own dataset to train an object detection model. Install jmd_imagescraper pip install jmd_imagescraper Get Images #Imports from jmd_imagescraper.core import * from pathlib import Path #Set Paths root_path =…
ArcFace Loss One of tha main challenges in feature learning using Deep Convolutional Neural Networks for large scale face recognition is designing the optimal loss function that enhance discriminative power. Before this paper, there were mainly 2 lines to train…
Deep Metric Learning Let’s say we want to create a model that can do face recognition(Face Identification and face verification). We CAN use traditional deep learning since it can perform really well but it requires a lot of training data.…
SILU SiLU is proposed as an activation function for neural network function approximation in reinforcement learning, and DSiLU is the derivative function for SiLU. DSiLU is a steeper and “overshot” version of the sigmoid function and it is proposed as…