∫ntegrabℓε ∂iﬀerentiαℓs · unorganised notes, code, and writings of random topics

Ren, He, Girshick, Sun (2016) Faster R-CNN

February 12, 2023 paper

This is the Faster R-CNN paper. It improves over Fast R-CNN and R-CNN in the sense that region proposals are generated from the image. It combines a region proposal network (RPN) with the Fast R-CNN detection network so the convolutional network part is shared. It eliminated the use of region... [more]

Girshick (2015) Fast R-CNN

February 8, 2023 paper

This is the Fast R-CNN paper. It improves over R-CNN. It is a single-stage training model (i.e., only one model to train) on the Pascal VOC 2012 dataset. It is faster than R-CNN by using the RoI pooling layer. [more]

Girshick et al (2014) Rich feature hierarchies for accurate object detection and semantic segmentation

February 4, 2023 paper

This is the R-CNN paper. It is a multi-stage process to take an image as input and produces bounding boxes and classification softmax vector as output. This is called localization of object within an image. The paper is based on Pascal VOC dataset (20 classes). Most of the technical details... [more]

Howard et al (2017) MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

January 30, 2023 paper

Convolutional networks such as AlexNet demonstrated the accuracy image recognition. However, latency as well as model size (i.e., memory) can be a concern. MobileNet proposed in this paper is to make these adjustable. [more]

Great Expectations

January 30, 2023 blog code

Great expectations is a Python library for data validation. It tries to profile a sample dataset for what are the attributes, and the statistics of the attributes (e.g., if numeric, what are min, max, mean, variance, etc.). If you read the documentation, it would suggest you to create a data... [more]