The goal of setting up this repo is to make full use of Stanford CS 231N course.
This repo mainly provides the following features:
- For review purpose : A more convenient visualization of jupyter notebooks without setting up notebook server locally.
- The references of papers which appear in the 5 courses as well as some notes about the papers
- Nicely commented code from helper functions to project architecture as well as a guideline of how to go through them
- Extend the project to end to end system: from data labeling to research diary
Recourse collection contributors: Michael Wang
[1] "Deep Visual-Semantic Alignments for Generating Image Descriptions"[pdf]
[1] "Object Recognition from Local Scale-Invariant Features"[pdf]
[2] "Histograms of Oriented Gradients for Human Detection"[pdf]
[3] "A Bayesian Hierarchical Model for Learning Natural Scene Categories"[pdf]
[1] "Object Recognition from Local Scale-Invariant Features"[pdf]
[2] "Histograms of Oriented Gradients for Human Detection"[pdf]
[1] "ImageNet Classification with Deep Convolutional Neural Networks"[pdf]
[2] "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks"[pdf]
[3] "DeepFace: Closing the Gap to Human-Level Performance in Face Verification"[pdf]
[4] "Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps"[pdf]
[5] "Two-Stream Convolutional Networks for Action Recognition in Videos"[pdf]
[6] "DeepPose: Human Pose Estimation via Deep Neural Networks"[pdf]
[7] "Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning"[pdf]
[8] "Breast Mass Classification from Mammograms using Deep Convolutional Neural Networks"[pdf]
[9] "Rotation-invariant convolutional neural networks for galaxy morphology prediction"[pdf]
[10] "Traffic Sign Recognition with Multi-Scale Convolutional Networks"[pdf]
[11] "Image Style Transfer Using Convolutional Neural Networks"[pdf]
[12] "Controlling Perceptual Factors in Neural Style Transfer"[pdf]
[1] Papers on weight initializations are on the lecture slide.
[2] "Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift"[pdf]
[1] "On the importance of initialization and momentum in deep learning"[pdf]
[2] "Adaptive Subgradient Methods for Online Learning and Stochastic Optimization"[pdf]
[3] "ADAM: A METHOD FOR STOCHASTIC OPTIMIZATION"[pdf]
[4] "On Optimization Methods for Deep Learning"[pdf]
[5] "SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS"[pdf]
[6] "SNAPSHOT ENSEMBLES: TRAIN 1, GET M FOR FREE"[pdf]
[7] "ACCELERATION OF STOCHASTIC APPROXIMATION BY AVERAGING*"[pdf]
[8] "Regularization of Neural Networks using DropConnect*"[pdf]
[9] "Deep Networks with Stochastic Depth"[pdf]
[10] "DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition"[pdf]
[11] "DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition"[pdf]
[12] "Fast RCNN"[pdf]
[1] "DEEP LEARNING WITH DYNAMIC COMPUTATION GRAPHS"[pdf]
[2] "Deep Compositional Question Answering with Neural Module Networks"[pdf]
[3] "Learning to Compose Neural Networks for Question Answering"[pdf]
[1] "ImageNet Classification with Deep Convolutional Neural Networks(AlexNet)"[pdf]
[2] "Visualizing and Understanding Convolutional Networks"[pdf]
[3] "VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION (VGG)"[pdf]
[4] "Going deeper with convolutions (Google Net, Inception Net)"[pdf]
[5] "Deep Residual Learning for Image Recognition (ResNet)"[pdf]
[6] "Network in Network (NiN)"[pdf]
[7] "Wide Residual Networks"[pdf]
[8] "Aggregated Residual Transformations for Deep Neural Networks (ResNext)"[pdf]
[9] "Wide Residual Networks"[pdf]
[10] "Deep Networks with Stochastic Depth"[pdf]
[11] "FRACTALNET: ULTRA-DEEP NEURAL NETWORKS WITHOUT RESIDUALS"[pdf]
[12] "Densely Connected Convolutional Networks"[pdf]
[13] "SQUEEZENET: ALEXNET-LEVEL ACCURACY WITH 50X FEWER PARAMETERS AND <0.5MB MODEL SIZE"[pdf]