![](https://img-blog.csdnimg.cn/20201014180756780.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
Stanford-cs231n
gdtop818
这个作者很懒,什么都没留下…
展开
-
[cv231n] Backpropagation for a Linear Layer
Backpropagation for a Linear Layer是李飞飞大牛学生Justin Johnson(注意断句),点击可以看他的个人主页发表的论文涉及了多个领域,博士期间在CVPR/ICCV/ECCV顶会上发表多达13篇论文。本文是他为cs231n课程撰写关于Backpropagation的notes,也是我迄今为止见过最简洁,数学推理比较完备的一篇notes。在本篇笔记中主要是...原创 2019-02-27 22:37:25 · 482 阅读 · 0 评论 -
[cs231n] Lecture 10 | Recurrent Neural Networks
Lecture 10 | Recurrent Neural NetworksRecurrent Neural Networks: Process Sequencesdifferent types of inputone to many : e.g. Image Captioning image -> sequence of wordsmany to one :e.g....原创 2019-03-12 20:18:14 · 547 阅读 · 0 评论 -
[cv231n] Lecture 12 | Visualizing and Understanding
Lecture 12 | Visualizing and UnderstandingVisualizingMaximally Activating PatchesSaliencySaliency Maps: Segmentation without supervisionIntermediate Features via (...原创 2019-03-17 11:50:07 · 367 阅读 · 0 评论 -
[cs231n] Lecture 14 | Deep Reinforcement Learning
Lecture 14 | Deep Reinforcement LearningWhat is Reinforcement Learning? Markov Decision Processes Q-Learning Policy GradientsCart-Pole ProblemRobot LocomotionAtari GamesGOMa...原创 2019-03-19 19:27:05 · 412 阅读 · 0 评论 -
[cv231n] Lecture 9 | CNN Architectures
Lecture 9 | CNN ArchitecturesReview: LeNet-5AlexNet(2012 first CNN)Input: 227x227x3 imagesFirst layer (CONV1): 96 11x11 filters applied at stride 4Q: what is the output volume size? Hi...原创 2019-03-11 00:00:51 · 283 阅读 · 0 评论 -
[cv231n] Lecture 8 | Deep Learning Software
Lecture 8 | Deep Learning SoftwareDeep Learning SoftwareGPUsare really good for parallel things where you need to do many things at the same time.the dot product of two vectorsCUDA c...原创 2019-03-09 20:54:07 · 225 阅读 · 0 评论 -
[cs231n] Lecture 11 | Detection and Segmentation
Lecture 11 | Detection and SegmentationSemantic Segmentation Object Detection Instance SegmentationSemantic SegmentationProblem: Very inefficient! Not reusing shared features between...原创 2019-03-14 10:54:13 · 276 阅读 · 0 评论 -
[cv231n] Lecture 4 | Introduction to Neural Networks
Lecture 4 | Introduction to Neural Networks原创 2019-02-26 21:50:45 · 222 阅读 · 0 评论 -
[cv231n] Lecture 3 | Loss Functions and Optimization
Lecture 3 | Loss Functions and Optimization关于优化方法补充一条2019/2/27新出的论文:AdaBound,论文中该方法被证明比SGD和ADAM有更好的优化速度,并且也更容易在训练中达到最优效果~~ 本篇文章如果有朋友看不懂的话可以留言给我,我会抽时间把cs231笔记再翻译成中文~Tools: Linear Classification Los...原创 2019-02-26 19:48:40 · 993 阅读 · 0 评论 -
[cv231n] Lecture 2 | Image Classification(2)
Lecture 2 | Image ClassificationPrerequisitesPython Numpy Tutorial Google Cloud Tutorial Image Classification Linear ClassificationAssignment 1K-Nearest Neighbor Linear classifiers: SVM, So...原创 2019-02-26 14:53:50 · 818 阅读 · 0 评论 -
[cv231n] Lecture 2 | Image Classification(1)
Lecture 2 | Image ClassificationPrerequisitesPython Numpy Tutorial Google Cloud Tutorial Image Classification Linear ClassificationAssignment 1K-Nearest Neighbor Linear classifiers: SVM, So...原创 2019-02-26 13:28:56 · 866 阅读 · 0 评论 -
[cv231n] Lecture 1 | Introduction to Convolutional Neural Networks for Visual Recognition
Lecture 1 | Introduction to Convolutional Neural Networks for Visual Recognition Feifei-LiPrerequisitesProficiency in Python, high-level familiarity in C/C++ All class assignments will be in Pyth...原创 2019-02-26 13:10:44 · 298 阅读 · 0 评论 -
[cv231n] Lecture 7 | Training Neural Networks II
Lecture 7 | Training Neural Networks IILearningIn the previous sections, we’ve discussed the static parts of a Neural Networks: how we can set up the network connectivity, the data, and the loss f...原创 2019-03-01 15:43:46 · 1192 阅读 · 0 评论 -
[cv231n] Lecture 6 | Training Neural Networks I (data processing&weight initializing&batch norm)
Lecture 6 | Training Neural Networks I[discussion]Setting up the data and the modelIn the previous section, we introduced a model of a Neuron, which computes a dot product following a non-linear...原创 2019-03-01 14:06:29 · 765 阅读 · 0 评论 -
[cv231n] Lecture 6 | Training Neural Networks I (activations)
Lecture 6 | Training Neural Networks IQuick introIt is possible to introduce neural networks without appealing to brain analogies. In the section on linear classification, we computed scores for d...原创 2019-03-01 11:48:22 · 921 阅读 · 0 评论 -
[cv231n] Lecture 5 | Convolutional Neural Networks
Lecture 5 | Convolutional Neural NetworksDiscussion:Backpropagation and GradientsAssignments:Assignment #1 due kNN, SVM, SoftMax, two-layer networkConvolutional Neural Networks (CNNs / ConvNe...原创 2019-02-28 14:10:51 · 931 阅读 · 0 评论 -
[cv231n] Lecture 4 | Introduction to Neural Networks
Lecture 4 | Introduction to Neural NetworksReference reading:Calculus on Computational Graphs: BackpropagationHow the backpropagation algorithm worksReviewWe want to find the gradient W of L....原创 2019-02-28 12:04:17 · 452 阅读 · 1 评论 -
[cv231n] Derivatives, Backpropagation, and Vectorization
Derivatives, Backpropagation, and Vectorization是李飞飞大牛学生Justin Johnson(注意断句),点击可以看他的个人主页发表的论文涉及了多个领域,博士期间在CVPR/ICCV/ECCV顶会上发表多达13篇论文。本文是他为cs231n课程撰写关于Derivatives, Backpropagation, and Vectorization的n...原创 2019-02-28 00:23:41 · 359 阅读 · 0 评论 -
[cv231n] Lecture 13 | Generative Models
Lecture 13 | Generative ModelsGenerative Models (Unsupervised Learning)PixelRNN and PixelCNN Variational Autoencoders (VAE) Generative Adversarial Networks (GAN)Supervised vs Unsupervised Learn...原创 2019-03-18 23:46:28 · 265 阅读 · 0 评论