AlexNet
노트
- In this paper, we compare the performance of CNN architectures, KCR-AlexNet and KCR-GoogLeNet.[1]
- The executable will create the results folder: /data/local/tmp/alexnet/output.[2]
- AlexNet competed in the ImageNet Large Scale Visual Recognition Challenge on September 30, 2012.[3]
- from_pretrained ( "alexnet" ) model .[3]
- AlexNet achieved a top-5 error around 16%, which was an extremely good result back in 2012.[4]
- AlexNet is made up of eight trainable layers, five convolution layers and three fully connected layers.[4]
- To address overfitting during training, AlexNet uses both data augmentation and dropout layers.[5]
- Training an image on the VGG network uses techniques similar to Krizhevsky et al., mentioned previously (i.e. the training of AlexNet).[5]
- AlexNet is one of the variants of CNN which is also referred to as a Deep Convolutional Neural Network.[6]
- Then the AlexNet applies maximum pooling layer or sub-sampling layer with a filter size 3×3 and a stride of two.[7]
- To achieve optimum optimal values of Deep Learning AlexNet structure, Modified Crow Search (MCS) is presented.[8]
- The selected Alexnet is ineffective to upgrade the model execution.[8]
- AlexNet, as a primary, average, fundamental, and a standout amongst the best DCNN engineering, was first proposed in our work.[8]
- This optimal Alexnet is anything but difficult to upgrade the quality and expanding the SR by standardized learning rates.[8]
- (the size used in the original AlexNet).[9]
- Following the convolutional layers, the original AlexNet had fully-connected layers with 4096 nodes each.[9]
- We saw earlier that Image Classification is a quite easy task thanks to Deep Learning nets such as Alexnet.[10]
- The convolutional part of Alexnet is used to compute the features of each region and then SVMs use these features to classify the regions.[10]
- AlexNet was the pioneer in CNN and open the whole new research era.[11]
- After competing in ImageNet Large Scale Visual Recognition Challenge, AlexNet shot to fame.[12]
- Before exploring AlexNet it is essential to understand what is a convolutional neural network.[12]
- AlexNet was trained on a GTX 580 GPU with only 3 GB of memory which couldn’t fit the entire network.[12]
- The authors of AlexNet used pooling windows, sized 3×3 with a stride of 2 between the adjacent windows.[12]
- The figure shown below shows us that with the help of ReLUs(solid curve), AlexNet can achieve a 25% training error rate.[12]
- The authors of AlexNet extracted random crops sized 227×227 from inside the 256×256 image boundary, and used this as the network’s inputs.[12]
- AlexNet was able to recognize off-center objects and most of its top 5 classes for each image were reasonable.[12]
- AlexNet was not the first fast GPU-implementation of a CNN to win an image recognition contest.[13]
- In short, AlexNet contains 5 convolutional layers and 3 fully connected layers.[14]
- AlexNet famously won the 2012 ImageNet LSVRC-2012 competition by a large margin (15.3% VS 26.2% (second place) error rates).[15]
- AlexNet uses Rectified Linear Units (ReLU) instead of the tanh function, which was standard at the time.[16]
- AlexNet allows for multi-GPU training by putting half of the model’s neurons on one GPU and the other half on another GPU.[16]
- AlexNet vastly outpaced this with a 37.5% top-1 error and a 17.0% top-5 error.[16]
- AlexNet is able to recognize off-center objects and most of its top five classes for each image are reasonable.[16]
- AlexNet is an incredibly powerful model capable of achieving high accuracies on very challenging datasets.[16]
- However, removing any of the convolutional layers will drastically degrade AlexNet’s performance.[16]
- AlexNet consists of eight layers: five convolutional layers, two fully-connected hidden layers, and one fully-connected output layer.[17]
- Second, AlexNet used the ReLU instead of the sigmoid as its activation function.[17]
- Besides, AlexNet changed the sigmoid activation function to a simpler ReLU activation function.[17]
- AlexNet controls the model complexity of the fully-connected layer by dropout (Section 4.6), while LeNet only uses weight decay.[17]
- The architecture used in the 2012 paper is popularly called AlexNet after the first author Alex Krizhevsky.[18]
- As mentioned above, AlexNet was the winning entry in ILSVRC 2012.[18]
- Random crops of size 227×227 were generated from inside the 256×256 images to feed the first layer of AlexNet.[18]
- An important feature of the AlexNet is the use of ReLU(Rectified Linear Unit) Nonlinearity.[18]
- The authors of AlexNet extracted random crops of size 227×227 from inside the 256×256 image boundary to use as the network’s inputs.[18]
- This can be understood from AlexNet, where FC layers contain approx.[19]
- AlexNet is a work of supervised learning and got very good results.[20]
- AlexNet was used as the basic transfer learning model.[21]
- We used AlexNet as the basic transfer learning model and tested different transfer configurations.[21]
- Original AlexNet was performed on two graphical processing units (GPUs).[21]
- Nowadays, researchers tend to use only one GPU to implement AlexNet.[21]
- Figure 2 illustrates the structure of AlexNet.[21]
- The details of learnable weights and biases of AlexNet are shown in Table 3.[21]
- Compared to traditional neural networks, there are several advanced techniques used in AlexNet.[21]
- Hence, we could not directly apply AlexNet as the feature extractor.[21]
- The AlexNet consists of five conv layers (CL1, CL2, CL3, CL4, and CL5) and three fully-connected layers (FCL6, FL7, FL8).[21]
- AlexNet can make full use of all its parameters with a big dataset.[21]
소스
- ↑ Variations of AlexNet and GoogLeNet to Improve Korean Character Recognition Performance
- ↑ Snapdragon Neural Processing Engine SDK: Running the AlexNet Model
- ↑ 3.0 3.1 alexnet-pytorch
- ↑ 4.0 4.1 Running AlexNet on Raspberry Pi with Arm Compute Library – Arm Developer
- ↑ 5.0 5.1 A Guide to AlexNet, VGG16, and GoogleNet
- ↑ Hands-on Guide To Implementing AlexNet With Keras For Multi-Class Image Classification
- ↑ AlexNet Implementation Using Keras
- ↑ 8.0 8.1 8.2 8.3 Images super-resolution by optimal deep AlexNet architecture for medical application: A novel DOCALN
- ↑ 9.0 9.1 Deep convolutional neural networks — The Straight Dope 0.1 documentation
- ↑ 10.0 10.1 What is Object Detection?
- ↑ Architecture of AlexNet and its current use
- ↑ 12.0 12.1 12.2 12.3 12.4 12.5 12.6 AlexNet: The First CNN to win Image Net
- ↑ Wikipedia
- ↑ ImageNet Classification with Convolutional Neural Networks
- ↑ A Walk-through of AlexNet
- ↑ 16.0 16.1 16.2 16.3 16.4 16.5 AlexNet: The Architecture that Challenged CNNs
- ↑ 17.0 17.1 17.2 17.3 7.1. Deep Convolutional Neural Networks (AlexNet) — Dive into Deep Learning 0.15.1 documentation
- ↑ 18.0 18.1 18.2 18.3 18.4 Understanding AlexNet
- ↑ ResNet, AlexNet, VGGNet, Inception: Understanding various architectures of Convolutional Networks – CV-Tricks.com
- ↑ Explanation of AlexNet and its leap for CNNs
- ↑ 21.0 21.1 21.2 21.3 21.4 21.5 21.6 21.7 21.8 21.9 Alcoholism Identification Based on an AlexNet Transfer Learning Model