K. Fujita's website

Dataset

dataset classes Objects Dimension
iris 3 150 4
wine 3 178 13
Libras Movement 15 360 91
spam 2 4601 57
CNAE-9 9 1080 857

Image datasets

dataset classes Objects Dimension
CIFAR-10 10 60000 (50000 training images and 10000 test images) 32 * 32
Handwritten digits 10 5620 8 * 8
Olivetti face images 40 400 92 * 112 (64 * 64)
MNIST 10 60,000 (training) 10,000 (test) 28 * 28
Stanford Dogs Dataset 120 20580 375 * 50
STL10 10 500 training images, 800 test images per class 96 * 96

GPU関連情報

  bus clock (GHz) sm CUDA cores Tensor Cores メモリGB メモリ帯域GB/s Single (Tflops) Double (Gflops) Half (Tflops) TDP (W)
RTX 3060 PCIe3x16 1.32 (1.777) 28 3584 112 12 360 12.74 199.0 12.74 170
RTX 3060Ti PCIe3x16 1.41 (1.665) 38 4864 152 8 448 16.2 263.1 16.2 200
RTX 3070 PCIe3x16 1.5 (1.725) 46 5888 184 8 448 20.31 317.4 20.31 220
RTX 3080 PCIe3x16 1.44 (1,71) 68 8704 272 10 760.3 29.77 465.1 29.77 320
RTX 3080Ti PCIe3x16 1.365 (1,665) 80 10240 320 12 912.4 34.10 532.8 34.10 350
RTX 3090 PCIe3x16 1.395 (1.695) 82 10496 328 24 936.2 35.58 556.0 35.58 350
  bus clock (MHz) sm CUDA cores Tensor Cores メモリGB メモリ帯域GB/s Single (Gflops) Double (Gflops) Half (Gflops) TDP (W)
GTX 1650 PCIe3x16 1485 14 896 - 4 128 2661 83 5322 75
RTX 2060 PCIe3x16 1365 30 1920 240 6 336 5242 164 10483 160
RTX 2060 Super PCIe3x16 1470 34 2176 272 8 448 6123 191 12246 175
RTX 2070 PCIe3x16 1410 36 2304 288 8 448 6497 203 12994 175
RTX 2070 Super PCIe3x16 1605 40 2560 320 8 448 8218 257 16435 215
RTX 2080 PCIe3x16 1515 46 2944 368 8 448 8920 279 17840 215
RTX 2080 Super PCIe3x16 1650 48 3072 384 8 496 10138 317 20275 250
RTX 2080Ti PCIe3x16 1350 68 4352 544 11 616 11750 367 23500 250
Titan RTX PCIe3x16 1350 72 4608 576 24 672 12442 389 24884 280
  bus clock (MHz) six, sm CUDA cores Tensor Cores メモリGB メモリ帯域GB/s Single (Gflops) Double (Gflops) Half (Gflops) TDP (W)
K20c PCIe2x16 706 13 2496 - 5 208 3524 1175 - 225
GTX750Ti PCIe3x16 1020 5 640 - 2 86.4 1306 40.8 - 60
k2200 PCIe2x16 1046 5 640 - 4 80 1300 40 - 68
GTX1050Ti PCIe3x16 1290 6 768 - 4 112 1981 62 - 75
GTX1060 6G PCIe3x16 1506 10 1280 - 6 192 3855 120 - 120
GTX1080 PCIe3x16 1607 20 2560 - 8 320 8228 257 - 180
GTX1080Ti PCIe3x16 1480 28 3584 - 11 484 10609 332 - 250
Titan V PCIe3x16 1200 80 5120 640 12 652.8 12288 6144 24576 250

適当な歴史年表 (気が向い時に増えていく)

機械学習など 人工ニューラルネットワーク その他
1700s ベイズ定理(Bayes)    
1795 最小二乗法 (Gauss)    
1865     Clausius’ entropy (Clausius)
1870s     Boltzmann’s entropy (Boltzmann)
1873     ゴルジ染色 (Golgi)
1888     Neuron doctrine (Cajal)
1901 PCA (Pearson)    
1905 Random walk (Pearson)   ブラウン運動 (Einstein)
1906     Receptive field (Sherrington)
1912-22 最尤推定 (Fisher)    
1920     Ising model (Lenz)
1938     Receptive field (Hartline)
1943     Neuron model (McCulloch and Pitts)
1948 Shannon’s entropy (Shannon)    
1949     Hebbian Learning (Hebb)
1952     Hodgkin Huxley model (Hodgkin and Huxley)
1956 Artificial Intelligence (Dartmouth Workshop)    
1957 Bellman equation (Bellman) Perceptron (Rosenblatt)  
1959     V1 (Hubel and Wiesel)
1960   Delta rule (Widrow and Hoff)  
1966 Hidden Markov Model (Baum and Petrie)    
1967 k-means (MacQueen)    
1969   Perceptrons (Minski and Papert)  
    Rectified linear (Fukushima)  
1973     Self organizing map (SOM) (von der Malsberg)
1977 EM Algorithm (Dempster et al.)    
1980     Neocognitron (Fukushima)
1980   SOM (Amari)  
1982   Hopfield Network (Hopfield) Vision (Marr)
1983   SOM (Kohonen)  
1986   Backpropagation (Rumelhart et al.)  
    Boltzmann Machine (Ackley, Hinton, Seinowski)  
1988   Autoencoder (Baldi and Hornik)  
    Autoencoder (Bourland and Lecun)  
1989   LeNet (LeCun)  
1994 ICA (Pierre)    
1995 Positive Matrix Factorization (Paatero and Tapper)    
  SVM (Vapnik)    
1996     Sparse Coding (Olshausen and Field)
1997 Deep Blue beats Kasparov. LSTM (Hochreiter and Schmidhuber)  
1998 Kernel PCA (Scholkopf et al.)    
  Quantum annealing (Kadowaki and Nishimori)    
1999 Non-Negative Factorization (Lee and Seung)    
2001 Topographic ICA (Hyvarinen et al.)   Echo state network (Jaeger)
2002     Liquid state machine (Maass and Markram)
2003     Izhikevich model (Izhikevich)
2006 IVA (Kim et al.) Deep Belief Network (Hinton and Salakhutdinov)  
  Monte Calro Tree Search (Coulom)    
2007   Stacked Autoencoder (Bengio et al.)  
2010   Rectufied linear (Nair et al.)  
2012   AlexNet (Krizhevsky, Sutskever, Hinton)  
    Droout(Hinton et al.)  
2014   GoogLeNet (Szegedy et al.)  
    VGG (Simonyan and Zisserman)  
2015   ResNet (He et al.)  
    Deep Q network (Mnih et al.)  
2016   Xception (Chollet)  
    AlphaGo (Silver et al.)  
2017   Squeeze-and-Excitation Networks (Hu et al.)  
    AlphaGo Zero (Silver et al.)  
2018   AlphaZero (Silver et al.)  
    R2D2  
2019   EfficientNet (Tan and Le)  

このページの技術的なコンセプト

  • markdownでページを作成する。
  • CGIを使わない。nanocでHTMLを生成する。
  • ページのデザインはBootstrapにお任せする。

Last Update: 2021/07/15