46. Dataset : Training
• Number of data : 60,000
– Variable : x (image)
• Type : Image
• Shape : 1, 28, 28
– Variable : y (label)
• Type : Scalar
pptxレポート出力機能を用いて出力した実際のレポート
47. Dataset : Examples of variable x in "Training"
pptxレポート出力機能を用いて出力した実際のレポート
48. Dataset : Validation
• Number of data : 10,000
– Variable : x (image)
• Type : Image
• Shape : 1, 28, 28
– Variable : y (label)
• Type : Scalar
pptxレポート出力機能を用いて出力した実際のレポート
49. Dataset : Examples of variable x in "Validation"
pptxレポート出力機能を用いて出力した実際のレポート
50. Network Architecture : Main
Type Value
Output 26,929
CostParameter 70,054
CostAdd 21,920
CostMultiply 3,144
CostMultiplyAdd 700,904
CostDivision 10
CostExp 110
CostIf 13,124
pptxレポート出力機能を用いて出力した実際のレポート
51. Training Procedure : Optimizer
• Optimize network "Main" using "Training" dataset.
– Batch size : 64
– Solver : Adam
• Learning rate(Alpha) : 0.001
• Beta1 : 0.9
• Beta2 : 0.999
• Epsilon : 1e-08
– Weight decay is not applied.
pptxレポート出力機能を用いて出力した実際のレポート
54. References
• Sony Corporation. Neural Network Console : Not just train and evaluate. You can design neural
networks with fast and intuitive GUI. https://dl.sony.com/
• Sony Corporation. Neural Network Libraries : An open source software to make research,
development and implementation of neural network more efficient. https://nnabla.org/
• BatchNormalization - Ioffe and Szegedy, Batch Normalization: Accelerating Deep Network
Training by Reducing Internal Covariate Shift. https://arxiv.org/abs/1502.03167
• Convolution - Chen et al., DeepLab: Semantic Image Segmentation with Deep Convolutional
Nets, Atrous Convolution, and Fully Connected CRFs. https://arxiv.org/abs/1606.00915, Yu et al.,
Multi-Scale Context Aggregation by Dilated Convolutions. https://arxiv.org/abs/1511.07122
• ELU - Clevart et al., Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs).
http://arxiv.org/abs/1511.07289
• Adam - Matthew D. Zeiler. ADADELTA: An Adaptive Learning Rate Method.
https://arxiv.org/abs/1212.5701
pptxレポート出力機能を用いて出力した実際のレポート