Skip to content
Go back

뉴럴 네트워크 디자인 팁

Updated:
Edit page

Table of contents

Open Table of contents

1. 기본적인 뉴럴 네트워크 구조

DL_1

Input neurons

DL_2

Output neurons

Hidden Layers and Neurons per Hidden Layers

Loss function

DL_4

Batch Size

Number of epochs

Scaling yout features

DL_5

2. 학습률(Learning Rate)

DL_6

3. 관성(Momentum)기법

DL_7

4. 기울기 소실, 폭발(Vanishing + Exploding Gradients)

DL_8

활성 함수(Activation functions)

Output Layer Activation

Weight initialization method

BatchNormalization

Early Stopping

5. Dropout

6. Optimizers

DL_9

7. Learning Rate Scheduling

DL_10

8. A Few More Things


Edit page
Share this post on:

Previous Post
ONNX란?
Next Post
딥러닝 면접 질문과 답변