1. Statistical Classification
Problem Type | Memo with Demo |
---|---|
Binary Classification |
|
Multi Class Classification |
|
Multi Label Classification |
|
2. Classification Problem
-
分类Classification和回归Regression(预测数字)是常见ML问题类型;
Topic | Content |
---|---|
0:NN Classification Architecture |
|
1:Getting Binary Classification Data Ready |
|
2:Building PyTorch Classification Model |
|
3:Fitting Model to Data(Training) |
|
4:Prediction and Evaluating(Inference) |
|
5:Improving Model |
|
6:Non Linearity |
|
7:Replicating Non Linearity Function |
|
8:Together MultiClass Classification |
|
4. Url
5. NNC Architecture
Hyperparameter | Binary Classification | Multiclass Classification |
---|---|---|
Input Layer Shape |
与特征数量相同,如心脏病预测: |
同binary classification |
Hidden Layer(s) |
特定问题,最小(minimum)=1, |
同binary classification |
Neuron Per Hidden Layer |
generally(通常)10 to 512 |
同binary classification |
Output Layer Shape |
1(one class or the other) |
1 per class, |
Hidden Layer Activation |
通常ReLU(整流线性单元); |
同binary classification |
Output Activation |
Sigmoid:torch.sigmoid |
Softmax:torch.softmax |
Loss Function |
Binary Crossentropy:torch.nn.BCELoss |
Cross Entropy:torch.nn.CrossEntropyLoss |
Optimizer |
SGD,Adam:torch.optim |
同binary classification |