phidnet 0.0.5

Phidnet

Phidnet

---

1. Introduction to phidnet

• Phidnet is a library developed for neural network construction and deep learning.

---

2. Install phidnet

• pip install phidnet
• PyPI url: https://pypi.org/project/phidnet/

---

3. Requirements of phidnet

• numpy
• matplotlib
• pandas(Optional)

---

4. Use phidnet

• Import phidnet
  • import phidnet
• Numpy
  • All data, such as matrix and vector, must be converted to numpy array object.
• Configuration of the Piednet
  • phidnet.activation
  • phidnet.optimizer
  • phidnet.load
  • phidnet.matrix
  • phidnet.set
  • phidnet.one_hot_encode
  • phidnet.model
• Define activation function
  • Sigmoid = phidnet.activation.Sigmoid()
  • Relu = phidnet.activation.Relu()
  • ect
• Define optimizer
  • SGD = phidnet.optimizer.SGD(lr=0.01) # lr: learning rate
  • Momentum = phidnet.optimizer.Momentum(lr=0.01, momentum=0.9)
  • AdaGrad = phidnet.optimizer.AdaGrad(lr=0.01)
• Set data
  • Set input data
    • phidnet.set.input_data(X)
  • Set output data
    • phidnet.set.target_data(T)
• Set weight and bias
  • phidnet.set.weight(row, column, layer=layer)
  • phidnet.set.bias(column, layer=layer)
  • phidnet.set.weight(2, 10, layer=1) # 2×10 matrix, 1st layer
  • phidnet.set.bias(10, layer=1) # 1×10 matrix, 1st layer
• Build neural network
  • phidnet.set.build_network(layer)
  • The number of layers is the total number of layers, excluding the input layer. For example, a network with one input layer, one output layer, and one hidden layer in between is a two-layer.
• Set activation function of neural network
  • phidnet.set.activation_function(function_list)
  • phidnet.set.activation_function([Sigmoid, Sigmoid]) # 1st layer: Sigmoid, 2nd layer: Sigmoid
  • The example is the activation functions of the two-layer and Sigmoid, an element of list, is the instance of phid.activation.Sigmoid() class
• Fit model
  • phidnet.model.fit(epoch=1000, optimizer=SGD, print_rate=100, save=True)
  • In the example, train the model for epoch. SGD is the instance of phid.optimizer.SGD() class. Every 100 epoch, print the loss, accuracy of model(print rate). If save= is true, save weight and bias in pickle. Default: save=False
• Predict
  • predicted = phidnet.model.predict(input, exponential=True, precision=2)
  • In the example, the model returns the predicted value in the predicted variable. If exponential= is True, the model returns exponential representation value like 1e-6. When exponential=False, The model returns the value represented by the decimal like 0.018193. The model returns precise values as set to precision. When output is 0.27177211, precision=3, output is 0.271.
• Load
  • phidnet.load.model('C:\examples')
  • If you set it to save=True and trained the model, there would be a file called saved_weight, saved_bias. If the file is in C:\examples\saved_... , you can load trained weight and bias as in the example.
• View fitting
  • phid.model.show_fit()
  • It shows a change in loss and accuracy.
• One hot encoding
  • phidnet.one_hot_encode.encode(number, length=length)
  • phidnet.one_hot_encode.encode(3, length=5) # [0, 0, 0, 1, 0]
  • phidnet.one_hot_encode.encode_array(array, length=length)
  • phidnet.one_hot_encode.encode_array([[1], [2], [3]], length=5) # [[0, 1, 0, 0, 0], [0, 0, 1, 0, 0], [0, 0, 0, 1, 0]]
  • phidnet.one_hot_encode.get_number(one_hot_encoded)
  • phidnet.one_hot_encode.get_number([0, 0, 1, 0, 0]) # 2
• Matrix operations
  • m = phid.matrix.matrix(list) # It converts the list into a matrix (※ phidnet matrix object. not numpy object)
  • ect.

---

5. Use phidnet's convolution neural network

• ect.

---

6. Example phidnet

• Refer to examples for details.