Sherry (#468)

* ch10-NonLinearBinaryClassification Level5_Pytorch * ch12-MultipleLayerNetwork Level4_Pytorch.py ch12-MultipleLayerNetwork Level4_Pytorch.py * ch05-MultiVariableLinearRegression Pytorch.py ch05-MultiVariableLinearRegression Pytorch.py
2026-09-17 17:20:01 +08:00 · 2020-03-10 07:21:03 +08:00
parent c2a0972958
commit 452028f364
1 changed files with 68 additions and 0 deletions
@@ -0,0 +1,68 @@
+import numpy as np
+import matplotlib.pyplot as plt
+from pathlib import Path
+import math
+from torch.utils.data import TensorDataset, DataLoader
+from HelperClass.NeuralNet_1_1 import *
+import torch.nn as nn
+import torch.nn.functional as F
+import torch
+from torch.optim import Adam
+import warnings
+warnings.filterwarnings('ignore')
+
+file_name = "../../data/ch05.npz"
+
+
+class Model(nn.Module):
+    def __init__(self, input_size):
+        super(Model, self).__init__()
+        self.fc = nn.Linear(input_size, 1)
+    def forward(self, x):
+        x = self.fc(x)
+        return x
+
+if __name__ == '__main__':
+    max_epoch = 500
+    num_category = 3
+    sdr = DataReader_1_1(file_name)
+    sdr.ReadData()
+    sdr.NormalizeX()
+    sdr.NormalizeY()
+
+    num_input = 2       # input size
+    # get numpy form data
+    XTrain, YTrain = sdr.XTrain, sdr.YTrain
+    torch_dataset = TensorDataset(torch.FloatTensor(XTrain), torch.FloatTensor(YTrain))
+
+    train_loader = DataLoader(          # data loader class
+        dataset=torch_dataset,
+        batch_size=32,
+        shuffle=True,
+    )
+
+    loss_func = nn.MSELoss()
+    model = Model(num_input)
+    optimizer = Adam(model.parameters(), lr=1e-4)
+
+    e_loss = []     # mean loss at every epoch
+    for epoch in range(max_epoch):
+        b_loss = []     # mean loss at every batch
+        for step, (batch_x, batch_y) in enumerate(train_loader):
+            optimizer.zero_grad()
+            pred = model(batch_x)
+            loss = loss_func(pred,batch_y)
+            b_loss.append(loss.cpu().data.numpy())
+            loss.backward()
+            optimizer.step()
+            b_loss.append(loss.cpu().data.numpy())
+        e_loss.append(np.mean(b_loss))
+        if epoch % 20 == 0:
+            print("Epoch: %d, Loss: %.5f" % (epoch, np.mean(b_loss)))
+    plt.plot([i for i in range(max_epoch)], e_loss)
+    plt.xlabel('Epoch')
+    plt.ylabel('Mean loss')
+    plt.show()
+
+
+