From a37bc17cc1fb4b335c7aa41cb430623929dba2bd Mon Sep 17 00:00:00 2001
From: higaki sota <79880317+sno-windy@users.noreply.github.com>
Date: Mon, 26 May 2025 00:36:05 +0900
Subject: [PATCH] =?UTF-8?q?[ch04]=E5=90=84epoch=E3=81=94=E3=81=A8=E3=81=AB?=
 =?UTF-8?q?=E3=81=99=E3=81=B9=E3=81=A6=E3=81=AE=E8=A8=93=E7=B7=B4=E3=83=87?=
 =?UTF-8?q?=E3=83=BC=E3=82=BF=E3=82=92=E8=A6=8B=E3=82=8B=E3=82=88=E3=81=86?=
 =?UTF-8?q?=E3=81=AB=E3=81=99=E3=82=8B?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 ch04/train_neuralnet.py | 55 ++++++++++++++++++++++-------------------
 1 file changed, 29 insertions(+), 26 deletions(-)

diff --git a/ch04/train_neuralnet.py b/ch04/train_neuralnet.py
index 0a0f70af5..528ee5f5a 100644
--- a/ch04/train_neuralnet.py
+++ b/ch04/train_neuralnet.py
@@ -11,7 +11,7 @@
 
 network = TwoLayerNet(input_size=784, hidden_size=50, output_size=10)
 
-iters_num = 10000  # 繰り返しの回数を適宜設定する
+epoch_num = 20
 train_size = x_train.shape[0]
 batch_size = 100
 learning_rate = 0.1
@@ -20,30 +20,33 @@
 train_acc_list = []
 test_acc_list = []
 
-iter_per_epoch = max(train_size / batch_size, 1)
-
-for i in range(iters_num):
-    batch_mask = np.random.choice(train_size, batch_size)
-    x_batch = x_train[batch_mask]
-    t_batch = t_train[batch_mask]
-    
-    # 勾配の計算
-    #grad = network.numerical_gradient(x_batch, t_batch)
-    grad = network.gradient(x_batch, t_batch)
-    
-    # パラメータの更新
-    for key in ('W1', 'b1', 'W2', 'b2'):
-        network.params[key] -= learning_rate * grad[key]
-    
-    loss = network.loss(x_batch, t_batch)
-    train_loss_list.append(loss)
-    
-    if i % iter_per_epoch == 0:
-        train_acc = network.accuracy(x_train, t_train)
-        test_acc = network.accuracy(x_test, t_test)
-        train_acc_list.append(train_acc)
-        test_acc_list.append(test_acc)
-        print("train acc, test acc | " + str(train_acc) + ", " + str(test_acc))
+iter_per_epoch = max(train_size // batch_size, 1)
+
+
+for epoch in range(epoch_num):
+    perm = np.random.permutation(train_size)
+
+    for i in range(iter_per_epoch):
+        batch_mask = perm[i * batch_size : (i + 1) * batch_size]
+        x_batch = x_train[batch_mask]
+        t_batch = t_train[batch_mask]
+
+        # 勾配の計算
+        #grad = network.numerical_gradient(x_batch, t_batch)
+        grad = network.gradient(x_batch, t_batch)
+
+        # パラメータの更新
+        for key in ('W1', 'b1', 'W2', 'b2'):
+            network.params[key] -= learning_rate * grad[key]
+
+        loss = network.loss(x_batch, t_batch)
+        train_loss_list.append(loss)
+
+    train_acc = network.accuracy(x_train, t_train)
+    test_acc = network.accuracy(x_test, t_test)
+    train_acc_list.append(train_acc)
+    test_acc_list.append(test_acc)
+    print("train acc, test acc | " + str(train_acc) + ", " + str(test_acc))
 
 # グラフの描画
 markers = {'train': 'o', 'test': 's'}
@@ -54,4 +57,4 @@
 plt.ylabel("accuracy")
 plt.ylim(0, 1.0)
 plt.legend(loc='lower right')
-plt.show()
\ No newline at end of file
+plt.show()