23-10-18

모델 save 시 score 확인 추가
배치 사이즈 오류 수정

fashion_mnist.py

@@ -33,23 +33,6 @@ def get_data():
     return x_train, y_train, x_test, y_test
 
 
-def get_data_test():
-    (x_train, y_train), (x_test, y_test) = fashion_mnist.load_data()
-    x_test = x_test / 255.0
-    x_test = x_test.reshape((10000, 28, 28, 1))
-
-    y_train, y_test = tf.one_hot(y_train, 10), tf.one_hot(y_test, 10)
-
-    x_train, x_test = tf.convert_to_tensor(
-        x_train), tf.convert_to_tensor(x_test)
-    y_train, y_test = tf.convert_to_tensor(
-        y_train), tf.convert_to_tensor(y_test)
-
-    print(f"x_test : {x_test[0].shape} | y_test : {y_test[0].shape}")
-
-    return x_test, y_test
-
-
 def make_model():
     model = Sequential()
     model.add(

fashion_mnist_tf.py

@@ -21,6 +21,7 @@ if gpus:
 
 def get_data():
     (x_train, y_train), (x_test, y_test) = fashion_mnist.load_data()
+    print(f"y_train : {y_train[0]} | y_test : {y_test[0]}")
 
     x_train, x_test = x_train / 255.0, x_test / 255.0
     x_train = x_train.reshape((60000, 28, 28, 1))
@@ -36,9 +37,9 @@ class _batch_generator:
     def __init__(self, x, y, batch_size: int = 32):
         self.batch_size = batch_size
         self.index = 0
-        dataset = tf.data.Dataset.from_tensor_slices((x, y))
-        self.dataset = list(dataset.batch(batch_size))
-        self.max_index = len(dataset) // batch_size
+        self.x = x
+        self.y = y
+        self.setBatchSize(batch_size)
 
     def next(self):
         self.index += 1
@@ -46,18 +47,50 @@ class _batch_generator:
             self.index = 0
         return self.dataset[self.index][0], self.dataset[self.index][1]
 
+    def getMaxIndex(self):
+        return self.max_index
+
+    def getIndex(self):
+        return self.index
+
+    def setIndex(self, index):
+        self.index = index
+
+    def getBatchSize(self):
+        return self.batch_size
+
+    def setBatchSize(self, batch_size):
+        self.batch_size = batch_size
+        self.dataset = list(
+            tf.data.Dataset.from_tensor_slices(
+                (self.x, self.y)).batch(batch_size)
+        )
+        self.max_index = len(self.dataset)
+
+    def getDataset(self):
+        return self.dataset
 
-# BEGIN: 5f8d9bcejpp
-from keras.applications.resnet50 import ResNet50
 
 def make_model():
-    model = ResNet50(weights=None, input_shape=(28, 28, 1), classes=10)
+    model = Sequential()
+    model.add(
+        Conv2D(32, kernel_size=(5, 5), activation="relu",
+               input_shape=(28, 28, 1))
+    )
+    model.add(MaxPooling2D(pool_size=(3, 3)))
+    model.add(Conv2D(64, kernel_size=(3, 3), activation="relu"))
+    model.add(MaxPooling2D(pool_size=(2, 2)))
+    model.add(Dropout(0.25))
+    model.add(Flatten())
+    model.add(Dense(256, activation="relu"))
+    model.add(Dense(10, activation="softmax"))
+
     return model
-# END: 5f8d9bcejpp
 
 
 model = make_model()
 x_train, y_train, x_test, y_test = get_data()
+print(x_train.shape)
 y_train = tf.one_hot(y_train, 10)
 y_test = tf.one_hot(y_test, 10)
 
@@ -67,13 +100,14 @@ model.compile(optimizer="adam", loss="mse", metrics=["accuracy"])
 
 count = 0
 
-while count < 50:
+while count < 100:
     x_batch, y_batch = dataset.next()
     count += 1
     print("Training model...")
     model.fit(x_batch, y_batch, epochs=1, batch_size=1, verbose=1)
 
 print(count)
+print(f"Max index : {dataset.getMaxIndex()}")
 
 print("Evaluating model...")
 model.evaluate(x_test, y_test, verbose=2)

pso/__init__.py

@@ -1,7 +1,7 @@
 from .optimizer import Optimizer as optimizer
 from .particle import Particle as particle
 
-__version__ = "0.1.8"
+__version__ = "0.1.9"
 
 __all__ = [
     "optimizer",

pso/optimizer.py

@@ -232,9 +232,9 @@ class Optimizer:
         def __init__(self, x, y, batch_size: int = 32):
             self.batch_size = batch_size
             self.index = 0
-            dataset = tf.data.Dataset.from_tensor_slices((x, y))
-            self.dataset = list(dataset.batch(batch_size))
-            self.max_index = len(dataset) // batch_size
+            self.x = x
+            self.y = y
+            self.setBatchSize(batch_size)
 
         def next(self):
             self.index += 1
@@ -242,6 +242,32 @@ class Optimizer:
                 self.index = 0
             return self.dataset[self.index][0], self.dataset[self.index][1]
 
+        def getMaxIndex(self):
+            return self.max_index
+
+        def getIndex(self):
+            return self.index
+
+        def setIndex(self, index):
+            self.index = index
+
+        def getBatchSize(self):
+            return self.batch_size
+
+        def setBatchSize(self, batch_size):
+            self.batch_size = batch_size
+            if self.batch_size > len(self.x):
+                self.batch_size = len(self.x)
+            print(f"batch size : {self.batch_size}")
+            self.dataset = list(
+                tf.data.Dataset.from_tensor_slices(
+                    (self.x, self.y)).batch(batch_size)
+            )
+            self.max_index = len(self.dataset)
+
+        def getDataset(self):
+            return self.dataset
+
     def fit(
         self,
         x,
@@ -575,7 +601,7 @@ class Optimizer:
             print(e)
 
         finally:
-            self.model_save(save_path)
+            self.model_save(x, y, save_path)
             print("model save")
             if save_info:
                 self.save_info(save_path)
@@ -658,7 +684,7 @@ class Optimizer:
         model = self.get_best_model()
         model.save_weights(save_path)
 
-    def model_save(self, save_path: str = "./result"):
+    def model_save(self, x, y, save_path: str = "./result"):
         """
         최고 점수를 받은 모델 저장
 
@@ -669,6 +695,8 @@ class Optimizer:
             (keras.models): 모델
         """
         model = self.get_best_model()
+        score = model.evaluate(x, y, verbose=1)
+        print(f"model acc : {score[1]}, loss : {score[0]}")
         model.save(
             f"./{save_path}/{self.day}/{self.n_particles}_{self.c0}_{self.c1}_{self.w_min}.h5"
         )