編集履歴

質問編集履歴

画像の前処理コードの反映

2021/09/22 06:54

投稿

ykykykk69

スコア2

test CHANGED Viewed

File without changes

test CHANGED Viewed

@@ -92,276 +92,458 @@
 ```
-指摘いただいた部分を修正
+下記に記載したのがエラー内容
 ```python
-imsize = (4, 4)
+WARNING:tensorflow:Model was constructed with shape (None, 4, 4, 512) for input KerasTensor(type_spec=TensorSpec(shape=(None, 4, 4, 512), dtype=tf.float32, name='flatten_input'), name='flatten_input', description="created by layer 'flatten_input'"), but it was called on an input with incompatible shape (None, 150, 150, 3).
+---------------------------------------------------------------------------
+ValueError                                Traceback (most recent call last)
+<ipython-input-12-baa28c867e4e> in <module>
+     35 model = load_model(keras_param)
+     36 img = load_image(testpic)
+---> 37 prd = model.predict(np.array([img]))
+     38 # print(prd) # 精度の表示
+     39 prelabel = np.argmax(prd, axis=1)
 ```
-下記に記載したのがエラー内容
+学習済みファイルの作成コード
 ```python
-WARNING:tensorflow:Model was constructed with shape (None, 4, 4, 512) for input KerasTensor(type_spec=TensorSpec(shape=(None, 4, 4, 512), dtype=tf.float32, name='flatten_input'), name='flatten_input', description="created by layer 'flatten_input'"), but it was called on an input with incompatible shape (None, 150, 150, 3).
----------------------------------------------------------------------------
-ValueError                                Traceback (most recent call last)
-<ipython-input-12-baa28c867e4e> in <module>
-     35 model = load_model(keras_param)
-     36 img = load_image(testpic)
----> 37 prd = model.predict(np.array([img]))
-     38 # print(prd) # 精度の表示
-     39 prelabel = np.argmax(prd, axis=1)
+from keras.applications.vgg16 import VGG16
+from keras.preprocessing.image import ImageDataGenerator
+from keras.layers import Input
+import numpy as np
+import os
+num_train =2000
+num_validation=800
+img_h,img_w=150,150
+channels=3
+batch_size=32
+train_data_dir='data/train'
+validation_data_dir='data/validation'
+result_dir='results'
+if not os.path.exists(result_dir):
+    os.mkdir(result_dir)
+def save_VGG16_outputs():
+    model=VGG16(
+    include_top=False,
+    weights='imagenet',
+    input_shape=(img_h,img_w,channels))
+    model.summary()
+    datagen=ImageDataGenerator(rescale=1.0/255)
+    train_generator=datagen.flow_from_directory(
+        train_data_dir,
+        target_size=(img_w,img_h),
+        batch_size=batch_size,
+        class_mode=None,
+        shuffle=False)
+    print('train-label:',train_generator.class_indices)
+    vgg16_train=model.predict_generator(
+        train_generator,
+        verbose=1)
+    np.save(os.path.join(result_dir,'vgg16_train.npy'),vgg16_train)
+    validation_generator=datagen.flow_from_directory(
+        validation_data_dir,
+        target_size=(img_w,img_h),
+        batch_size=batch_size,
+        class_mode=None,
+        shuffle=False)
+    print('test-label:',validation_generator.class_indices)
+    vgg16_test=model.predict(
+        validation_generator,
+        verbose=1)
+    np.save(os.path.join(result_dir,'vgg16_test.npy'),vgg16_test)
+save_VGG16_outputs()
+import os
+import numpy as np
+from keras.models import Sequential
+from keras import optimizers
+from keras.layers import Activation,Dropout,Flatten,Dense
+def train_FClayer():
+    train_data=np.load(
+        os.path.join(result_dir,'vgg16_train.npy'))
+    print(train_data.shape)
+    train_labels=np.array(
+    [0]*int(num_train/2)+[1]*int(num_train/2))
+    validation_data=np.load(
+    os.path.join(result_dir,'vgg16_test.npy'))
+    print(validation_data.shape)
+    validation_labels=np.array(
+    [0]*int(num_validation/2)+[1]*int(num_validation/2))
+    model=Sequential()
+    model.add(Flatten(input_shape=train_data.shape[1:]))
+    model.add(Dense(256,activation='relu'))
+    model.add(Dense(1,activation='sigmoid'))
+    model.compile(
+        loss='binary_crossentropy',
+        metrics=['accuracy'],
+        optimizer=optimizers.SGD(lr=1e-4,momentum=0.9),)
+    epoch=60
+    batch_size=32
+    history=model.fit(train_data,
+                      train_labels,
+                      epochs=epoch,
+                      batch_size=batch_size,
+                      verbose=1,
+                      validation_data=(validation_data,validation_labels))
+    with open('model.json','w') as json_file:
+        json_file.write(model.to_json())
+    model.save('model_and_weight.h5')
+    return history
+history=train_FClayer()
 ```
-修正後のエラーコード
+画像の処理コード
 ```python
-WARNING:tensorflow:Model was constructed with shape (None, 4, 4, 512) for input KerasTensor(type_spec=TensorSpec(shape=(None, 4, 4, 512), dtype=tf.float32, name='flatten_input'), name='flatten_input', description="created by layer 'flatten_input'"), but it was called on an input with incompatible shape (None, 4, 4, 3).
+from keras.applications.vgg16 import VGG16
+from keras.preprocessing.image import ImageDataGenerator
+from keras.layers import Input
+import numpy as np
+import os
+num_eva =1
+img_h,img_w=150,150
+channels=3
+batch_size=32
+eva_data_dir='data/evaluation'
+result_dir='results'
+if not os.path.exists(result_dir):
+    os.mkdir(result_dir)
+def save_VGG16_outputs():
+    model=VGG16(
+    include_top=False,
+    weights='imagenet',
+    input_shape=(img_h,img_w,channels))
+    model.summary()
+    datagen=ImageDataGenerator(rescale=1.0/255)
+    eva_generator=datagen.flow_from_directory(
+        eva_data_dir,
+        target_size=(img_w,img_h),
+        batch_size=batch_size,
+        class_mode=None,
+        shuffle=False)
+    print('eva-label:',eva_generator.class_indices)
+    vgg16_eva=model.predict_generator(
+        eva_generator,
+        verbose=1)
+    np.save(os.path.join(result_dir,'vgg16_eva.npy'),vgg16_eva)
+    print(vgg16_eva.shape)
+save_VGG16_outputs()
+Model: "vgg16"
+_________________________________________________________________
+Layer (type)                 Output Shape              Param #
+=================================================================
+input_2 (InputLayer)         [(None, 150, 150, 3)]     0
+_________________________________________________________________
+block1_conv1 (Conv2D)        (None, 150, 150, 64)      1792
+_________________________________________________________________
+block1_conv2 (Conv2D)        (None, 150, 150, 64)      36928
+_________________________________________________________________
+block1_pool (MaxPooling2D)   (None, 75, 75, 64)        0
+_________________________________________________________________
+block2_conv1 (Conv2D)        (None, 75, 75, 128)       73856
+_________________________________________________________________
+block2_conv2 (Conv2D)        (None, 75, 75, 128)       147584
+_________________________________________________________________
+block2_pool (MaxPooling2D)   (None, 37, 37, 128)       0
+_________________________________________________________________
+block3_conv1 (Conv2D)        (None, 37, 37, 256)       295168
+_________________________________________________________________
+block3_conv2 (Conv2D)        (None, 37, 37, 256)       590080
+_________________________________________________________________
+block3_conv3 (Conv2D)        (None, 37, 37, 256)       590080
+_________________________________________________________________
+block3_pool (MaxPooling2D)   (None, 18, 18, 256)       0
+_________________________________________________________________
+block4_conv1 (Conv2D)        (None, 18, 18, 512)       1180160
+_________________________________________________________________
+block4_conv2 (Conv2D)        (None, 18, 18, 512)       2359808
+_________________________________________________________________
+block4_conv3 (Conv2D)        (None, 18, 18, 512)       2359808
+_________________________________________________________________
+block4_pool (MaxPooling2D)   (None, 9, 9, 512)         0
+_________________________________________________________________
+block5_conv1 (Conv2D)        (None, 9, 9, 512)         2359808
+_________________________________________________________________
+block5_conv2 (Conv2D)        (None, 9, 9, 512)         2359808
+_________________________________________________________________
+block5_conv3 (Conv2D)        (None, 9, 9, 512)         2359808
+_________________________________________________________________
+block5_pool (MaxPooling2D)   (None, 4, 4, 512)         0
+=================================================================
+Total params: 14,714,688
+Trainable params: 14,714,688
+Non-trainable params: 0
+_________________________________________________________________
+Found 1 images belonging to 1 classes.
+eva-label: {'1': 0}
+1/1 [==============================] - 2s 2s/step
+(1, 4, 4, 512)
 ```
-学習済みファイルの作成コード
-```python
-from keras.applications.vgg16 import VGG16
-from keras.preprocessing.image import ImageDataGenerator
-from keras.layers import Input
-import numpy as np
-import os
-num_train =2000
-num_validation=800
-img_h,img_w=150,150
-channels=3
-batch_size=32
-train_data_dir='data/train'
-validation_data_dir='data/validation'
-result_dir='results'
-if not os.path.exists(result_dir):
-    os.mkdir(result_dir)
-def save_VGG16_outputs():
-    model=VGG16(
-    include_top=False,
-    weights='imagenet',
-    input_shape=(img_h,img_w,channels))
-    model.summary()
-    datagen=ImageDataGenerator(rescale=1.0/255)
-    train_generator=datagen.flow_from_directory(
-        train_data_dir,
-        target_size=(img_w,img_h),
-        batch_size=batch_size,
-        class_mode=None,
-        shuffle=False)
-    print('train-label:',train_generator.class_indices)
-    vgg16_train=model.predict_generator(
-        train_generator,
-        verbose=1)
-    np.save(os.path.join(result_dir,'vgg16_train.npy'),vgg16_train)
-    validation_generator=datagen.flow_from_directory(
-        validation_data_dir,
-        target_size=(img_w,img_h),
-        batch_size=batch_size,
-        class_mode=None,
-        shuffle=False)
-    print('test-label:',validation_generator.class_indices)
-    vgg16_test=model.predict(
-        validation_generator,
-        verbose=1)
-    np.save(os.path.join(result_dir,'vgg16_test.npy'),vgg16_test)
-save_VGG16_outputs()
-import os
-import numpy as np
-from keras.models import Sequential
-from keras import optimizers
-from keras.layers import Activation,Dropout,Flatten,Dense
-def train_FClayer():
-    train_data=np.load(
-        os.path.join(result_dir,'vgg16_train.npy'))
-    print(train_data.shape)
-    train_labels=np.array(
-    [0]*int(num_train/2)+[1]*int(num_train/2))
-    validation_data=np.load(
-    os.path.join(result_dir,'vgg16_test.npy'))
-    print(validation_data.shape)
-    validation_labels=np.array(
-    [0]*int(num_validation/2)+[1]*int(num_validation/2))
-    model=Sequential()
-    model.add(Flatten(input_shape=train_data.shape[1:]))
-    model.add(Dense(256,activation='relu'))
-    model.add(Dense(1,activation='sigmoid'))
-    model.compile(
-        loss='binary_crossentropy',
-        metrics=['accuracy'],
-        optimizer=optimizers.SGD(lr=1e-4,momentum=0.9),)
-    epoch=60
-    batch_size=32
-    history=model.fit(train_data,
-                      train_labels,
-                      epochs=epoch,
-                      batch_size=batch_size,
-                      verbose=1,
-                      validation_data=(validation_data,validation_labels))
-    with open('model.json','w') as json_file:
-        json_file.write(model.to_json())
-    model.save('model_and_weight.h5')
-    return history
-history=train_FClayer()
-```

指摘いただいた部分の修正箇所を反映

2021/09/22 06:54

投稿

ykykykk69

スコア2

test CHANGED Viewed

File without changes

test CHANGED Viewed

@@ -92,6 +92,16 @@
 ```
+指摘いただいた部分を修正
+```python
+imsize = (4, 4)
+```
 下記に記載したのがエラー内容
@@ -118,7 +128,13 @@
 ```
+修正後のエラーコード
+```python
+WARNING:tensorflow:Model was constructed with shape (None, 4, 4, 512) for input KerasTensor(type_spec=TensorSpec(shape=(None, 4, 4, 512), dtype=tf.float32, name='flatten_input'), name='flatten_input', description="created by layer 'flatten_input'"), but it was called on an input with incompatible shape (None, 4, 4, 3).
+```
 学習済みファイルの作成コード

学習済みファイルの作成コードの掲載

2021/09/22 06:22

投稿

ykykykk69

スコア2

test CHANGED Viewed

File without changes

test CHANGED Viewed

@@ -117,3 +117,235 @@
      39 prelabel = np.argmax(prd, axis=1)
 ```
+学習済みファイルの作成コード
+```python
+from keras.applications.vgg16 import VGG16
+from keras.preprocessing.image import ImageDataGenerator
+from keras.layers import Input
+import numpy as np
+import os
+num_train =2000
+num_validation=800
+img_h,img_w=150,150
+channels=3
+batch_size=32
+train_data_dir='data/train'
+validation_data_dir='data/validation'
+result_dir='results'
+if not os.path.exists(result_dir):
+    os.mkdir(result_dir)
+def save_VGG16_outputs():
+    model=VGG16(
+    include_top=False,
+    weights='imagenet',
+    input_shape=(img_h,img_w,channels))
+    model.summary()
+    datagen=ImageDataGenerator(rescale=1.0/255)
+    train_generator=datagen.flow_from_directory(
+        train_data_dir,
+        target_size=(img_w,img_h),
+        batch_size=batch_size,
+        class_mode=None,
+        shuffle=False)
+    print('train-label:',train_generator.class_indices)
+    vgg16_train=model.predict_generator(
+        train_generator,
+        verbose=1)
+    np.save(os.path.join(result_dir,'vgg16_train.npy'),vgg16_train)
+    validation_generator=datagen.flow_from_directory(
+        validation_data_dir,
+        target_size=(img_w,img_h),
+        batch_size=batch_size,
+        class_mode=None,
+        shuffle=False)
+    print('test-label:',validation_generator.class_indices)
+    vgg16_test=model.predict(
+        validation_generator,
+        verbose=1)
+    np.save(os.path.join(result_dir,'vgg16_test.npy'),vgg16_test)
+save_VGG16_outputs()
+import os
+import numpy as np
+from keras.models import Sequential
+from keras import optimizers
+from keras.layers import Activation,Dropout,Flatten,Dense
+def train_FClayer():
+    train_data=np.load(
+        os.path.join(result_dir,'vgg16_train.npy'))
+    print(train_data.shape)
+    train_labels=np.array(
+    [0]*int(num_train/2)+[1]*int(num_train/2))
+    validation_data=np.load(
+    os.path.join(result_dir,'vgg16_test.npy'))
+    print(validation_data.shape)
+    validation_labels=np.array(
+    [0]*int(num_validation/2)+[1]*int(num_validation/2))
+    model=Sequential()
+    model.add(Flatten(input_shape=train_data.shape[1:]))
+    model.add(Dense(256,activation='relu'))
+    model.add(Dense(1,activation='sigmoid'))
+    model.compile(
+        loss='binary_crossentropy',
+        metrics=['accuracy'],
+        optimizer=optimizers.SGD(lr=1e-4,momentum=0.9),)
+    epoch=60
+    batch_size=32
+    history=model.fit(train_data,
+                      train_labels,
+                      epochs=epoch,
+                      batch_size=batch_size,
+                      verbose=1,
+                      validation_data=(validation_data,validation_labels))
+    with open('model.json','w') as json_file:
+        json_file.write(model.to_json())
+    model.save('model_and_weight.h5')
+    return history
+history=train_FClayer()
+```

説明文にコードのことを追加

2021/09/22 02:24

投稿

ykykykk69

スコア2

test CHANGED Viewed

File without changes

test CHANGED Viewed

@@ -1,6 +1,8 @@
 下記のようなコードで画像分類器を作成したいと考えています。
 しかし、エラーが発生してしまい実行できません。
+コードは同じ本から採用しています。
 解決方法がありましたら、教えていただけますでしょうか。