kerasを試してみたいがエラーが出る

Question

前提

サイトの丸写しをしていたところ、エラーが発生しました。
https://www.sejuku.net/blog/53512

実現したいこと

エラーが発生しなくなり、サイトのコードが動作すること。

発生している問題・エラーメッセージ

x_train shape: (60000, 28, 28, 1)
60000 train samples
10000 test samples
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
/var/folders/1b/4r83ph916kb86bjsvm74d7740000gn/T/ipykernel_1047/3293630069.py in <module>
     42 model.add(Dense(num_classes, activation='softmax'))
     43 
---> 44 model.compile(loss=keras.losses.categorical_crossentropy,
     45               optimizer=Adadelta(),
     46               metrics=['accuracy'])

/usr/local/lib/python3.9/site-packages/tensorflow/python/keras/engine/training.py in compile(self, optimizer, loss, metrics, loss_weights, weighted_metrics, run_eagerly, steps_per_execution, **kwargs)
    571       self._run_eagerly = run_eagerly
    572 
--> 573       self.optimizer = self._get_optimizer(optimizer)
    574       self.compiled_loss = compile_utils.LossesContainer(
    575           loss, loss_weights, output_names=self.output_names)

/usr/local/lib/python3.9/site-packages/tensorflow/python/keras/engine/training.py in _get_optimizer(self, optimizer)
    609       return opt
    610 
--> 611     return nest.map_structure(_get_single_optimizer, optimizer)
    612 
    613   @trackable.no_automatic_dependency_tracking

/usr/local/lib/python3.9/site-packages/tensorflow/python/util/nest.py in map_structure(func, *structure, **kwargs)
    867 
    868   return pack_sequence_as(
--> 869       structure[0], [func(*x) for x in entries],
    870       expand_composites=expand_composites)
    871 

/usr/local/lib/python3.9/site-packages/tensorflow/python/util/nest.py in <listcomp>(.0)
    867 
    868   return pack_sequence_as(
--> 869       structure[0], [func(*x) for x in entries],
    870       expand_composites=expand_composites)
    871 

/usr/local/lib/python3.9/site-packages/tensorflow/python/keras/engine/training.py in _get_single_optimizer(opt)
    600 
    601     def _get_single_optimizer(opt):
--> 602       opt = optimizers.get(opt)
    603       if (loss_scale is not None and
    604           not isinstance(opt, lso.LossScaleOptimizer)):

/usr/local/lib/python3.9/site-packages/tensorflow/python/keras/optimizers.py in get(identifier)
    129     return deserialize(config)
    130   else:
--> 131     raise ValueError(
    132         'Could not interpret optimizer identifier: {}'.format(identifier))

ValueError: Could not interpret optimizer identifier: <keras.optimizer_v2.adadelta.Adadelta object at 0x139ae80d0>

該当のソースコード

'''Trains a simple convnet on the MNIST dataset.
Gets to 99.25% test accuracy after 12 epochs
(there is still a lot of margin for parameter tuning).
16 seconds per epoch on a GRID K520 GPU.
'''
 
from __future__ import print_function
from tensorflow.keras.utils import to_categorical
from tensorflow.keras.optimizers import Adadelta
from tensorflow.python.keras.datasets import mnist
from tensorflow.python.keras.models import Sequential
from tensorflow.python.keras.layers import Dense, Dropout, Flatten
from tensorflow.python.keras.layers import Conv2D, MaxPooling2D
from tensorflow.python.keras import backend as K

batch_size = 128
num_classes = 10
epochs = 12
 
# input image dimensions
img_rows, img_cols = 28, 28
 
# the data, split between train and test sets
(x_train, y_train), (x_test, y_test) = mnist.load_data()
 
if K.image_data_format() == 'channels_first':
    x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)
    x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)
    input_shape = (1, img_rows, img_cols)
else:
    x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1)
    x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1)
    input_shape = (img_rows, img_cols, 1)
 
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
print('x_train shape:', x_train.shape)
print(x_train.shape[0], 'train samples')
print(x_test.shape[0], 'test samples')
 
# convert class vectors to binary class matrices
y_train = to_categorical(y_train, num_classes)
y_test = to_categorical(y_test, num_classes)
 
model = Sequential()
model.add(Conv2D(32, kernel_size=(3, 3),
                 activation='relu',
                 input_shape=input_shape))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes, activation='softmax'))
 
model.compile(loss=keras.losses.categorical_crossentropy,
              optimizer=Adadelta(),
              metrics=['accuracy'])
 
model.fit(x_train, y_train,
          batch_size=batch_size,
          epochs=epochs,
          verbose=1,
          validation_data=(x_test, y_test))
score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])

試したこと

以下のサイトを参考にエラーを直していきました。
https://keras.io/ja/optimizers/#adadelta
https://bobbyhadz-com.translate.goog/blog/python-attributeerror-module-keras-utils-has-no-attribute-to-categorical?_x_tr_sl=en&_x_tr_tl=ja&_x_tr_hl=ja&_x_tr_pto=sc

補足情報（FW/ツールのバージョンなど）

jupyter notebookで作業しています。

import tensorflow as tf

print(tf.__version__) # 1.5.0

を実行したところ2.6.0と出ました。

Comments

[jbpb0]

from tensorflow.python.keras.datasets import mnist
from tensorflow.python.keras.models import Sequential
from tensorflow.python.keras.layers import Dense, Dropout, Flatten
from tensorflow.python.keras.layers import Conv2D, MaxPooling2D
from tensorflow.python.keras import backend as K
↓ 変更
import tensorflow as tf
from tensorflow.keras.datasets import mnist
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout, Flatten
from tensorflow.keras.layers import Conv2D, MaxPooling2D
from tensorflow.keras import backend as K

model.compile(loss=keras.losses.categorical_crossentropy,
↓ 変更
model.compile(loss=tf.keras.losses.categorical_crossentropy,

で、どうでしょうか？

[miraimirai]

変更点
from __future__ import print_function
from tensorflow.keras.utils import to_categorical
from tensorflow.keras.optimizers import Adadelta
from tensorflow.keras.datasets import mnist
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout, Flatten
from tensorflow.keras.layers import Conv2D, MaxPooling2D
from tensorflow.keras import backend as K
from tensorflow.keras.losses import categorical_crossentropy

model.compile(loss=categorical_crossentropy(),
optimizer=Adadelta(),
metrics=['accuracy'])

上記のように変更したところ以下のエラーが発生しました。
x_train shape: (60000, 28, 28, 1)
60000 train samples
10000 test samples
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
/var/folders/1b/4r83ph916kb86bjsvm74d7740000gn/T/ipykernel_1109/3977368573.py in <module>
42 model.add(Dense(num_classes, activation='softmax'))
43
---> 44 model.compile(loss=categorical_crossentropy(),
45 optimizer=Adadelta(),
46 metrics=['accuracy'])

/usr/local/lib/python3.9/site-packages/tensorflow/python/util/dispatch.py in wrapper(*args, **kwargs)
204 """Call target, and fall back on dispatchers if there is a TypeError."""
205 try:
--> 206 return target(*args, **kwargs)
207 except (TypeError, ValueError):
208 # Note: convert_to_eager_tensor currently raises a ValueError, not a

TypeError: categorical_crossentropy() missing 2 required positional arguments: 'y_true' and 'y_pred'

[miraimirai]

loss=categorical_crossentropyに変更したところ実行できました。

[miraimirai]

loss=categorical_crossentropyの意味を教えてもらえませんか？

[miraimirai]

https://teratail.com/questions/228397
は読んだのですが、関数そのものを渡してどうなっているのかがわかりません。

[miraimirai]

kerasのドキュメントを読んだら解決しました。損失関数を設定しているのですね。

Answer 1

from tensorflow.python.keras.datasets import mnist
from tensorflow.python.keras.models import Sequential
from tensorflow.python.keras.layers import Dense, Dropout, Flatten
from tensorflow.python.keras.layers import Conv2D, MaxPooling2D
from tensorflow.python.keras import backend as K

↓ 変更

import tensorflow as tf
from tensorflow.keras.datasets import mnist
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout, Flatten
from tensorflow.keras.layers import Conv2D, MaxPooling2D
from tensorflow.keras import backend as K

　

model.compile(loss=keras.losses.categorical_crossentropy,

↓ 変更

model.compile(loss=tf.keras.losses.categorical_crossentropy,