<div id="readability-page-1" class="page"><div>
텐서플로우에서 다음과 같은 간단한 MNIST 예제를 실행시켜보면 매번 학습 결과가 다른 것을 확인할 수 있습니다.



실행할때마다 일정한 학습 결과를 얻는 방법을 다룹니다.&nbsp;







2021. 7. 29&nbsp; – 최초작성







<pre><code># https://www.tensorflow.org/tutorials/quickstart/beginner
import tensorflow as tf


mnist = tf.keras.datasets.mnist

(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0

model = tf.keras.models.Sequential([
 tf.keras.layers.Flatten(input_shape=(28, 28)),
 tf.keras.layers.Dense(128, activation='relu'),
 tf.keras.layers.Dropout(0.2),
 tf.keras.layers.Dense(10, activation='softmax')
])

model.compile(optimizer='adam',
 loss='sparse_categorical_crossentropy',
 metrics=['accuracy'])

model.fit(x_train, y_train, epochs=5)

model.evaluate(x_test, y_test, verbose=2)</code></pre>







세번 연달아 실행시킨 결과인데 학습 결과가 일정하지 않습니다.&nbsp;



Epoch 2/5



1875/1875 [==============================] – 5s 2ms/step – loss: 0.1442 – accuracy: 0.9568&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 3/5



1875/1875 [==============================] – 5s 2ms/step – loss: 0.1085 – accuracy: 0.9668&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 4/5



1875/1875 [==============================] – 5s 2ms/step – loss: 0.0873 – accuracy: 0.9742&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 5/5



1875/1875 [==============================] – 5s 2ms/step –&nbsp;loss: 0.0757 – accuracy: 0.9765&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



313/313 – 1s – loss: 0.0733 – accuracy: 0.9774







Epoch 2/5



1875/1875 [==============================] – 5s 3ms/step – loss: 0.1469 – accuracy: 0.9560&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 3/5



1875/1875 [==============================] – 4s 2ms/step – loss: 0.1097 – accuracy: 0.9669&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 4/5



1875/1875 [==============================] – 4s 2ms/step – loss: 0.0881 – accuracy: 0.9724&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 5/5



1875/1875 [==============================] – 4s 2ms/step –&nbsp;loss: 0.0751 – accuracy: 0.9772&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



313/313 – 1s – loss: 0.0756 – accuracy: 0.9761







Epoch 2/5



1875/1875 [==============================] – 5s 3ms/step – loss: 0.1424 – accuracy: 0.9578



Epoch 3/5



1875/1875 [==============================] – 5s 3ms/step – loss: 0.1087 – accuracy: 0.9674



Epoch 4/5



1875/1875 [==============================] – 5s 3ms/step – loss: 0.0905 – accuracy: 0.9718



Epoch 5/5



1875/1875 [==============================] – 5s 3ms/step –&nbsp;loss: 0.0776 – accuracy: 0.9761



313/313 – 1s – loss: 0.0735 – accuracy: 0.9771







KERAS FAQ에서 소개하는 다음 코드를 파이썬 코드 첫줄에 추가한 후, 다시 MNIST 예제를 실행해봅니다.



출처 –&nbsp;<a href="https://keras.io/getting_started/faq/#how-can-i-obtain-reproducible-results-using-keras-during-development">https://keras.io/getting_started/faq/#how-can-i-obtain-reproducible-results-using-keras-during-development</a>&nbsp;



<pre><code>import numpy as np
import tensorflow as tf
import random as python_random

# The below is necessary for starting Numpy generated random numbers
# in a well-defined initial state.
np.random.seed(123)

# The below is necessary for starting core Python generated random numbers
# in a well-defined state.
python_random.seed(123)

# The below set_seed() will make random number generation
# in the TensorFlow backend have a well-defined initial state.
# For further details, see:
# https://www.tensorflow.org/api_docs/python/tf/random/set_seed
tf.random.set_seed(1234)</code></pre>







Epoch 2/5



1875/1875 [==============================] – 4s 2ms/step – loss: 0.1408 – accuracy: 0.9584&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 3/5



1875/1875 [==============================] – 4s 2ms/step – loss: 0.1090 – accuracy: 0.9667&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 4/5



1875/1875 [==============================] – 4s 2ms/step – loss: 0.0892 – accuracy: 0.9722&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 5/5



1875/1875 [==============================] – 4s 2ms/step –&nbsp;loss: 0.0763 – accuracy: 0.9765&nbsp; &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



313/313 – 0s – loss: 0.0729 – accuracy: 0.9776







Epoch 2/5



1875/1875 [==============================] – 4s 2ms/step – loss: 0.1408 – accuracy: 0.9584&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 3/5



1875/1875 [==============================] – 4s 2ms/step – loss: 0.1090 – accuracy: 0.9667&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 4/5



1875/1875 [==============================] – 4s 2ms/step – loss: 0.0892 – accuracy: 0.9724&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



Epoch 5/5



1875/1875 [==============================] – 4s 2ms/step –&nbsp;loss: 0.0767 – accuracy: 0.9761&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;



313/313 – 1s – loss: 0.0727 – accuracy: 0.9768







Epoch 2/5



1875/1875 [==============================] – 4s 2ms/step – loss: 0.1408 – accuracy: 0.9584



Epoch 3/5



1875/1875 [==============================] – 4s 2ms/step – loss: 0.1090 – accuracy: 0.9667



Epoch 4/5



1875/1875 [==============================] – 4s 2ms/step – loss: 0.0892 – accuracy: 0.9723



Epoch 5/5



1875/1875 [==============================] – 4s 2ms/step –&nbsp;loss: 0.0766 – accuracy: 0.9762



313/313 – 1s – loss: 0.0737 – accuracy: 0.9771







학습 결과가 앞에서 했던것 보다는 일정합니다. 완전히 일치하지는 않네요.



Dropout을 사용해서 영향을 준듯합니다.&nbsp;



지금 사용한 MNIST 코드보다는 데이터셋을 무작위로 뒤섞은 후, 사용하는 경우에 효과가 더 좋았습니다.





</div></div>

멈춤보단 천천히라도

webnautes

텐서플로우에서 다음과 같은 간단한 MNIST 예제를 실행시켜보면 매번 학습 결과가 다른 것을 확인할 수 있습니다. 실행할때마다 일정한 학습 결과를 얻는 방법을 다룹니다.  2021. 7. 29  – 최초작성 세번 연달아 실행시킨 결과인데 학습 결과가 일정하지 않습니다.  Epoch 2/5 1875/1875 [==============================] – 5s 2ms/step – loss: 0.1442 – accuracy: 0.9568        Epoch 3/5 1875/1875 [==============================] – 5s 2ms/step… 더 보기 »Tensorflow에서 재현 가능한 결과(reproducible results) 얻기
The post Tensorflow에서 재현 가능한 결과(reproducible results) 얻기 first appeared on 멈춤보단 천천히라도.

Tensorflow에서 재현 가능한 결과(reproducible results) 얻기

댓글