예제 3-1 필기체를 분류하는 DNN 구현

In [2]:

    

# %load /home/sjkim/.jupyter/head.py
%matplotlib inline
%load_ext autoreload 
%autoreload 2
from importlib import reload

import matplotlib.pyplot as plt
import numpy as np

import pandas as pd
import os
#os.environ["CUDA_VISIBLE_DEVICES"]="0"

# seaborn
#import seaborn as sns
#sns.set( style = 'white', font_scale = 1.7)
#sns.set_style('ticks')
#plt.rcParams['savefig.dpi'] = 200

# font for matplotlib
#import matplotlib
#import matplotlib.font_manager as fm
#fm.get_fontconfig_fonts()
#font_location = '/usr/share/fonts/truetype/nanum/NanumGothicBold.ttf'
#font_name = fm.FontProperties(fname=font_location).get_name()
#matplotlib.rc('font', family=font_name)


    

    




The autoreload extension is already loaded. To reload it, use:
  %reload_ext autoreload

In [2]:

    

import ex3_1_dnn_mnist_cl


    

    




Using TensorFlow backend.






    




Train on 48000 samples, validate on 12000 samples
Epoch 1/5
48000/48000 [==============================] - 3s - loss: 0.3812 - acc: 0.8904 - val_loss: 0.1867 - val_acc: 0.9466
Epoch 2/5
48000/48000 [==============================] - 2s - loss: 0.1565 - acc: 0.9552 - val_loss: 0.1380 - val_acc: 0.9605
Epoch 3/5
48000/48000 [==============================] - 2s - loss: 0.1113 - acc: 0.9664 - val_loss: 0.1156 - val_acc: 0.9662
Epoch 4/5
48000/48000 [==============================] - 2s - loss: 0.0890 - acc: 0.9738 - val_loss: 0.1057 - val_acc: 0.9685
Epoch 5/5
48000/48000 [==============================] - 2s - loss: 0.0698 - acc: 0.9789 - val_loss: 0.0984 - val_acc: 0.9712
 7700/10000 [======================>.......] - ETA: 0sTest Loss and Accuracy -> [0.091476730333524756, 0.97250000655651092]

In [ ]: