9 minutes read (About 1383 words)

Tensorflow를 활용한 회귀 모델링

자동차 연비 예측하기
1
2
3
4
5
6
7
8
import pathlib
import matplotlib.pyplot as plt
import seaborn as sns
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers

print(tf.__version__)
2.4.0-dev20200724
Auto MPG 데이터셋
UCI 머신러닝 저장소에서 다운로드를 받자!
1
2
3
dataset_path = keras.utils.get_file("auto-mpg.data", "http://archive.ics.uci.edu/ml/\
                                    machine-learning-databases/auto-mpg/auto-mpg.data")
dataset_path
'/Users/wglee/.keras/datasets/auto-mpg.data'
1
2
3
4
5
6
7
# 데이터 불러오기
column_names = ['MPG', 'Cylinders', 'Displacement', 'Horsepwer', 'Weight', 'Acceleration',\
               'Model_year', 'Origin']

dataset = pd.read_csv(dataset_path, names=column_names, na_values='?', comment='\t', sep=' ',\
                     skipinitialspace=True)
df = dataset.copy()
1
df.tail(2)

MPG Cylinders Displacement Horsepwer Weight Acceleration Model_year Origin
396 28.0 4 120.0 79.0 2625.0 18.6 82 1
397 31.0 4 119.0 82.0 2720.0 19.4 82 1 
1
df['Origin'].unique()
array([1, 3, 2])



null값 확인 결과 6개의 데이터가 누락된 것을 확인하였고 제거 정제 작업을 하였습니다.
1
df.isnull().sum()
MPG             0
Cylinders       0
Displacement    0
Horsepwer       6
Weight          0
Acceleration    0
Model_year      0
Origin          0
dtype: int64
1
df.dropna(inplace=True)
1
2
import missingno as msno
msno.matrix(df, figsize=(8, 2))
<matplotlib.axes._subplots.AxesSubplot object at 0x7faaa5864f10>
output_10_1 
"Origin" 열은 수치형이 아니고 범주형이므로 원-핫 인코딩(one-hot encoding)으로 변환
1
origin = df.pop('Origin')
1
2
3
df['USA'] = (origin == 1) * 1.0
df['Europe'] = (origin == 2) * 2.0
df['Japan'] = (origin == 3) * 3.0
1
df.tail(2)

MPG Cylinders Displacement Horsepwer Weight Acceleration Model_year USA Europe Japan
396 28.0 4 120.0 79.0 2625.0 18.6 82 1.0 0.0 0.0
397 31.0 4 119.0 82.0 2720.0 19.4 82 1.0 0.0 0.0 
데이터셋 분리 (train, test)
1
2
train_df = df.sample(frac=0.7, random_state=0)
test_df = df.drop(train_df.index)
1
len(train_df)
274



데이터 EDA를 통해 데이터의 분포 및 통계치를 확인합니다
1
2
sns.pairplot(train_df[['MPG','Cylinders','Displacement','Weight']], diag_kind='kde')
plt.show()
output_19_0 
1
2
3
4
train_stats = train_df.describe()
# train_stats.pop("MPG")
train_stats = train_stats.T #transpose
train_stats

count mean std min 25% 50% 75% max
MPG 274.0 23.323358 7.643458 10.0 17.0 22.0 29.000 46.6
Cylinders 274.0 5.467153 1.690530 3.0 4.0 4.0 8.000 8.0
Displacement 274.0 193.846715 102.402201 68.0 105.0 151.0 260.000 455.0
Horsepwer 274.0 104.135036 37.281034 46.0 76.0 93.0 128.000 225.0
Weight 274.0 2976.879562 829.860536 1649.0 2250.5 2822.5 3573.000 4997.0
Acceleration 274.0 15.590876 2.714719 8.0 14.0 15.5 17.275 24.8
Model_year 274.0 75.934307 3.685839 70.0 73.0 76.0 79.000 82.0
USA 274.0 0.635036 0.482301 0.0 0.0 1.0 1.000 1.0
Europe 274.0 0.335766 0.748893 0.0 0.0 0.0 0.000 2.0
Japan 274.0 0.591241 1.195564 0.0 0.0 0.0 0.000 3.0 
이번에는 train, test 분리가 아니라 feature와 label를 분리시켜 줍니다.
1
2
train_labels = train_df['MPG']
test_labels = test_df['MPG']
데이터 정규화
feature의 크기와 범위가 다르면 정규화(normalization)를 하는 것이 권장됩니다. 정규화를 하지 않아도 모델링이 가능하지만 훈련시키기 어렵고 입력 단위에 의존적인 모델이 만들어지게 됩니다.
1
2
3
4
5
# # 데이터 정규화
# from sklearn.preprocessing import StandardScaler
# scaler = StandardScaler()
# train_df = scaler.fit_transform(train_df)
# test_df = scaler.fit_transform(test_df)
1
2
3
4
def norm(x):
    return (x - train_stats['mean']) / train_stats['std']
normed_train_data = norm(train_df)
normed_test_data = norm(test_df)
모델링
모델을 구성해 보죠. 여기에서는 두 개의 완전 연결(densely connected) 은닉층으로 Sequential 모델을 만들겠습니다. 출력 층은 하나의 연속적인 값을 반환합니다. 나중에 두 번째 모델을 만들기 쉽도록 build_model 함수로 모델 구성 단계를 감싸겠습니다.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
# 모델링
def build_model():
    model = keras.Sequential([
        layers.Dense(128, activation='relu', input_shape=[len(train_df.keys())]),
        layers.Dense(64, activation='relu'),
        layers.Dense(1)
    ])

    optimizer = tf.keras.optimizers.RMSprop(0.001)

    model.compile(loss='mse',
                optimizer=optimizer,
                metrics=['mae', 'mse'])
    return model
1
model = build_model()
모델 확인
.summary() 메서드를 사용하여 모델의 간단한 정보를 출력해줍니다.
1
print(model.summary())
Model: "sequential_15"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_45 (Dense)             (None, 128)               1408      
_________________________________________________________________
dense_46 (Dense)             (None, 64)                8256      
_________________________________________________________________
dense_47 (Dense)             (None, 1)                 65        
=================================================================
Total params: 9,729
Trainable params: 9,729
Non-trainable params: 0
_________________________________________________________________
None


모델을 한번 실행해 보죠. training 세트에서 10개의 샘플을 하나의 배치로 만들어 model_predict 메서드를 호출해 보겠습니다.
1
2
3
example_batch = normed_train_data[:10]
example_result = model.predict(example_batch)
example_result
WARNING:tensorflow:5 out of the last 15 calls to <function Model.make_predict_function.<locals>.predict_function at 0x7faa9a8f0200> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/tutorials/customization/performance#python_or_tensor_args and https://www.tensorflow.org/api_docs/python/tf/function for  more details.





array([[-0.03285253],
       [-0.01362434],
       [-0.48285854],
       [ 0.01581845],
       [ 0.08219826],
       [ 0.08362657],
       [ 0.15519306],
       [ 0.28581452],
       [ 0.07680693],
       [ 0.01200353]], dtype=float32)
모델 훈련
에포크가 끝날 때마다 점(.)을 출력해 훈련 진행 과정을 표시합니다.
1
2
3
4
5
6
7
8
9
10
11
class PrintDot(keras.callbacks.Callback):
    def on_epoch_end(self, epoch, logs):
        if epoch % 100 == 0: print('')
        print('.', end='')

EPOCHS = 1000

history = model.fit(
  normed_train_data, train_labels,
  epochs=EPOCHS, validation_split = 0.2, verbose=0,
  callbacks=[PrintDot()])
....................................................................................................
....................................................................................................
....................................................................................................
....................................................................................................
....................................................................................................
....................................................................................................
....................................................................................................
....................................................................................................
....................................................................................................
....................................................................................................

acc : 훈련 정확도
loss : 훈련 손실값
val_acc : 검증 정확도
val_loss : 검증 손실값
1
2
3
hist = pd.DataFrame(history.history)
hist['epoch'] = history.epoch
hist.tail()

loss mae mse accuracy val_loss val_mae val_mse val_accuracy epoch
995 0.102436 0.234032 0.102436 0.0 0.165683 0.324213 0.165683 0.0 995
996 0.124358 0.292103 0.124358 0.0 0.263786 0.404004 0.263786 0.0 996
997 0.130789 0.295300 0.130789 0.0 0.212862 0.362374 0.212862 0.0 997
998 0.116644 0.275093 0.116644 0.0 0.054454 0.196261 0.054454 0.0 998
999 0.106241 0.280440 0.106241 0.0 0.121306 0.281089 0.121306 0.0 999 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
def plot_history(history):
       
    hist = pd.DataFrame(history.history)
    hist['epoch'] = history.epoch

    plt.figure(figsize=(8,8))
    
    plt.subplot(2,1,1)
    plt.plot(hist['epoch'], hist['mae'], label='Train Error')
    plt.xlabel('Epoch')
    plt.ylabel('Mean Abs Error [MPG]')
    plt.plot(hist['epoch'], hist['val_mae'],
             label = 'Val Error')
    plt.ylim([0,5])
    plt.legend()
    
    plt.subplot(2,1,2)
    plt.xlabel('Epoch')
    plt.ylabel('Mean Square Error [$MPG^2$]')
    plt.plot(hist['epoch'], hist['mse'],
             label='Train Error')
    plt.plot(hist['epoch'], hist['val_mse'],
             label = 'Val Error')
    plt.ylim([0,20])
    plt.legend()
    plt.show()

plot_history(history)
output_37_0 
1
2
3
4
5
6
7
8
9
model = build_model()

# patience 매개변수는 성능 향상을 체크할 에포크 횟수입니다
early_stop = keras.callbacks.EarlyStopping(monitor='val_loss', patience=10)

history = model.fit(normed_train_data, train_labels, epochs=EPOCHS,
                    validation_split = 0.2, verbose=0, callbacks=[early_stop, PrintDot()])

plot_history(history)
......................................................................................
output_38_1
모델 검증
테스트 세트의 모델 성능을 확인
1
2
3
loss, mae, mse = model.evaluate(normed_test_data, test_labels, verbose=2)

print("테스트 세트의 평균 절대 오차: {:5.2f} MPG".format(mae))
4/4 - 0s - loss: 0.4875 - mae: 0.5579 - mse: 0.4875
테스트 세트의 평균 절대 오차:  0.56 MPG
예측
테스트 세트에 있는 샘플을 이용해 MPG 값 예측
1
2
3
4
5
6
7
8
9
10
test_predictions = model.predict(normed_test_data).flatten()

plt.scatter(test_labels, test_predictions)
plt.xlabel('True Values [MPG]')
plt.ylabel('Predictions [MPG]')
plt.axis('equal')
plt.axis('square')
plt.xlim([0,plt.xlim()[1]])
plt.ylim([0,plt.ylim()[1]])
_ = plt.plot([-100, 100], [-100, 100])
output_42_0 
1
2
3
4
error = test_predictions - test_labels
plt.hist(error, bins = 25)
plt.xlabel("Prediction Error [MPG]")
_ = plt.ylabel("Count")
output_43_0 
1
2


You need to set install_url to use ShareThis. Please set it in _config.yml.

Like this article? Support the author with

You forgot to set the business or currency_code for Paypal. Please set it in _config.yml.

Comments

You forgot to set the shortname for Disqus. Please set it in _config.yml.
Follow

Links

Recent

Archives

Tags

fastcampus1
git1
python1

Subscribe to Updates

You need to set client_id and slot_id to show this AD unit. Please set it in _config.yml.
×