딥러닝 기술 및 응용 (Fall 2018, UST, 15984)

오늘 주제 토론 (kNN) 발표 자료입니다.

발표자 : 최붕기 박사님

knn.key.pdf

CNN 실습 소스 코드 입니다. (from learnopencv.com )

cifar10 이미지 분류 dataset 대상

import numpy as np

import matplotlib.pyplot as plt

%matplotlib inline

from __future__ import print_function

import keras

from keras.models import Sequential

from keras.layers import Dense, Conv2D, MaxPooling2D, Dropout, Flatten

from keras.datasets import cifar10

(train_images, train_labels), (test_images, test_labels) = cifar10.load_data()

from keras.utils import to_categorical

print('Training data shape : ', train_images.shape, train_labels.shape)

print('Testing data shape : ', test_images.shape, test_labels.shape)

# Find the unique numbers from the train labels

classes = np.unique(train_labels)

nClasses = len(classes)

print('Total number of outputs : ', nClasses)

print('Output classes : ', classes)

plt.figure(figsize=[4,2])

# Display the first image in training data

plt.subplot(121)

plt.imshow(train_images[0,:,:], cmap='gray')

plt.title("Ground Truth : {}".format(train_labels[0]))

# Display the first image in testing data

plt.subplot(122)

plt.imshow(test_images[0,:,:], cmap='gray')

plt.title("Ground Truth : {}".format(test_labels[0]))

# Find the shape of input images and create the variable input_shape

nRows,nCols,nDims = train_images.shape[1:]

train_data = train_images.reshape(train_images.shape[0], nRows, nCols, nDims)

test_data = test_images.reshape(test_images.shape[0], nRows, nCols, nDims)

input_shape = (nRows, nCols, nDims)

# Change to float datatype

train_data = train_data.astype('float32')

test_data = test_data.astype('float32')

# Scale the data to lie between 0 to 1

train_data /= 255

test_data /= 255

# Change the labels from integer to categorical data

train_labels_one_hot = to_categorical(train_labels)

test_labels_one_hot = to_categorical(test_labels)

def createModel():

    model = Sequential()

    # The first two layers with 32 filters of window size 3x3

    model.add(Conv2D(32, (3, 3), padding='same', activation='relu', input_shape=input_shape))

    model.add(Conv2D(32, (3, 3), activation='relu'))

    model.add(MaxPooling2D(pool_size=(2, 2)))

    model.add(Dropout(0.25))

    model.add(Conv2D(64, (3, 3), padding='same', activation='relu'))

    model.add(Conv2D(64, (3, 3), activation='relu'))

    model.add(MaxPooling2D(pool_size=(2, 2)))

    model.add(Dropout(0.25))

    model.add(Conv2D(64, (3, 3), padding='same', activation='relu'))

    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(512, activation='relu'))

    model.add(Dropout(0.5))

    model.add(Dense(nClasses, activation='softmax'))

    return model

model1 = createModel()

batch_size = 256

epochs = 50

model1.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy'])

model1.summary()

history = model1.fit(train_data, train_labels_one_hot, batch_size=batch_size, epochs=epochs, verbose=1, 

                   validation_data=(test_data, test_labels_one_hot))

model1.evaluate(test_data, test_labels_one_hot)

plt.figure(figsize=[8,6])

plt.plot(history.history['loss'],'r',linewidth=3.0)

plt.plot(history.history['val_loss'],'b',linewidth=3.0)

plt.legend(['Training loss', 'Validation Loss'],fontsize=18)

plt.xlabel('Epochs ',fontsize=16)

plt.ylabel('Loss',fontsize=16)

plt.title('Loss Curves',fontsize=16)

plt.figure(figsize=[8,6])

plt.plot(history.history['acc'],'r',linewidth=3.0)

plt.plot(history.history['val_acc'],'b',linewidth=3.0)

plt.legend(['Training Accuracy', 'Validation Accuracy'],fontsize=18)

plt.xlabel('Epochs ',fontsize=16)

plt.ylabel('Accuracy',fontsize=16)

plt.title('Accuracy Curves',fontsize=16)

time series binary classification.

Home assignment용 데이터 (자세한 사항은 수업시간에 설명하겠습니다.)

출처: Alcoholism EEG dataset from UCI machine learning data repository

TR_allTrialEEGavgGlobalNormLabeled.csv.zip

6번째 수업 자료 입니다.

class06.pptx

주제 토론 발표 자료 입니다.

주제: Random Forest

발표자: 서신원

Ramdom Forest ver.2.pdf

5번째 강의 슬라이드 입니다.

class05.pdf

오늘 주제 발표 2 (Naïve Bayes Classification)에 사용된 슬라이드와 예제 소스코드 입니다.

발표자: 윤빛나리

Example_NBC.ipynb 

20181012_Naive Bayes Classifier_YUN.pdf

NBC.py

1. 하드에 있는 모델을 로딩하자

2. 로딩된 모델에 새로운 테스트 데이터를 입력하여 예측값을 얻어보자.

3. 이 소스에서는 학습을 수행하지 않고 이미 학습된 모델을 로딩하여 이용하는 것을 확인하자.

import numpy as np

from keras.models import Sequential

from keras.layers import Dense

from keras.utils import np_utils

np.random.seed(5)

dataset = np.loadtxt("diabetes.csv", delimiter=",", skiprows = 1)

data_test = dataset[700:, 0:8]

label_test = dataset[700:, 8]

#data_test = data_test(768,8).astype('float32')

label_test = np_utils.to_categorical(label_test)

# load model

from keras.models import load_model

model = load_model('diabetes_mlp_model.h5')

# use the model

label_pred = model.predict_classes(data_test)

for i in range(30):

    print("Input: ",data_test[i,:]," Predict:"+str(label_pred[i]))

1. 아래 코드를 입력하여 epoch 별 학습 상태를 그래프로 확인하자.

2. 저장된 모델 파일의 존재를 확인하고 모델 파일을 loading하여 새로운 테스트 데이터를 입력하여 prediction을 수행하자.

import numpy as np

from keras.models import Sequential

from keras.layers import Dense

np.random.seed(5)

dataset = np.loadtxt("diabetes.csv", delimiter=",", skiprows = 1)

# split

data_val = dataset[500:, 0:8]

label_val = dataset[500:, 8]

data_train = dataset[:700, 0:8]

label_train = dataset[:700, 8]

data_test = dataset[700:, 0:8]

label_test = dataset[700:, 8]

# shuffle

train_rand_idxs = np.random.choice(700, 600)

val_rand_idxs = np.random.choice(268, 260)

data_train = data_train[train_rand_idxs]

label_train = label_train[train_rand_idxs]

data_val = data_val[val_rand_idxs]

label_val = label_val[val_rand_idxs]

# convert labels to one-hot representation

model = Sequential()

model.add(Dense(12, input_dim=8, activation='relu'))

model.add(Dropout(0.3))

model.add(Dense(8, activation = 'relu'))

model.add(Dropout(0.3))

model.add(Dense(1, activation='sigmoid'))

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

hist = model.fit(data_train, label_train, epochs = 150, batch_size = 64, validation_data=(data_val, label_val))

scores = model.evaluate(data_test, label_test)

print("%s: %.2f%%" %(model.metrics_names[1], scores[1]*100))

# save the trained model into file

#from keras.models import load_model

#model.save('diabetes_mlp_model.h5')

# show the learning process

%matplotlib inline

import matplotlib.pyplot as plt

fig, loss_ax = plt.subplots()

acc_ax = loss_ax.twinx()

loss_ax.plot(hist.history['loss'], 'y', label='train loss')

loss_ax.plot(hist.history['val_loss'], 'r', label='val loss')

acc_ax.plot(hist.history['acc'], 'b', label='train acc')

acc_ax.plot(hist.history['val_acc'], 'g', label='val acc')

loss_ax.set_xlabel('epoch')

loss_ax.set_ylabel('loss')

acc_ax.set_ylabel('accuracy')

loss_ax.legend(loc='upper left')

acc_ax.legend(loc='lower left')

plt.show()

import numpy as np

from keras.models import Sequential

from keras.layers import Dense

np.random.seed(5)

dataset = np.loadtxt("diabetes.csv", delimiter=",", skiprows = 1)

data_train = dataset[:700, 0:8]

label_train = dataset[:700, 8]

data_test = dataset[700:, 0:8]

label_test = dataset[700:, 8]

model = Sequential()

model.add(Dense(12, input_dim=8, activation='relu'))

model.add(Dense(8, activation = 'relu'))

model.add(Dense(1, activation='sigmoid'))

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

model.fit(data_train, label_train, epochs = 150, batch_size = 64)

scores = model.evaluate(data_test, label_test)

print("%s: %.2f%%" %(model.metrics_names[1], scores[1]*100))