from keras.models import Sequential
import read_n_write as rw
import utils
import models
base_dir = os.path.dirname(__file__)
DATA_PATH = sys.argv[1]
LOG_FOLDER = './model_log'
CHECK_POINT_NAME = LOG_FOLDER + '/model.{epoch:04d}-{val_acc:.4f}.h5'
LOG_NAME = LOG_FOLDER + '/log.csv'
PATIENCE = 50
EPOCH = 2000
feats, lables, _ = rw.read_dataset(DATA_PATH, shuffle=True)
X_train, Y_train, X_val, Y_val = utils.validation(feats, lables, 0.05)
train_data_gen = ImageDataGenerator(rotation_range=40,
width_shift_range=0.2,
height_shift_range=0.2,
shear_range=0.2,
zoom_range=0.2,
horizontal_flip=True,
fill_mode='nearest')
train_data_gen.fit(X_train)
val_data_gen = ImageDataGenerator()
callback = [
TensorBoard(),
from keras.models import load_model, Sequential
from keras.utils import to_categorical
from sklearn.metrics import confusion_matrix
import itertools
import numpy as np
import matplotlib.pyplot as plt
import utils
import read_n_write as rw
MODEL_PATH = './model_re_bst.h5'
DATA_PATH = './data/train.csv'
model = load_model(MODEL_PATH)
x, y, _ = rw.read_dataset(DATA_PATH)
y = utils.to_nb_class(y)
y_pred = model.predict_classes(x)
# y_pred = to_categorical(y_pred)
# y_pred = utils.to_nb_class(y_pred)
print(y_pred)
cnf_mat = confusion_matrix(y, y_pred)
def plot_confusion_matrix(cm,
classes,
title='Confusion matrix',
cmap=plt.cm.jet):
"""
This function prints and plots the confusion matrix.
import os
import sys
from keras.models import Sequential, load_model
import utils
import read_n_write as rw
TEST_PATH = sys.argv[1]
ANS_PATH = sys.argv[2]
MODELS_PATH = ['./model_re_bst.h5',
'./model_ta.h5',
'./model_x.h5']
WEIGHT = [1.06, 1, 1]
x_test = rw.read_dataset('data/test.csv', labeled_data = False)
model_ls = []
result_ls = []
num_model = len(MODELS_PATH)
for i, w, mod_path in zip(range(num_model), WEIGHT, MODELS_PATH):
# io processing
model = load_model(mod_path)
model_ls.append(model)
result = model.predict(x_test)
result_ls.append(result)
if i == 0:
result_total = w * result
else:
result_total += w * result
# print(result_total)
import numpy as np
from keras.preprocessing.image import ImageDataGenerator
from keras.callbacks import EarlyStopping, CSVLogger, ModelCheckpoint, TensorBoard, ReduceLROnPlateau
from keras.models import Sequential
import read_n_write as rw
import utils
import models
base_dir = os.path.dirname(__file__)
TRAIN_FOLDER = './model_chk/dnn1'
CHECK_POINT_NAME = TRAIN_FOLDER + '/model.{epoch:04d}-{val_acc:.4f}.h5'
LOG_NAME = TRAIN_FOLDER + '/log.csv'
PATIENCE = 50
feats, lables, _ = rw.read_dataset('data/train.csv', shuffle=True)
X_train, Y_train, X_val, Y_val = utils.validation(feats, lables, 0.1)
train_data_gen = ImageDataGenerator(rotation_range=40,
width_shift_range=0.2,
height_shift_range=0.2,
shear_range=0.2,
zoom_range=0.2,
horizontal_flip=True,
fill_mode='nearest')
train_data_gen.fit(X_train)
val_data_gen = ImageDataGenerator()
# early_stopping = EarlyStopping(patience = 50)
callback = [
import os import sys from keras.models import Sequential, load_model import utils import read_n_write as rw base_dir = os.path.dirname(__file__) MODEL_PATH = './model_re_bst.h5' TEST_PATH = sys.argv[1] SAVE_PATH = sys.argv[2] # io processing model = load_model(MODEL_PATH) x = rw.read_dataset(TEST_PATH, labeled_data=False) y_pred = model.predict(x) y_pred = utils.to_nb_class(y_pred) rw.save_dir(y_pred, path=SAVE_PATH)