python: save () отсутствует 1 обязательный позиционный аргумент: 'filepath' - PullRequest
0 голосов
/ 13 сентября 2018

Я создаю сверточную нейронную сеть, используя данные датчиков.Теперь я хочу сохранить модель и вес.Но при запуске кода возникает вопрос.Я искал блог и документ, но понятия не имею.Я надеюсь, что кто-то может помочь мне.неправильно:

TypeError                                 

    201 plt.savefig('cmatrix.png')
    202 #plt.show();
--> 203 Model.save("model.h5")
    204 Model.save_weights("model_weights")
    TypeError: save() missing 1 required positional argument: 'filepath'

набор данных введите описание ссылки здесь .код выглядит следующим образом:

import matplotlib
    matplotlib.use('Agg')
    import timing
    from keras.layers import Input,Dense,Conv2D,MaxPooling2D,UpSampling2D,Flatten
    from keras.models import Model
    from keras import backend as K
    from keras.utils.np_utils import to_categorical
    import numpy as np
    import pandas as pd
    import seaborn as sns
    from keras.models import Sequential# 导入Sequential
    from keras.utils import np_utils, generic_utils
    from keras.callbacks import LearningRateScheduler
    import os
    from keras.layers import Dropout
    from keras.backend.tensorflow_backend import set_session
    import tensorflow as tf
    from sklearn.model_selection import train_test_split,  cross_val_score
    from sklearn.cross_validation import KFold, StratifiedKFold
    from keras.wrappers.scikit_learn import KerasClassifier
    from sklearn.preprocessing import LabelEncoder
    from sklearn import metrics
    import time
    from scipy import stats
    from keras import optimizers
    import matplotlib.pyplot as plt
    from keras import regularizers
    import keras
    from keras.callbacks import TensorBoard
    config = tf.ConfigProto(allow_soft_placement=True)
    gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=1)
    config.gpu_options.allow_growth = True
    sess = tf.Session(config=config)
    os.environ['TF_CPP_MIN_LOG_LEVEL']='2'
    time1 = time.time()
    class LossHistory(keras.callbacks.Callback):
        def on_train_begin(self, logs={}):
            self.losses = {'batch':[], 'epoch':[]}
            self.accuracy = {'batch':[], 'epoch':[]}
            self.val_loss = {'batch':[], 'epoch':[]}
            self.val_acc = {'batch':[], 'epoch':[]}

        def on_batch_end(self, batch, logs={}):
            self.losses['batch'].append(logs.get('loss'))
            self.accuracy['batch'].append(logs.get('acc'))
            self.val_loss['batch'].append(logs.get('val_loss'))
            self.val_acc['batch'].append(logs.get('val_acc'))

        def on_epoch_end(self, batch, logs={}):
            self.losses['epoch'].append(logs.get('loss'))
            self.accuracy['epoch'].append(logs.get('acc'))
            self.val_loss['epoch'].append(logs.get('val_loss'))
            self.val_acc['epoch'].append(logs.get('val_acc'))

        def loss_plot(self, loss_type):
            iters = range(len(self.losses[loss_type]))
            plt.figure()
            # acc
            plt.plot(iters, self.accuracy[loss_type], 'r', label='train acc')
            # loss
            plt.plot(iters, self.losses[loss_type], 'g', label='train loss')
            if loss_type == 'epoch':
                # val_acc
                plt.plot(iters, self.val_acc[loss_type], 'b', label='val acc')
                # val_loss
                plt.plot(iters, self.val_loss[loss_type], 'k', label='val loss')
            plt.grid(True)
            plt.xlabel(loss_type)
            plt.ylabel('acc-loss')
            plt.legend(loc="center")
            plt.show()
            #plt.savefig('common.png')


    #dataset
    RANDOM_SEED = 42
    def read_data(file_path):
        column_names = ['user-id', 'activity', 'timestamp', 'x-axis', 'y-axis', 'z-axis']
        m = pd.read_csv(file_path,names=column_names, header=None,sep=',')
        return m
    def feature_normalize(dataset):
        mu = np.mean(dataset,axis=0)
        sigma = np.std(dataset,axis=0)
        return (dataset-mu)/sigma

    dataset1 = read_data('ab.txt')
    dataset = pd.DataFrame(dataset1)
    dataset['x-axis'] = feature_normalize(dataset['x-axis'])
    dataset['y-axis'] = feature_normalize(dataset['y-axis'])
    dataset['z-axis'] = feature_normalize(dataset['z-axis'])

    N_TIME_STEPS = 200
    N_FEATURES = 3
    step = 200
    segments = []
    labels = []
    for i in range(0, len(dataset) - N_TIME_STEPS, step):
        xs = dataset['x-axis'].values[i: i + N_TIME_STEPS]
        ys = dataset['y-axis'].values[i: i + N_TIME_STEPS]
        zs = dataset['z-axis'].values[i: i + N_TIME_STEPS]
        label = stats.mode(dataset['activity'][i: i + N_TIME_STEPS])[0][0]
        segments.append([xs, ys, zs])
        labels.append(label)
    print("reduced size of data", np.array(segments).shape)
    reshaped_segments = np.asarray(segments,dtype=np.float32).reshape(-1,1, N_TIME_STEPS, 3)
    print("Reshape the segments", np.array(reshaped_segments).shape)
    #x_train1, x_val_test, y_train1, y_val_test = train_test_split(reshaped_segments, labels, test_size=0.25, random_state=RANDOM_SEED)

    batch_size = 128
    num_classes =6

    def create_model():
        input_shape = Input(shape=(1,200,3))
        #x = Conv2D(7, kernel_size=(1, 1), padding='valid')(input_shape)
        #x1 = keras.layers.concatenate([input_shape, x], axis=-1)

        x = Conv2D(50, kernel_size=(1, 7),padding='valid',
                     kernel_initializer='glorot_uniform',
        kernel_regularizer = keras.regularizers.l2(0.0015),name='conv1')(input_shape)


        x = keras.layers.core.Activation('relu',name='act1')(x)
        x = MaxPooling2D(pool_size=(1, 2),name='maxpool1')(x)
        x = Conv2D(50, kernel_size=(1, 7),padding='valid',kernel_initializer='glorot_uniform',
               kernel_regularizer=keras.regularizers.l2(0.0015),name='conv2')(x)
        x = keras.layers.core.Activation('relu',name='act2')(x)
        x = MaxPooling2D(pool_size=(1, 2),name='maxpool2')(x)

        x = Flatten()(x)
        x = Dropout(0.9)(x)
        output = Dense(num_classes, activation='softmax',kernel_initializer='glorot_uniform',name='dense1')(x)
        model = Model(inputs=input_shape,outputs=output)
        model.summary()

        sgd = optimizers.SGD(lr=0.005,decay=1e-6,momentum=0.9,nesterov=True)
        model.compile(loss=keras.losses.categorical_crossentropy,                        
                  optimizer=sgd,
                  metrics=['accuracy'])
        return model
    history = LossHistory()
    epochs = 4000

    #setting learning rate
    def scheduler(epoch):
        if epoch > 0.75 * epochs:
            lr = 0.0005
        elif epoch > 0.25 * epochs:
            lr = 0.001
        else:
            lr = 0.005
        return lr

    scheduler = LearningRateScheduler(scheduler)
    estimator = KerasClassifier(build_fn=create_model)

    scores = []
    confusions = []   #list of confusion matrix
    sign = ['DOWNSTAIRS','JOGGING','SITTING','STANDING','UPSTAIRS','WALKING']
    encoder = LabelEncoder()
    encoder_y = encoder.fit_transform(labels)

    train_labels = to_categorical(encoder_y,num_classes=None)

    kfold = StratifiedKFold(labels,n_folds=10,shuffle=True,random_state=42)
    for train_index,test_index in kfold:
        x_train, x_test = reshaped_segments[train_index], reshaped_segments[test_index]
        y_train, y_test = train_labels[train_index], train_labels[test_index]
        estimator.fit(x_train,y_train,callbacks=[scheduler,history],epochs=2,batch_size=128,verbose=0)
        scores.append(estimator.score(x_test,y_test))
        pred_test = estimator.predict(x_test)  
        confusions.append(metrics.confusion_matrix(np.argmax(y_test,axis=1),pred_test))

    matrix = [[0,0,0,0,0,0],[0,0,0,0,0,0],[0,0,0,0,0,0],[0,0,0,0,0,0],[0,0,0,0,0,0],[0,0,0,0,0,0]]

    for i in np.arange(3):
        for j in np.arange(6):
            for k in np.arange(6):
                matrix[j][k] = matrix[j][k] + confusions[i][j][k]   

    #Model.save("model.h5")
    #Model.save_weights("model_weights")    
    print('score:',scores)
    scores = np.mean(scores)
    print('mean:',scores)

    time2 = time.time()
    time3 = time2-time1
    print(time3)

    plt.figure(figsize=(16,14))     
    sns.heatmap(matrix, xticklabels=sign, yticklabels=sign, annot=True, fmt="d");
    plt.title("CONFUSION MATRIX : ")
    plt.ylabel('True Label')
    plt.xlabel('Predicted label')
    plt.savefig('cmatrix.png')
    #plt.show();
    Model.save("model.h5")
    Model.save_weights("model_weights")

1 Ответ

0 голосов
/ 17 сентября 2018

Я нахожу ответ. Я хочу вызвать созданную модель, но модель определяется в функции create_model (). Поэтому я должен определить модель как глобальную, а затем использовать model.save () для сохранения модели.

...