ModelCheckpoint 不保存 hdf5 文件

Question

我在构建语音到文本模型时遇到了一个奇怪的问题。

模型已训练和创建，但是当我尝试将其保存为 hdf5 文件时，为了将其重新用于测试目的，操作根本不起作用，控制台中也没有弹出任何错误…

完整代码如下：

import os
import librosa   
import IPython.display as ipd
import matplotlib.pyplot as plt
import numpy as np
from scipy.io import wavfile 
import warnings
from sklearn.preprocessing import LabelEncoder
from keras.utils import np_utils
from sklearn.model_selection import train_test_split
from keras.layers import Dense, Dropout, Flatten, Conv1D, Input, MaxPooling1D
from keras.models import Model
from keras.callbacks import EarlyStopping, ModelCheckpoint
from keras import backend as K
from keras.models import load_model
import random

warnings.filterwarnings("ignore")

samples, sample_rate = librosa.load(r'C:\Users\makra\OneDrive\Desktop\Conda\Speech-to-text\train\audio\yes[=10=]a7c2a8d_nohash_0.wav', sr = 16000)

ipd.Audio(samples, rate=sample_rate)
samples = librosa.resample(samples, sample_rate, 8000)
ipd.Audio(samples, rate=8000)

labels=os.listdir(r'C:\Users\makra\OneDrive\Desktop\Conda\Speech-to-text\train\audio')

labels=["yes", "no", "one", "two", "three", "four", "five", "six", "seven", "eight", "nine", "zero"]

all_wave = []
all_label = []
for label in labels:
    print(label)
    waves = [f for f in os.listdir(r'C:\Users\makra\OneDrive\Desktop\Conda\Speech-to-text\train\audio' + '/'+ label) if f.endswith('.wav')]
    for wav in waves:
        samples, sample_rate = librosa.load(r'C:\Users\makra\OneDrive\Desktop\Conda\Speech-to-text\train\audio' + '/' + label + '/' + wav, sr = 16000)
        samples = librosa.resample(samples, sample_rate, 8000)
        if(len(samples)== 8000) : 
            all_wave.append(samples)
            all_label.append(label)
            
le = LabelEncoder()
y=le.fit_transform(all_label)
classes= list(le.classes_)

y=np_utils.to_categorical(y, num_classes=len(labels))
all_wave = np.array(all_wave).reshape(-1,8000,1)

x_tr, x_val, y_tr, y_val = train_test_split(np.array(all_wave),np.array(y),stratify=y,test_size = 0.2,random_state=777,shuffle=True)

K.clear_session()

inputs = Input(shape=(8000,1))

#First Conv1D layer
conv = Conv1D(8,13, padding='valid', activation='relu', strides=1)(inputs)
conv = MaxPooling1D(3)(conv)
conv = Dropout(0.3)(conv)

#Second Conv1D layer
conv = Conv1D(16, 11, padding='valid', activation='relu', strides=1)(conv)
conv = MaxPooling1D(3)(conv)
conv = Dropout(0.3)(conv)

#Third Conv1D layer
conv = Conv1D(32, 9, padding='valid', activation='relu', strides=1)(conv)
conv = MaxPooling1D(3)(conv)
conv = Dropout(0.3)(conv)

#Fourth Conv1D layer
conv = Conv1D(64, 7, padding='valid', activation='relu', strides=1)(conv)
conv = MaxPooling1D(3)(conv)
conv = Dropout(0.3)(conv)

#Flatten layer
conv = Flatten()(conv)

#Dense Layer 1
conv = Dense(256, activation='relu')(conv)
conv = Dropout(0.3)(conv)

#Dense Layer 2
conv = Dense(128, activation='relu')(conv)
conv = Dropout(0.3)(conv)

outputs = Dense(len(labels), activation='softmax')(conv)

model = Model(inputs, outputs)
model.summary()

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

es = EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=10, min_delta=0.0001) 
mc = ModelCheckpoint('best_model.hdf5', monitor='val_acc', verbose=1, save_best_only=True, mode='max')

history=model.fit(x_tr, y_tr ,epochs=100, callbacks=[es,mc], batch_size=32, validation_data=(x_val,y_val))

非常感谢您的帮助，

马克拉姆

Answer 1

我建议您在 ModelCheckpoint 回调中使用 save_weights_only=True 选项并使用 API model.load_weights(checkpoint_path)。这将加载最新的模型权重，如 here 所示。这只是一个临时解决方案，我认为您的模型精度没有提高，这就是您的 save_best_only 选项无法保存任何模型的原因在 ModelCheckpoint 回调中设置 accuracy 应该增加 的条件。

ModelCheckpoint 不保存 hdf5 文件

ModelCheckpoint not saving the hdf5 file

speech-recognition

text-to-speech

speech-to-text

keras

tensorflow