python - Keras 没有验证数据集

Question

我正在尝试按照语音识别权重和偏差的教程进行操作：

https://github.com/lukas/ml-class/tree/master/videos/cnn-audio

我做了教程中的所有操作，但收到以下错误消息：

wandb: WARNING No validation_data set, pass a generator to the callback.

您也可以通过 GitHub 链接查找代码，它非常相似（我只更改了标签的名称）。

预处理.py：

import librosa
import os
from sklearn.model_selection import train_test_split
from tensorflow.keras.utils import to_categorical
import numpy as np
from tqdm import tqdm

DATA_PATH = "./data/"


# Input: Folder Path
# Output: Tuple (Label, Indices of the labels, one-hot encoded labels)
def get_labels(path=DATA_PATH):
    labels = os.listdir(path)
    label_indices = np.arange(0, len(labels))
    return labels, label_indices, to_categorical(label_indices)


# convert file to wav2mfcc
# Mel-frequency cepstral coefficients
def wav2mfcc(file_path, n_mfcc=20, max_len=11):
    wave, sr = librosa.load(file_path, mono=True, sr=None)
    wave = np.asfortranarray(wave[::3])
    mfcc = librosa.feature.mfcc(wave, sr=16000, n_mfcc=n_mfcc)

    # If maximum length exceeds mfcc lengths then pad the remaining ones
    if (max_len > mfcc.shape[1]):
        pad_width = max_len - mfcc.shape[1]
        mfcc = np.pad(mfcc, pad_width=((0, 0), (0, pad_width)), mode='constant')

    # Else cutoff the remaining parts
    else:
        mfcc = mfcc[:, :max_len]
    
    return mfcc


def save_data_to_array(path=DATA_PATH, max_len=11, n_mfcc=20):
    labels, _, _ = get_labels(path)

    for label in labels:
        # Init mfcc vectors
        mfcc_vectors = []

        wavfiles = [path + label + '/' + wavfile for wavfile in os.listdir(path + '/' + label)]
        for wavfile in tqdm(wavfiles, "Saving vectors of label - '{}'".format(label)):
            mfcc = wav2mfcc(wavfile, max_len=max_len, n_mfcc=n_mfcc)
            mfcc_vectors.append(mfcc)
        np.save(label + '.npy', mfcc_vectors)


def get_train_test(split_ratio=0.6, random_state=42):
    # Get available labels
    labels, indices, _ = get_labels(DATA_PATH)

    # Getting first arrays
    X = np.load(labels[0] + '.npy')
    y = np.zeros(X.shape[0])

    # Append all of the dataset into one single array, same goes for y
    for i, label in enumerate(labels[1:]):
        x = np.load(label + '.npy')
        X = np.vstack((X, x))
        y = np.append(y, np.full(x.shape[0], fill_value= (i + 1)))



    assert X.shape[0] == len(y)

    return train_test_split(X, y, test_size= (1 - split_ratio), random_state=random_state, shuffle=True)


def prepare_dataset(path=DATA_PATH):
    labels, _, _ = get_labels(path)
    data = {}
    for label in labels:
        data[label] = {}
        data[label]['path'] = [path  + label + '/' + wavfile for wavfile in os.listdir(path + '/' + label)]

        vectors = []

        for wavfile in data[label]['path']:
            wave, sr = librosa.load(wavfile, mono=True, sr=None)
            # Downsampling
            wave = wave[::3]
            mfcc = librosa.feature.mfcc(wave, sr=16000)
            vectors.append(mfcc)

        data[label]['mfcc'] = vectors

    return data


def load_dataset(path=DATA_PATH):
    data = prepare_dataset(path)

    dataset = []

    for key in data:
        for mfcc in data[key]['mfcc']:
            dataset.append((key, mfcc))

    return dataset[:100]


# print(prepare_dataset(DATA_PATH))

音频.ipynb：

from preprocess import *
import keras
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten, Conv2D, MaxPooling2D, LSTM
from tensorflow.keras.utils import to_categorical
import wandb
from wandb.keras import WandbCallback
import matplotlib.pyplot as plt

wandb.init()
config = wandb.config

config.max_len = 11
config.buckets = 20

# Save data to array file first
save_data_to_array(max_len=config.max_len, n_mfcc=config.buckets)

labels=["off", "on", "stop"]

# Loading train set and test set
X_train, X_test, y_train, y_test = get_train_test()

# Feature dimension
channels = 1
config.epochs = 100
config.batch_size = 100

num_classes = 3

X_train = X_train.reshape(X_train.shape[0], config.buckets, config.max_len, channels)
X_test = X_test.reshape(X_test.shape[0], config.buckets, config.max_len, channels)

plt.imshow(X_train[100, :, :, 0])
print(y_train[100])

y_train_hot = to_categorical(y_train)
y_test_hot = to_categorical(y_test)

X_train = X_train.reshape(X_train.shape[0], config.buckets, config.max_len)
X_test = X_test.reshape(X_test.shape[0], config.buckets, config.max_len)

model = Sequential()
model.add(Flatten(input_shape=(config.buckets, config.max_len)))
model.add(Dense(num_classes, activation='softmax'))
model.compile(loss="categorical_crossentropy",
                  optimizer="adam",
                  metrics=['accuracy'])

wandb.init()
model.fit(X_train, y_train_hot, epochs=config.epochs, validation_data=(X_test, y_test_hot), callbacks=[WandbCallback(data_type="image", labels=labels)])

这对我来说是一种逻辑，为什么会发生这个错误，因为我在代码中找不到它说这是验证数据的位置的地方。我也想使用我下载的验证数据，但不知道如何。

您能帮我解决验证数据的错误吗？

wandb 版本：0.10.30

标签： pythonkerasspeech-recognition