tfrecord：从 .tfrecord 文件打印图像

Question

我从头开始为我的张量流模型创建了一个数据集。我正在使用 TensorFlow 2.4.0。为了加速，我决定将数据存储在 .tfrecord 文件类型中，现在我想检查它是否正确存储在 .tfrecord 文件中。我写了一段代码来打印 .tfrecord 文件中的一张图片，但出现以下错误：

    imageRaw = imageFeautre['image/width'].numpy()
TypeError: 'TakeDataset' object is not subscriptable

我以官方tensorflow教程为导向(https://www.tensorflow.org/tutorials/load_data/tfrecord#write_the_tfrecord_file)

我可以加载数据集并读取它，内容是正确的。我无法在其中打印一张图片，我只想在其中打印一张图片我在网上没有找到解决方案。

这是我的代码：

import tensorflow as tf
import numpy as np
import IPython.display as display





  tf.compat.v1.enable_eager_execution()
  tfrecordPath='/home/adem/PycharmProjects/dcganAlgorithmus/dataHandler/preparedData/train.tfrecord'



rfrecordDataSet=tf.data.TFRecordDataset(tfrecordPath)


imageFeatureDescription ={
    'image/width:':tf.io.FixedLenFeature([],tf.int64),
    'image/height':tf.io.FixedLenFeature([], tf.int64),
    'image/xmin':tf.io.FixedLenFeature([], tf.int64),
    'image/ymin':tf.io.FixedLenFeature([],tf.int64),
    'image/xmax':tf.io.FixedLenFeature([],tf.int64),
    'image/ymin':tf.io.FixedLenFeature([],tf.int64),
}


def _parse_image_function(example_proto):
  # Parse the input tf.train.Example proto using the dictionary above.
  return tf.io.parse_single_example(example_proto, imageFeatureDescription)



ParsedImageDataset = rfrecordDataSet.map(_parse_image_function)
imageFeautre=ParsedImageDataset.take(1)
imageRaw = imageFeautre['image/width'].numpy()
display.display(display.Image(data=imageRaw))

Answer 1

您提到打印图像，但您的示例显示提取宽度。这是一个显示两者的示例。

feature_description = {
    'image/width': tf.io.FixedLenFeature([], tf.int64, default_value=0),
    'image/encoded': tf.io.FixedLenFeature([], tf.string, default_value=''),
    ...
}

tfrecord_file = 'myfile.tfrecord'
raw_dataset = tf.data.TFRecordDataset(tfrecord_file)
for raw_record in raw_dataset.take(num_records_to_plot):
    example = tf.train.Example()
    example.ParseFromString(raw_record.numpy())
    record = tf.io.parse_single_example(raw_record, feature_description)

    width = record['image/width'].numpy()
    image = record['image/encoded'] 

    # Convert image from raw bytes to numpy array
    image_decoded = tf.image.decode_image(image)
    image_decoded_np = image_decoded.numpy()
    ....

您可能还想确保存储的宽度有效。以下是我创建 TFRecord 的方式：

from PIL import Image

im = Image.open(file_path)
image_w, image_h = im.size

with tf.io.gfile.GFile(file_path, 'rb') as fid: 
    encoded_jpg = fid.read()

tf_example = tf.train.Example(features=tf.train.Features(feature={
    'image/width': dataset_util.int64_feature(image_w),
    'image/encoded': dataset_util.bytes_feature(encoded_jpg),
    ...
} 
writer.write(tf_example.SerializeToString())
``

tfrecord：从 .tfrecord 文件打印图像

tfrecord: print image from .tfrecord file

python

tensorflow

tensorflow-datasets

tensorflow2.0