TFRECORD features extraction

Community Help
1

i installed the dataset call Mollusca with the extension .tfrecord (Mollusca Object Detection Dataset and Pre-Trained Model by mollus) but i m not able to find the organizaation of the features in it
import tensorflow as tf
import numpy as np

Definizione della struttura dei dati

def _parse_function(example_proto):
features_description = { # Definizione della struttura dei dati
‘image’: tf.io.FixedLenFeature(, tf.string),
‘label’: tf.io.FixedLenFeature(, tf.int64),

}
paresed_example=tf.io.parse_single_example(example_proto, features_description)  # Parsing dei dati
image = tf.image.decode_image(paresed_example['image'], channels=3)  # Decodifica dell'immagine da stringa a ten
label = paresed_example['label']  # Estrazione dell'etichetta
return image, label

dataset = tf.data.TFRecordDataset(“percorso”) # Caricamento del dataset
dataset = dataset.map(_parse_function) # Parsing dei dati
X_train =
y_train =
X_test =
y_test =

for image, label in dataset:
X_train.append(image.numpy())
y_train.append(label.numpy())

InvalidArgumentError Traceback (most recent call last) Cell In[48], line 6 3 X_test = 4 y_test = ----> 6 for image, label in dataset: 7 X_train.append(image.numpy()) 8 y_train.append(label.numpy()) File File :3, in raise_from(value, from_value) InvalidArgumentError: Feature: image (data type: string) is required but could not be found. [[{{node ParseSingleExample/ParseExample/ParseExampleV2}}]] [Op:IteratorGetNext]
what can i do?