TensorFlow Datasets is a collection of ready to use datasets for Text, Audio, image and many other ML applications.
In this post we will load "titanic" dataset with tf.data.Datasets.Run below code in either Jupyter notebook or in google Colab.
Intsall TensorFlow datasets
pip install tensorflow-datasets
Import modules and load "titanic" dataset with TensorFlow Datsets
import tensorflow as tf
import tensorflow_datasets as tfds
# Construct a tf.data.Dataset
ds = tfds.load('titanic', split='train', shuffle_files=True)
ds = ds.shuffle(1024).batch(32).prefetch(tf.data.experimental.AUTOTUNE)
for example in ds.take(1):
# Print features for the batch input
print(" Input Features")
print(example['features'])
# Print labels for batch input
print(" Input Labels")
print(example['survived'])
