Home > database >  Trying to combine two MapDatasets to become one MapDataset
Trying to combine two MapDatasets to become one MapDataset

Time:11-23

I am trying to combine these two mapdatasets into one. In other words, I am trying to expand my dataset.

ds = ds.map(lambda x, y: (load_aug(tf.image.resize(x, size)), y))
ds1 = ds1.map(lambda x, y: (tf.image.resize(x, size), y))

ds is an augmented version of a dataset an ds1 is a regular unaugmented dataset. The shape of the images in both datasets are the same.

CodePudding user response:

You could use zip or concatenate to combine both MapDatasets:

import tensorflow as tf

ds = tf.data.Dataset.range(1, 4)
ds1 = tf.data.Dataset.range(4, 8) 

train_dataset = tf.data.Dataset.zip((ds, ds1))
# or 
train_dataset = ds.concatenate(ds1)
  • Related