i'm using tensorflow_io_bigquery_client in order to read data from bq. each record has 4 integer elements.
but i need to transform it into tensor (4 integer each).
so i have something like:
tensorflow_io_bigquery_client = BigQueryClient()
read_session = tensorflow_io_bigquery_client.read_session(..)
dataset = read_session.parallel_read_rows()
dataset = dataset.map (transofrom_row)
#todo convert it into tensor (but i guess it should be an iterator or something like this because the dataset is hugh?
k_means_estimator = tf.compat.v1.estimator.experimental.KMeans(num_clusters = num_clusters, use_mini_batch = False, relative_tolerance = 1)
fit = k_means_estimator.train(input_fn=lambda: to_tensor(dataset), steps=1000)
# todo convert dataset into tensor